(I used to work for an autonomous truck company, and when you factor in the cost of roads in addition to the development of the trucks, it makes absolutely no sense to do autonomous trucking when you could do trains. As a culture, we've been brainwashed not to fund trains. We collectively spend billions and billions on roads but would not dare spend money to build more tracks. It is shocking and ludicrous, but that's what happens when you suck up a century of propaganda from the fossil fuel and automotive industry).
I think the problem few people talk about is the small, in-city rail lines that used to exist that don't anymore. 100 years ago the city I live in had dedicated rail lines going to every major factory in the city - there was little need for last-mile trucking. AFAIK there are no western cities that are organized like this anymore, so the best you can do is deliver goods to a rail station and truck it from there - this requires a bunch of temporary storage space to transfer cargo and delays everything.
I live in a city that has some of these going through it, and I can say that my opinions changed dramatically after moving here, with respect to this idea. Our real estate agent, in hindsight, had employed numerous subtle tactics to discourage us from taking serious interest in homes that were within a quarter mile or so of these lines, although I was completely oblivious to what they were doing at the time.
What I did not appreciate, but they did, is that these rails can bring an absolutely astonishing amount of noise into a town, and behind that a long trail of associated social ills. There is a gradual sifting of residents within a certain distance of these trains, based on who is either loud themselves and thus doesn't mind, or else must tolerate the amount of noise these things make.
Our town is old enough that some of the 'nicer' neighborhoods predate the introduction of the rails, so there isn't a strong confounding signal of 'bad' neighborhood correlation at work here. In at least one case there was a house we pushed past our agent's scheduling machinations to see for ourselves, and would likely have closed on had we not happened to visit at just the right time, when a multi-engined repositioning train (which we had no idea was a very common guest on what we thought was a mild mannered commuter rail line) came through. Though unseen and multiple blocks away, it still shook the floorboards as it passed.
Not all locomotives are equal, electric is much quieter than diesel
Both older DC and newer AC trains are effectively silent when coasting on a straight
The main sources of noise are cornering and braking. I live in a rare city that has narrow gauge for all of our rail, and am less than 200m from the busiest corridor in our state
I regret chosing an apartment on a tight bend (narrow gauge's bends are especially loud)
But if I were to live 500m away parallel to the track (as in, still 200m away from the track) the mostly AC electric trains going by every few minutes would be largely inaudible (<40dB)
My main point is tracks aren't loud, tight corners and areas around stops can be loud
I fear city planners aren't testing their decisions closely enough, though
Also, as a friend noticed when he bought a ver nice apartment next to the railroad is that they have to (at least in Sweden) maintain the tracks once a month. They do this by going very slowly with a special train that basically runs with the brakes on throwing sparks all over with all the noise you can imagine. And since it's going to slowly, they have to do it when there are the fewest number of other trains running, i.e. at three in the morning.
I grew up half a block from a freight train line, and you could hear it, but it wasn’t really an issue. Trains have gotten a lot louder in the US recently due to lack of rail maintenance. (Concretely, the transbay BART tube is screeches was above osha safety limits inside the car these days, but was quiet when new. The problem is that you need to grind rails to have a rounded top or they screech (and wear out train wheels), but BART doesn’t bother.
Anyway, noise pollution has a real impact on health, but so does particulate carbon and benzene from freeways. Before moving near a major road, find out what sort of PM 2.5 is typical, and check to see if everything is coated in black dust. If so, and you can afford some other place, move there instead.
I got a friend who did the first grind ever in NZ a few years back, they borrowed the grinder locos from Aussie. Every train driver he talked to commented how amazingly better the rails were after grinding.
I live in a small town and the loudest sounds besides the train itself is the extremely loud horn it blows before every crossing. Apparently nobody wants to pay for automatic booms at the road crossings so it legally has to use the horn that blasts through the area.
I live next to an intermodal train yard. Literally outside my windows [1].
I love the noise and the trains. You get used to the sounds, and it becomes comforting. Like a loud rain on a rooftop. I feel strange when it's quiet.
There are some homes by Georgia Tech that are literally feet from the tracks [2], and I think those are awesome. I also hope these trains don't derail frequently like the ones in Hulsey Yard do.
What kinds of noise do the trains produce? Do you think there are any ways to lessen it, and/or have any been implemented? (I'm thinking something like sound barriers).
Not a rail expert, not even a casual spectator. But my unstudied impression is that the noise that penetrates the most is the ultra-low frequency rumble that comes from the diesel side of the drivetrain, with a secondary source being steel on steel squealing of bogeys around bends at the minimal range of their tolerance. But the latter doesn't have the penetrating power of the former, not by a long shot, so to eliminate that I'm not sure if barrier would really have much of an effect. A lot of these in-town runs actually do have some sort of barrier already.
Instead powertrain would have to somehow switch entirely to battery while moving through sound propagating areas, and I'm assuming (maybe generously) that this is not already being done for some thorny technical reason and not just due to systemic inertia and complacency.
The trains that make the most noise are diesel-electric, not electric; electric cross country freight trains aren't really a thing here. I don't DE engines have a battery in the conventional sense, although arguably their fuel tender full of diesel is a sort of battery. But in order for a train to be fully electric you would need to have a fully electrified infrastructure for freight rail, which we do not have and would be far-fetched as a solution compared to equipping such trains with a local battery tender when they come into town, in order to move them through the town (somewhat) quietly. after which they can switch back to diesel-electric power for the long cross country runs.
Electric freight trains make about as much noise as diesel electrics, since the vast majority of the noise either makes is the wheels hitting the rails and going screech. I had a hotel room in Switzerland next to a rail yard and learned this the hard way.
> But in order for a train to be fully electric you would need to have a fully electrified infrastructure for freight rail, which we do not have and would be far-fetched as a solution compared to equipping such trains with a local battery tender when they come into town, in order to move them through the town (somewhat) quietly. after which they can switch back to diesel-electric power for the long cross country runs.
Okay, I understand now.
An alternative proposition to installing batteries in DE engines is modifying them to be able to switch between electric lines (infrastructure) and the diesel engine.
This means that towns can lay down the infrastructure, so that the diesel is switched off when the engine is going through a town, and then switches on again when the infrastructure is no longer there.
The added benefit to this would be then that rolling out infrastructure gradually is not a problem: keep adding infrastructure over long distances and after a few decades the diesel engine won't be needed anymore.
Sadly I think even installing infrastructure for local electrification would not get out of the starting gate, because you need to both modify all of the engines involved, as well as add infrastructure to every affected town with all of the predictable pushback from people who will react negatively the notion of installing a bunch of extremely high power, highly visible catenary on already-dangerous rail paths.
But fortunately we can take advantage of a useful characteristic of trains, which is that it's quite natural to hook up a car called a tender, normally carrying fuel such as diesel or (in the old days) coal but in this case would be a giant battery on wheels that's been pre-charged and positioned on a siding outside of town at the crest of any convenient rise, ready for connection to incoming freight trains. These can then switch off (or idle) their diesel engines and draw on the battery for power as it passes into and out of town. Obviously a fee would be charged, but towns have all the leverage here- it's not as if the freight can just take some other set of train tracks that go around the town to avoid payong it.
This has the benefit of being extremely easy to get by in on, as it makes very little demand of both freight train operators and towns respectively. operators need to add nothing more than a simple transfer switch to the engine (which may already exist) and for the towns, the same high power electric circuit as catenary, minus all the catenar, and located on a short siding safely well out of town. And of course a couple great big batteries, but they are mobile and can go where the yrains are, thus far cheaper than putting poles in the ground everywhere the trains run.
I’ve lived near tracks a couple times and the loudest by far is the horn from at-grade crossings. If you are a heavy sleeper and a few blocks away it might not matter, but if you are nearby it is incredibly loud.
For what it’s worth, I lived 50 yards from an at-grade freight track in the city, where from 10pm-7am the trains did not blow their horns, relying on the crossing protective lights and gates only, and that worked for me. Charming train horns during the day (yes they occasionally interrupted me when speaking during video meetings, but it was quaint and no big deal), peace and quiet at night. Great neighborhood, quite desirable, and I would be happy to live there again in the future.
Not sure the safety track record, but if it’s favorable, perhaps this kind of program could be expanded.
I live about 150m from a line that goes over a bridge. Twice a day 4pm and 4am. It is not often the night train wakes me, often I wait to hear the click clacks before thinking it might be an earthquake.
The day train usually slows since coming the other way it is approaching a pedestrian crossing. I kinda love it because when it speeds up again you can hear it's giant turbo spooling... Sounds so good.
The rumbling also sounds kinda cool, sometimes it'll shake the house, I thought about setting up an accelerometer on and Arduino to swee what output I'd see.
Often however I think I hear the train but it's just the sea. Or I can hardly hear the train for the noise of the sea.
Zoning is a problem, but it's not the only problem there. I think there are at least two more good ones we could discuss:
1. We actually do not have a market mechanic that exists in the United States that allows people to choose properly dense homes or neighborhoods because of a number of reasons. Search tools and things like that focus on metrics like square footage or number of bathrooms, and so people select for that "bigger is better" "more is better".
2. State departments of highways (they don't do anything besides that in most states) consist of individuals and contractor companies who have become adept and normal with building highways, and so all of the organizational inertia is on building anything that looks like a highway for the safety of drivers, to the detriment of everyone else. Since only highways get funded and built in most of the country, we end up having other companies like fast-food restaurants build in ways that cater to the driving demographic.
Sorry 1 does not make sense. It's not like I bid on a house based on its square footage, and then only find out the address when I win.
Even in the US, walkability scores, satellite photos, maps of transit and bike lanes, commute estimates, are all common enough on real estate sites. I think you just be seeing may be observing people's expressed preferences.
When you go online to use a search tool like Zillow, the main categories and features used to search are price, square footage, number of bedrooms, and number of bathrooms.
There are no filters or options for things like proximity to a park or coffee shop, neighborhood demographics, whether the neighborhood has events or a civic organization, or other relevant information. There is no scorecard for whether or not a home is architecturally sound (I.e. symmetrical, built with proper materials for the environment, etc.) either. You actually have to look through dozens or hundreds of homes yourself.
Items like walkability scores which are displayed on websites like Zillow are good, but they don’t really tell you much about how actually walkable an area is. They’re also not a primary feature in the search utilities.
Hm. I did a quick search on freight percentages, EU vs US and google gave me this snippet off the top hit. "46 percent of European freight goes by truck while only 11 percent goes by rail, while in the United States more than 40 percent goes by rail while just 30 percent goes on the highway."
Seems US is doing pretty well on prioritising freight rail transport, despite brainwashing, at least compared to europe.
Which is probably a good thing. Until that figure is a lot higher, passenger rail is just one more thing reducing the efficiencies of freight.
The reason is probably because, in the EU, passenger rail transport is prioritised over cargo. For example in Switzwerland rail cargo is mainly transported during the night because thats the only time the network isn't working at it's limit for the passenger trains.
In Canada, to my understanding, it's the other way round where passenger trains have been reduced because of the need for more cargo train trips. The USA might be similar.
In the USA most railroad are owned by the railroad company. When Amtrak wants to use a piece of rail they pay the railway.
But the USA is VAST. Freight trains are LONG. Not much of America's West is actually double track. And when you have a shorter Amtrack and a long freight train the Amtrack loses. The Amtrack has to wait until the freight train has passed.
As the freight trains run on PSR [1] there is no fixed time as to when they pass and thus the Amtrack can not run reliable.
As such it is easier for me to drive from West Los Angeles to Las Vegas, or even to drive to LAX and then fly to LAS than it is to take the train.
I believe you, but why does your source claim otherwise?
"Under PSR, freight trains operate on fixed schedules, much like passenger trains, instead of being dispatched whenever a sufficient number of loaded cars are available."
So do they leave on a fixed schedule, or is there no fixed time? It sounds like the previous system had no fixed time (leave when ready) whereas in the PSR world - there is in fact a set schedule.
They don't leave on fixed time. This is a big part of the union complaints. The engineer needs to be on standby and be there in 2 (?) hours when the train is 'full'.
That sentence doesn't match with the idea of a fixed schedule. A fixed schedule doesn't require you to be on-call, because you know when the train will leave.
That's 40% of "ton-miles" - trains disproportionately move heavy, cheap goods (i.e. aggregate, coal), which makes that number potentially misleading. The percentage of freight moved by rail by "value" is closer to 15% (trucking is ~65%).
Why would one want to emphasize value over weight? If I load up a train with extremely expensive metals, jewels, artwork, etc. couldn’t I raise a countries rail by value measurement overnight?
I agree it's good to consider the whole picture, but ton-miles are probably the most important to consider given the increased diesel required to truck heavy loads, correct?
So perhaps a priority could be to build more medium/high speed passenger rail, and get it off the freight lines?
Travelers and commuters win, cargo wins!
It is funny though, the US built it's current freight network in a comparatively low tech era, yet I doubt it could do it again in the current era due to funding and beauracracy.
This is a weird post. Could any highly industrialised nation build a new freight network today? Probably not. Mostly, it would be blocked by noise complaints and lack of land. Most nation-sized freight rail networks were built more than 100 years ago when population density was far lower and noise controls barely existed.
Also, have you heard of the Alameda Corridor is Los Angeles County? That project is incredible. The amount of soil moved for that project must be staggering. The dimensions: <<The centerpiece of the new Alameda Corridor would be the "Mid-Corridor Trench" a below-ground, triple-tracked rail line that is 10 miles (16 km) long, 33 feet (10 m) deep, and 50 feet (15 m) wide.>>
Freight trains don't have to be louder than passenger trains. (Though from what I've heard, the US has incredible noisy trains, thanks also to lots of horn blaring?)
And quite a few industrialised countries around the world are building more rail lines.
The US already has an extensive freight rail network. It's the one part of their rail transport that works fairly well. Expanding that should be possible.
Especially because freight typically travels slower than passengers, so your track doesn't need to be as straight, so you have more leeway in planning.
Freight is noisy in the US for two reasons: (1) much heavier than trucks on poorly maintained tracks and (2) required horn blaring for safety at road crossings.
My own adopted home of Singapore is highly industrialised, and building more rail lines. China is reasonably industrialised and building more rail lines. Spain is building more, etc.
You can do some web searches. Alas, it's a bit hard, because Google seems to insists that I am only interested in new _high-speed_ rail lines, instead of all new rail lines. But if you are willing to only look at those, https://en.wikipedia.org/wiki/Proposed_high-speed_rail_by_co... has a list.
The original post suggested to build more passenger rails to reduce burden upon freight rails. Does Singapore move any freight by rail?
Also, the original post wrote:
> the US built it's current freight network in a comparatively low tech era, yet I doubt it could do it again in the current era due to funding and beauracracy.
My reply was specific to that. Saying: The original post says little because _no_ highly industrialised nation could build a freight network from scratch given current land and noise constraints.
If you are building new freight rail in your country, you can avoid problems (1) and (2) from that comment. Freight rail doesn't take more land than other rail (if anything, it has lower requirements, because lower speeds mean tighter curves are possible). And freight rail doesn't have to be noisier than passenger rail.
It's certainly not a US only problem. My city tore up all it's rail and now there are houses in the way so we can't put it back. On top of that, labour and materials are so expensive we probably couldn't afford it anyway.
Thanks for showing me that project, I had not heard of it. I am also excited to watch what Bright Rail are doing and if that translates into a country wide approach to high speed rail.
That doesn't change the fact they are transporting half their freight by truck while the US transports a small fraction of that.. Also, US does have rivers and canal transport as well. Mississippi and Erie canal.
No one (generally) bats an eye when voting at the city level for a $100mm road package.
But trains?
Freight should move mostly by train. We could build rail lines where there are already highways, just swap a vehicle lain for a train rail, all that 18 wheeler traffic can be reduced significantly.
Not that simple, at least in the U.S. Rail has far stricter requirements on turn radius and grade. As an example the highways around here have 6%+ grade segments (after tunneling!) and rail is rarely over 1%. I haven’t even mentioned the incorrect subgrade and the massive safety envelope (approx 20x20 feet) which is about 2 highway lanes wide and far too tall for the thousands of highway overpasses.
Rail truly is a unique infrastructure concern and needs to be designed and laid on its own.
For most heavy passenger rail like Amtrak and things like Caltrain in the Bay Area, it's because those trains use the same tracks as freight companies and the freight traffic takes priority.
Caltrain tracks and right of way was purchased in 1991 from Southern Pacific (now Union Pacific) by Peninsula Corridor Joint Powers Board which is basically Muni, SamTrans and VTA for most of the route.
While it is true that Southern Pacific owned the line for most its history it was always built as commuter line. Union Pacific is only allowed to operate 3 freight lines per day on weekdays now.
Freight does not take priority, historically freight did not fund the development of the line, Characterizing it a freight line with passengers added on to it is not a fair representation.
--
I cannot speak for Amtrack, I don't know the network personally and it is also big and very composite a network to make any general statements like for Caltrain. However in general commuter heavy lines were built as commuter lines very early one because there was economic case to do so. It is the only the long distance lines which do not have the economic viability on their own that can only run on freight lines.
We already have a great interstate transport network that reaches every address, and where this cargo already travels.
I don't know what building a second parallel railroad system to take over this task would cost, but I'd guess several Iraq Wars.
Making the trucks autonomous doubles their utilization, which means we need half as many trucks, and makes freight cheaper, which makes prices lower, benefitting everyone, especially the poorest.
Having a railroad without a highway (and a highway won't come along and serve itself) will mean that you won't get to any city on your own. Wait for the train, maybe the railroad authorities will deign to launch a passenger train once a week, of such quality that you won't want to ride it. Like in some places in the Russian Federation. Or buy an off-road vehicle and go camping instead of traveling.
What sources do you have to cite on development costs and regular maintenance? Autonomous trucking is an add-on to existing trucks which are already being built regularly. We already perform road maintenance and account for it in local, state, and federal budgets as it services more vehicle's and destinations than just "autonomous trucks."
Freight railroads are alive and well in the US. Incidentally, it's also the side of the railroad business that is mostly deregulated and privatised. The passenger business is nationalised, and frankly a basket case.
(it's not a basket case because it's nationalised, it's nationalised because it's a basket case - thank the automotive industry - and it's a national security asset)
Agreed but it can’t happen anymore. Rights to the now-unused tracks have reverted to property owners. Impossible to restore w/out massive eminent domain issues.
A rail line is single use, single destination. commuter and cargo cannot mix (one goes high speed and one goes super slow) if they do mix it shuts the entire segment of the track down until it’s off that track.
Roads are multi use , multi location. It seems unfair to compare the cost based on road maintenance vs track maintenance. I would expect something to account for the sheer difference in volume each does
Why not road trains ? One lead driver with few of autonomous trailers ?
In a ideal world more rail roads in places it makes sense to develop and maintain dedicated track infrastructure, cheaper running costs ( fuel, tires, wear and tear etc ) justify the upfront costs.
Road trains are both intermediate solution everywhere and also can operate in routes where trains are not economical enough, even if there was a will to do it.
The billionaires in question (I think Warren Buffet is one of them) would benefit from more and better rail infrastructure. But the billionaires don't have enough influence to make that happen.
Similar for eg more open immigration, or not shutting down the US government down every so often.
The issue with braking distance and jackknifing could be resolved with smaller trucks. The 18 wheeler we know is optimal because it maximize the cubic volume driven by one person. If you don't need the human driver, two smaller trucks might be better than one.
The cab, sensors, and compute are also expensive, not to mention other variable costs like staff for remote assistance, maintenance, and first/last mile
Sensors and compute will get rapidly cheaper over time, because they are on the cost curve of electronics. (And they are getting cheaper even if autonomous trucking has only small production numbers, just like batteries will continue to get better, even if nobody builds electric cars: the whole electronics industry is working on those technologies regardless.)
The cab can be a lot cheaper, if you don't have to keep a human comfortable and safe inside. Also keep in mind that an autonomous truck can drive 24 hours a day, and doesn't need to take regular breaks throughout the day nor sleep. (They will need maintenance, but probably not more than a manned truck.)
You are right that the costs are real. Things like first/last mile (or loading and unloading) would probably need a major reshake of the industry, if the truck doesn't bring its own labour, in the form of a driver, with it.
I think that’s a great idea. It will increase the cost of the fleet but if it solves the blocking issues it may be worth it. Might even help with logistics and loading and unloading.
Heck, maybe they don’t even need to be articulated. They could look like slightly larger Amazon or UPS delivery vans. No need to worry about jackknifing.
If it’s a fleet maybe battery packs can be swappable so when it arrives at a hub to load/unload it can pick up a charged battery pack.
The only issue is for cargo that can’t be split up.
> No guarantee of a timely response from remote operators or backend services.
> Therefore, all safety-critical decisions must be made by the onboard computer alone.
Why is the requirement that all safety-critical decisions must be made on board, versus the seemingly-simplifying assumption that only some or most decisions would have to be made on board, because a remote operator or backend service could be available a lot of the time? It doesn't seem unreasonable to me to have a single operator remotely monitoring multiple vehicles that are autonomous under ideal conditions (driving along a straight road in good weather) and then taking over when necessary. Let's say you would only use such a system for major routes with solid satellite visibility, not last-mile routes hauling heavy equipment on a dirt road in the boonies, or something like that. Maybe this wouldn't work, but it's not obviously ridiculous to me, so I wonder why he just starts out saying the truck most be fully autonomous with no human ever in the loop.
Yes, and that has to be safely possible. From my reading, the article has no problems with remote operators giving command. The truck just also needs the ability to both safely execute them and come to some safe state by itself if necessary commands don't arrive [in time].
The interesting question is how complicated those safe states are to realize. Just hitting the breaks on a freeway seems hardly acceptable. Generally, this being harder than for cars in cities is a core conclusion of the article.
Author here. At the mean time between failures needed to exceed human performance, the uptime of the network connection quickly becomes a limiting factor. It’s possible to pick only routes that have great cell coverage but this limits commercial viability.
Satellite internet is fast becoming good enough and cheap enough.
As an additional safeguard, you can make your trucks go into 'safety' mode when connection becomes spotty or when too many operators become too busy with other trucks.
'Safety mode' could mean slowing down the trucks or even stopping some of them. And in general, letting the autonomous systems err on the side of caution more often.
I believe the MTBF argument still applies to Starlink.
Regarding the minimal risk condition / fallback behavior, a central point of the article was that slowing or stopping are almost always unacceptable on freeways because of the speeds involved
A centralized control like that coupled with a stop as a safe default would be a very juicy attack target. Want to cripple a city / country or create some chaos? Just jam or hack the control system.
Remote does not mean just sitting in a room far away. It could be simply one truck with a lead driver and few automated ones following it closely. Road trains are an option. Even just two autonomous trailers to a lead truck will reduce the driver costs by 2/3rds.
Unlike autonomous cars where everyone needs to go wherever, inter city trucks have fixed well known routes with predicable volume of cargo that can be easily chained together.
Ya, I wonder the same thing. In an extreme case it seems like you could build dedicated repeaters/cell phone towers/something along specific stretches of highway, and then start testing along just that stretch with remote operators on call. I suspect remote operators are cheaper than devoted drivers, potentially making a profit from that transition alone, and also collect training data to develop full autonomy.
You need at least a minute to get into the context and have full situational awareness, to imagine what it would be like as a remote driver with less lead time than that, think about a scenario where you suddenly wake up with your hands on the wheel, with no idea where you are and no idea what's going on, except the situation is probably bad.
Any kind of remote assistance will most likely not include steering the truck in real time over the internet.
Rather, things like "unknown obstacle on the road, can't continue", with the operator instructing it to just run over that empty cardboard box. All realtime safety decision-making should have safe autonomous fallbacks, a human cannot be expected to drop in and take charge in seconds. The entire level 3 / 4 "semiautomation" market is a bunch of make believe.
Cell networks are even longer lead time and more capital intensive than autonomous driving ;) even if we only consider the fcc + local permitting time, it already makes this option difficult.
This is the route May Mobility https://maymobility.com/ is taking with their AVs. I'm not sure where they're at now, but (I believe) they started by servicing old-age communities where the trips went to a few known destinations in low density areas.
If the AV ran into an issue, it could tele-help and have a human operator take over.
Completely makes sense that you could have a person monitoring multiple AV trucks and take over driving when needed.
Whoa hold on, table 1 of stopping distances is calculated for 2.5 second reaction time which is 10x longer than conventional human reaction time. Then the stopping distances are compared to radar/lidar/camera to argue that AVs can struggle to stop within detection range.
It's possible a computer might need 2.5s to make a decision to stop. But the current analysis isn't based on that.
This analysis seems really suspect to me. Any clarification would be appreciated.
Author here. the shorter reaction times you mentioned are collected under ideal conditions, like the person knows they are being tested and only needs to push a button or whatever. In driving, the reaction time is end to end, including perception, decision-making, and actuation (moving your foot, pressing pedal all the way, shifting, etc.)
Also recommend checking out the citation. It is an accepted value used in American highway design.
A human decision to stop is not binary 1/0. If I am approaching a hazard, first thing to do is raise alertness. Also, lift foot off accelerator pedal. Then, put foot on brake. Then, apply some pressure to brake. All this stuff happens on a gradient depending how quickly I need to react and stop.
So between which endpoints shall you measure "reaction time"? I'm already priming my reaction by actions which lead up to it. Obviously, an automated driving system doesn't need some of these steps, but it's still running through the same scenario and "thought processes" to reach a decision and course of action.
Once at lunchtime, I had two coworkers in my car. I was driving a manual-transmission Acura Integra in light rain, heavy traffic in downtown Palo Alto. I had a left turn signal and I proceeded through the intersection, when an oncoming car was coming straight at us and couldn't stop. I punched the accelerator and sped through the turn without incident. My coworkers both congratulated me on making the correct decision: to hesitate, or to hit the brakes, would've surely crashed us. But what was my reaction time? My foot already on the gas, I simply made a snap decision in the moment to follow through.
I had a guy do the exact opposite: he slammed on the brakes doing a left turn in front of me. I t-boned him. If he had had a passenger they would have been severely injured: he was driving a Kia, I was driving an Isuzu Trooper. My entire family was in the Trooper, and we were fine. The Trooper's radiator (and air conditioning) were broken, but we put in a new radiator and drove it another 150 thousand miles. That Trooper Would Not Die.
Sensors and their ranges aren't the right thing to point to. Off-the-shelf options are typically geared towards the ranges useful for passenger vehicles because that's where the volume is, but with money and time one can design something different. It's possible to achieve a sensible link budget for lidar or radar at much-longer ranges. The sensors will be bigger, they'll consume more power, and they'll cost more. But it's totally achievable.
There are a lot of differences between passenger vehicles and trucks. The physical dynamics of articulated vehicles, the mission profile, and social dynamics come to mind. How does a robotruck place cones or flares while it awaits rescue?
Personally, I expect autonomous trucking to be a force-multiplier for humans who were formerly drivers. Such trucks will have sleeper cabs and the human will be there to maintain the vehicle and handle the long tail of tasks (filling tires, cleaning, refueling, repairs, rigging, whatever). You'll get 24-hour operation out of a single human employee because they'll be able to sleep and do other things most of the time. Maybe they'll work a second job as a remote call-center operator.
Most high end sensors, especially lidars, are targeted at L4 applications. Otherwise the price cannot be justified. It’s a safe bet that sensor makers are including AV developers in their market research.
For lidar, the range is also limited by power limits + physics, which cannot be overcome by increasing money/power/device size. Some dependencies on semiconductor manufacturing tech or better signal processing might be possible to solve with more money.
LiDAR transmit power is practically only limited by eye safety. And retinas are fixed size while the aperture of your transceiver isn't. Get a big lens and you can transmit a lot of power and collect a lot of reflected photos.
Long-haul trucking is potentially easier but we already have a great system for delivering large amounts of cargo along fixed, guided pathways. The only problems that making it run on roads instead of rails solves is political - the US is allergic to owning and funding rails but will happily dump tens of billions into its government-owned roads every year.
Short-haul trucking (in-city delivery etc) is where the value is in having agile, self-driving vehicles moving cargo away from supply points… but it’s also the hardest to implement because there are so many dynamic elements in city driving. And largely these light delivery vehicles are already being electrified where it makes sense (UPS, USPS, etc). You are mostly talking about it any advantage coming from removing a human, not the act of rolling electric vehicles into it like taxis where it’s primarily disrupting internal combustion with electric.
Also, any system is eventually going to be utilized at its full capacity. The number of vehicles supervised per agent will increase until the average number of incidents at any one time meets the average number of agents, or beyond, until there is a regulation. Capitalism is a cost optimizer and no cost will ever go unsqueezed - much like the internet is designed to route around network damage, capitalism is designed to route around any such inefficiency as morals or safety standards until otherwise compelled by regulation.
> The only problems that making it run on roads instead of rails solves is political - [...]
You say it like this is a point against autonomous trucking?
Technical problems are solvable. Political and social problems are basically intractable in practice.
If autonomous driving can turn 'delivering large amounts of cargo along fixed, guided pathways' from a political into a technical problem, then this would be a jackpot.
> [...] capitalism is designed to route around any such inefficiency as morals or safety standards until otherwise compelled by regulation.
No? Plenty of eg cars exceed minimum safety standards. Some brands, like Volvo, are even explicitly sold on safety. Most companies explicitly talk up their morals and ethics, too. Look at almost any old advertisement for examples. Ethical brands are quite popular, and whenever one is found to be only pretending, there's usually a big scandal. So many of them actually practice what they preach.
Similarly, most people are paid more than regulated minimums. According to your theory, that shouldn't happen.
>No? Plenty of eg cars exceed minimum safety standards. Some brands, like Volvo, are even explicitly sold on safety.
In the context you're discussing, that form of safety isn't an inefficiency but a marketable feature. That doesn't mean that safety in general is a marketable feature.
Those safety standards tend to omit the safety of any pedestrians being hit - otherwise, cars would be far lighter (and with bonnets far lower), because a heavier car benefits the occupants in a crash at the expense of whoever they're hitting.
You are right. I agree that cars should be taxed (at least partially) based on weight.
However, you don't really need new regulations for this. Tightening liability for drivers, instead of letting them get away with a slap on the wrist when they kill a pedestrian or cyclists etc would probably suffice.
At the moment, we mostly do victim blaming instead. When you operate a death machine, you should be responsible for any kills, and not be able to hide behind 'it was an accident'.
> Tightening liability for drivers, instead of letting them get away with a slap on the wrist when they kill a pedestrian or cyclists etc would probably suffice.
That really can only happen if the roads are improved enough that drivers can’t cast reasonable doubt around when an accident occurs. Like a road used as a main car route, but has lots of on street parking to obscure pedestrians and bicycles, and also has a lot of them. Yes, the car should be liable, but when the report is taken they find that the car’s view f the victim was completely obscured anyways. We have to really go Dutch on our road design before we can go Dutch on our car liability enforcement.
What do you mean by reasonable doubt? If you are in a car and you hit someone, it should be your fault.
If the road is bad, you should drive slow enough that it's not dangerous.
If you can't see anything, because there are cars parked, perhaps you shouldn't drive, or only drive very slowly. It should be your responsibility as a driver to only drive when the view ain't obscured.
When I'm eg running around with a knife, or discharging a gun and injure someone, I don't get to make excuses like the above either. Why should drivers be allowed to blame the victim or circumstances?
Again, the fact that the roads are designed so poorly for this still gives drivers too much sympathy from authorities, regulators, and the general public. It literally can happen to anyone who decides to drive at all, given that visibility is so messed up.
The ditch will heavily sanction a driver and go through a likely redesign when an accident happens. You can’t just expect drivers to do the right thing, you have to also make right things possible. The American approach of “let’s require drivers to develop precognition” is just ridiculously stupid.
The American approach is just idiotic all around. It gives plenty of plausible “accident” explanation given crappy road designs and the lack of will to fix them.
I tell my kid that it will be the car’s fault if they hit you when you are walking out into the street behind some parked cars, but even if they pay some money you’ll still be dead. This is not just about cars being liable (like seriously, get rid of on street parking if you are going to have lots of cyclists and pedestrians on that busy through fare, Seattle sucks at this).
This is a video from our firefighters here in slovenia... Humans are shitty drivers, sure, but how the hell would a computer react to a chaos like this?
Pulling to the side when the traffic jam starts, not only when the firefighter walks up and knocks on the window.
Also, since it's a relatively time-insensitive situation (as in, tens of seconds to deal with it rather than seconds), once it identifies that it's an exceptional situation, it can have someone connect to the truck remotely.
> Pulling to the side when the traffic jam starts, not only when the firefighter walks up and knocks on the window.
And get overtaken by other drivers :) Then a lot of honking and swerving would be needed to get back in the correct lane. Also they sometimes need to back up a bit, to let a 'late' driver to move infront of them, or behind them, move to the other side, interpret what honking means to whom, etc... or drive forwards a bit, or whatever the firefighter knocking on the window tells them, speech recognition is still not perfect and interpreting abstract orders is even worse.
> it can have someone connect to the truck remotely
This is covered in TFA, and noted that one cannot rely on wireless connectivity along all highways. Certainly not immediate responses, and doubly-so when the vehicle may be in a deadzone; either a natural one or one caused by overloading the local towers (such as one might find at a remote accident with lots of other motorists) will do.
I think there are some good points here but IMHO this is overly focused on flawless
100% of the time L5 autonomy. Many autonomous trucking companies can become economically viable without perfection because 1. They aren’t dealing with consumers directly 2. Can control or focus on specific well understood shipping lanes 3. Can provide more human in the loop assistance for tricky situations. In this way trucking is easier than ride sharing because there is a longer on ramp (no pun intended) for companies to improve tech while being viable businesses
I often see logic along the lines of "if autonomous vehicles are safer than people, then let's deploy them" and that logic straight-up does not fly in the real world. In the real world, among the techno-pessimists, that is an outrageously low bar. Most accidents and deaths that happen because of human drivers occur because the driver was doing something illegal, which means the bar that we have for even humans is higher than "safer than human drivers" -- like, we wouldn't allow someone to be an Uber or Lyft or truck driver if they were candid and said "I'm going to text on my phone, and be drunk driving and sleepy and distracted as often as the average motorist".
Also, I feel like there's a lot of talking past one another in these conversations because one person will say "Let's see an autonomous truck shipping hazmat to Pittsburgh in February with freeway lanes shut down" and another person might say "that's a rare instance" but I really don't feel like society will accept anything other than trucks / vehicles that are able to operate under all conditions, with greater safety than the safest human driver. We tolerate human failures but to use them as the benchmark for autonomous systems would be perceived as unethical, because autonomous systems are deliberately designed and any failures by them would be seen as an intentional oversights and errors, and no one at Waymo or Tesla or where ever is ever going to be charged with vehicular manslaughter for an autonomous vehicle error. We'd demand a way higher standard because these companies don't really have any skin in the game, except for financial penalties which we now understand is not a deterrent for anything. My observations are only moderately related but I'm anticipating the same well-trod talking points coming up and want to address them.
I agree achieving human safety equivalent is the minimum bar. Ex: We can all agree that if your system is below human safety, it is definitely unacceptable.
But issue is that human safety is kind of long tailed. Eg more than a third of all fatal crashes are DUI. I’m guessing if you take out high risk behaviours then the rate will be an order of magnitude lower. What we care about is not being better then an average driver (which basically is bad due to high risk individuals), but better than median driver.
I think better automated collision avoidance as well as automated “you’re driving erratic take a rest in a safe location, we can help get you there” are clear wins. But forced autopilot is definitely not there yet and would require a lot of improvement over average driver (because I don’t want to increase my personal risk in order to decrease risk of high risk individual).
Yes but this misses it, autopilot being better than a drunk driver but worse than the median driver isn't enough unless they only use it when they were going to drive drunk which is kinda a political nonstarter. Otherwise the overwhelmingly sober usage of the feature makes it a net loss in driver quality weighted by miles driven.
The problem is that it is better than average human performance but it needs to be better than specific human performance for any human to be incentivized to switch.
For now what you could do is to demand that anybody that has a DUI or other such item on their record to mandatory only be allowed to be in vehicles that have self driving if the manufacturer is willing to assume liability. And if that doesn't happen then they might as well take a regular cab.
Yeah, I should clarify that my response is in anticipation of the comment sections I often see on hackernews about self driving cars, rather than arguments from the self driving car companies themselves. Waymo won't say "we're slightly better than people, let us on your roadways" but I feel like every time I see something about self driving on Hackernews there's a handful of commenters taking the hyper-utilitarian viewpoint of "they're better than people, we all need to let the robots do driving for us" which will never convince anyone outside these comment sections.
They don't operate in snow now, but everybody was like "This is great, but it doesn't work in the rain!" this time last year. I've now taken many flawless rides in outright downpours in SF. It might take a year or two, but I have no doubt snow will be something that is a non-issue in the near future.
Sensing tech has to improve to get through the noise. As far as I'm aware, you can't see a yellow line or white line on a highway if it's covered in dirty snow.
We could also rework our roadways to include better sensing design and tech (passive or active!), but we are a ways out still from willingness to pay for that.
Do what humans do and drive without lines while trying to stay to one side of the road. This doesn’t seem like a major problem to self driving snow cars. More like, how do you deal with other drivers who are RWD with summer tires and are fish tailing all over the place.
If the sensors are up for it... Mountain roads covered in snow pack are a white blur, especially in certain light. Best bets for actually knowing where the road is might be previous tracks. I've dipped a wheel in a ditch more than once, not from carelessness, but snow drift obscuring the road... A road that I drive 500 times a year
I don’t drive at all when it snows as a human. Our whole city basically shuts down then, and we are pretty far north as far as cities go.
Even in places that expect snow, cars are moving much more slowly and cautiously than normal. It almost seems like self driving cars would do much better in those situations given the speeds and caution involved.
That's the smart move. I've been quite astounded at family members without a driving license pushing family members with driving licenses to drive when they thought it wasn't safe. If it snows and it isn't 911 level urgent then simply stay put.
What is the point you're trying to make? Waymo isn't making claims about their driving performance in snow. And they're not comparing their performance to using a basket of human performance in all weather conditions, if you read the linked blog post they're specifically comparing to human driving performance in the regions in California and Arizona they're testing in.
Honestly, I never want a self-driving vehicle of any sort. I like driving, and if my going on about 8 years since even a speeding ticket is anything to go by, I'm pretty good at it. Driving is genuinely a leisure activity for me (only complicated by other, shit drivers) both in the usual way, and in the track-day way. So you could say I'm a car guy for sure.
I am all for autonomous vehicles for other people. And you can call that elitist and I frankly couldn't even argue with you, but god damn, so many people have utterly no business being in control of a car. I travel quite a bit both for work and leisure, almost always by motorway, and some of the shit I see is just mind-bending. So much inattentive driving, so much pointless risk-taking, as they say it's time for Wisconsin's favorite game: Why's that car being driven badly? Old, stupid, drunk, or all three?
And like, there's not really an answer for a lot of this. As much as I love them, cars suck on a society scale. They're basically a poor tax as even the most low-income people in my neck of the woods must have them to get around, there's a bus network but it's shit and it has very low ridership, and that's all my city has apart from taxis. And, old people use them for the same reason, even if they're damn well aware that they shouldn't really be driving anymore, if they don't have someone to get their groceries n such, what are they supposed to do?
So yeah, self-driving cars are awesome. In theory. But then I see video of someone using the lane-guidance on a Tesla where the car just sees... who knows, something, and suddenly jerks to the right, right at pedestrians. Or we get the stories of autonomous cars just shutting down and refusing to move, even for emergency vehicles. Still others, they end up parked on someone's legs, and the stupid support system can't allow a remote operator to move the vehicle off said leg.
At this point, as much as I hope we get it, I also kind of recognize that truly autonomous cars simply might not be viable yet. I think a much better idea is to just de-carify our societies. I don't want them gone entirely, but people need more options than JUST buying or leasing these enormous, polluting, dangerous machines that they do not want to learn to operate well. And like, I don't even judge people for that necessarily? I love them, to be clear, but that's something that's true for me and not necessarily everyone else, and if you don't care about something, you're not going to put your best effort into it. It becomes a chore. A necessary step to doing what you actually want to do, and as a result, people don't try and they just suck at it.
Edit: I'd much prefer and think it's more practical to create cities where cars are an option, not a requirement. If you WANT one, you're welcome to have one. And maybe it drives itself, or maybe you drive it. With those options being on the table, we can then raise the bar for driving tests which we badly need to do, there needs to be a higher skill floor for driving and more stringent requirements, especially as people age. Maybe once you get to a certain point in your life, as most of us probably will, you just can't have a manual car anymore. Then you can buy one that drives itself, or just join people on a robust public transit system.
This doesn't need to be hard, but it's made hard because tons of entities involved have a financial stake in keeping us dependent on cars. I think that's why self-driving cars are getting so much traction and funding, because then they get to sell you a new, more expensive car, instead of actually solving the problems of mass transit. And that just sucks.
> but I really don't feel like society will accept anything other than trucks / vehicles that are able to operate under all conditions
Trucks are different because they're just for transportation of goods. As challenging weather arrives they can just wait at the nearest rest stop, pull over on the shoulder, etc.
Your shipping will be delayed, but outside of that I don't think society will care. Totally acceptable as long as weather delays only last as long as a storm does.
That's different from rideshare where people do expect to be able to call a Waymo even if it's snowing lightly.
Society is literally willing to let truck drivers die to maintain shipping times. See the responses during the height of COVID-19, or even just the demands placed on every day operators cross-referenced with their average health and mortality rates. Our treatment of truck drivers is already subhuman, I don’t think we’re likely to extend a lot of understanding to the robots.
That's a bit of hyperbole. Lots of jobs take their toll on your health. That's nothing special about truck drivers, and it doesn't mean we "let them die".
And you're ignoring the fact that not requiring a driver is significantly cheaper. And when it's speed versus cost, cost generally wins. So yes, society is absolutely going to be fine with trucks pausing in bad weather, because it means cheaper shipping.
(And stuff that is urgent and requires next-day delivery is sent across the country by plane anyways, not by truck.)
On truckers dying on the job: you'd probably agree that there's arrangements that could save more truckers' lives but are not applied because of cost and lack of social pressure. For instance driver fatigue is said to be the primary cause of death among trucker, and that's obviously an issue we could throw money at to at least alleviate (but it wouldn't always be cheaper, that's the point)
And to your point, yes, that same logic would aplly to a ton of high risk occupations. There's building construction techniques that cost more but have less risk, but we won't apply them where there's no pressure to do so.
Are we letting them die then ? I don't see the hyperbole in saying yes, as a society we're not willing to put the money to reduce these deaths. I see it as a part of life, and don't think we'll change much even as autonomous vehicle come (if they kill 1 in 10000 people but cost way less than now, we'll swallow the pill collectively)
Driver fatigue is taken way more seriously than it used to be and measures and laws have been put in place to minimize it by forcing maximum driving and minimum resting periods.
So, ignoring the fact that society doesn't just get together every year and decide to let people die, I'd say it's absolute cynical hyperbole to say society is "letting truckers die".
"Taken way more seriously" is a step on the spectrum. Where did we put the cursor ? At the point where the cost vs the number of deaths kinda makes sense.
I think it's a poisonous to ignore that tradeoff and wash our hands of the whole issue. It's like eating meat ignoring where it came from. That doesn't mean we all take extreme positions and become vegetarians, but it's not something we can completely ignore saying it's hyperbole or cynical to recognize as a fact.
For the same reason, we'll likely be absolutely fine with them driving a bit slower as well once there's no driver to pay by the hour - that'll save money not just on the automation but on fuel cost as well.
That might be more a result of capitalist motivations i.e: funnel money to the share holders / owners and C-suite and away from workers. Which is much more direct than someone buying legos or a stove.
We can either legislate that or they need an effective union.
"That might be more a result of capitalist motivations"
People wanting to get more stuff faster is a universal motivation not limited (in theory or in practice) to capitalism. To the extent that anyone is exploited, that's just selfishness, again a universal motivation.
This is a bizarre framing to support the argument people don't like faster delivery of goods.
But in the spirit of good faith I will give some:
- Movement of fish/meat to markets before it spoils
- Delivery of military messages around the battlefield. Also supply trains for military expeditions
- Projects like the Grand Canal in China (https://en.wikipedia.org/wiki/Grand_Canal_(China)). Pretty much anywhere in history where people opted for using water transport of goods over land transport was done for speed/cost reasons.
I mean the copper ingot guy was sort of complaining that he hadn’t gotten enough (good) copper ingots and that his couriers had had to make multiple trips and had come back empty handed each time. This isn’t exactly “more stuff faster,” but it is pretty close.
The free-gratis supply of bread; as a daily life-preserving staple; to a portion of the population doesn't appear to match the spirit of "more stuff faster".
Surely basic food supply is "the bare minimum of stuff" and the rate is fixed to 'each day' (and the delivery to grain stores is 'each harvest in the supplying region'? Yes, it's more as populations grow and centralise, but that appears to be occurring the context of the question, no?
>Trucks are different because they're just for transportation of goods. As challenging weather arrives they can just wait at the nearest rest stop, pull over on the shoulder, etc.
Trucks? waiting? In "just in time" culture of storage management, where buffers are low and delays are nightmare?
The same problem exists with manned and autonomous vehicles here though, it seems. Only for autonomous vehicles you don't have to worry about the safety of personnel. They might need digging out come the end of the storm?
There are way more trucks on the road than will fit in parking at rest areas, and simply parking on the shoulder is a hazard in itself, which is why trucks are required to set out warning triangles, etc. when they have to do it.
> There are way more trucks on the road than will fit in parking at rest areas
Not the ones that are taking multi-day trips. By definition, we've already made room for them since truckers generally sleep at night all at the same time.
This isn't about trucks on a 3-hour haul -- those just won't head out at all until the forecast is good.
And parking on the shoulder is a worst-case scenario if something suddenly happens that wasn't in the weather forecast for the next couple of horus.
The bar for commercial driving should definitely be higher than for “civilian”.
But I think the bar for civilian is also woefully out of date now. Giant trucks should require a commercial license. Make it easier to get than a delivery truck license, but weed out the soccer moms and bring back the station wagon. If you’re a professional tree trimmer, general contractor, or a forester, renewing your license is on the clock and not a big deal.
First of all for vehicles with total mass of above 3500 kg you need extra license. And also separate one if you have a trailer. This is get it and keep it, until certain age. But good enough often.
And then in general commercial operations also need more licensing. Which needs to be renewed after certain time.
This is a great idea and the first I've heard of it. Set a maximum weight of 5,000 lbs for an ordinary license. Minivans will come in just under the line. Small SUVs like the CRV or RAV4 will be allowed, as will small trucks like the Ford Maverick. Indeed, most of those vehicles are under 4,000 lbs. Even some larger SUVs like the Honda Pilot or (just barely) the Toyota 4runner get under the bar. Even a low-specced F-150. But nothing bigger than that.
Truth is, I'd prefer the bar at 4,000 lbs, which would limit us to Camrys and CRVs, but 5,000 would really allow just about any reasonable vehicle.
And big heavy EVs with massive acceleration are just too powerful for somebody who's got a Starbucks in one hand, a cell phone in the other, their knee on the wheel, and shouting kids in the back seat. "Pedal misapplication" will go tragic really fast.
Absolutely. Those iconic 70’s car profiles will never be seen again because they suck pedestrians under the car and kill them. Trucks have some loopholes for this.
Even the Cooper Mini had to get taller about 10 years ago so the hood angle wouldn’t flip people head first into the windshield. And for side impact ratings. Those two are why the mini is so huge now. (I parked my midsized sedan next to a countryman the other day and when I came back I staggered and rolled my eyes because it is 20% bigger than my car in height and almost in width. Wtf).
It would also help to make it expensive to maintain/own vehicles over certain wait. California charges weight fees, but only for pickups, no car or SUV gets that weight fees.
All other things being equal, that's true. All other things are not equal though. The 4000lb+ vehicles often have worse visibility, so you have to pay more attention to make sure a kid didn't wander in front of your truck while you were waiting to make a right turn.
>A distracted driver can kill pedestrians just as easily in a 3999lb vehicle.
No they can't - a lighter vehicle has a slightly harder time killing people. It's still incredibly easy, but the pedestrian fatality rates do go down the lighter the vehicle is.
So it's OK to have more deaths so long as we can punish someone with no expectation that doing so will fix the problem? I suspect you are right, but I don't think it reflects well on us.
I think a way to think about it is that we (at a societal level) wouldn't accept a desk calculator to be sold as a consumer product if it did correct arithmetic almost all of the time but then would spit out a close but wrong answer 1 time out of 1000, even if that would be significantly better than the average person doing arithmetic. Our expectation would be that it could (and should) perform flawlessly. If our technological progress was such that the only calculator we could ever hope to produce would still have that kind of error 1 time in 1000, would it be unethical to prevent that from being sold? That's hard to say!
One of the heuristics built into us (because we're mortal beings living beings in a competitive, historically resource-poor environment) is that we trust the devil we know more than the devil we don't, and so unless there's a strongly compelling reason to trust autonomous driving devices a lot more than humans, there will be some inertia against using them, even if the calculations are that it will save X number of lives. I mean, inherent in that calculation is a level of uncertainty and people don't necessarily trust that number, because they don't have a reason to trust it, because they haven't really seen enough to trust it. Why take a company's word for it that it's safer when they have a financial incentive to do some creative stuff to get their marketing pitch? I would say that if you feel it doesn't reflect well on us, it's because it hasn't been thought about enough.
I think intuitively the desk calculator analogy makes sense, but from a technology perspective it does not. A calculator has a limited number of operations and operating modes, and you can feasibly test all of them, quickly, easily, and in an automated fashion.
An autonomous vehicle has an uncountable number of operating modes, and it is not feasible (perhaps not even possible) to test it in all possible conditions and states. Even if you could, doing so for (say) every single software change would take years each time.
Maybe that does mean that this is a fool's errand, and we just shouldn't be building autonomous cars, at least not until we have AGI that can think and act faster and with better judgment than a human.
I personally do think that "better record than a human driver" should be sufficient (perhaps with some significant, TBD margin; 0.1% better is probably not enough), but I accept and agree with your toplevel comment that sort of thing won't fly in the real world. The bar is really more like the self-driving car has to avoid making the specific kinds of mistakes and illegal/unsafe acts that a human driver would do (all while not creating new classes of mistakes that a human driver would not make), and, on top of that, be better than a human driver in situations where crash would not be deemed that human driver's fault.
I don't disagree that a calculator and a car are different things, but the important thing for the analogy is that the wider society expects the same thing of them. When John Q Public thinks about self-driving cars, he thinks to himself "How hard can it be? Drive the speed limit, stop at stop signs and traffic signals, go when lights turn green, stop at obstacles and stay between the lane dividing markers. If these vehicles can't do that then what good are they? Why risk letting those things on the road if I can't even trust them to do that?" Obviously you and I both know that the base things like "recognizing a stop sign" required decades of research to become reliable, but like, back in the 1960s before the problem had work started on it, even PhD holding computer scientists were thinking that object recognition was basically a done deal.
What I want to keep reminding the software developers with consequentialist ethics is that the entire rest of the world operates under a totally different mental framework than the techno-utopian one. In that mental model, a self driving car is an appliance and it will be judged by the standard of all appliances, which is that any catastrophic failures that happen during their normal operation are evidence of a defective, untrustworthy product. The problem they're trying to solve and its complexity has little bearing in that judgment. What incentive does anyone have to be generous and forgiving about potentially fatal errors when you're not taking the consequentialist viewpoint for granted? That's the thrust of my observation.
I was thinking about this and one way of thinking is what u said. We already allow sale of spaceships which have non-zero failure rate. So we don't necessarily need zero failure rate self driving.
But a problem with cars is, usually ur malfunctioning calculator just harms u, but a malfunctioning car will affect people who don't agree with ur choice of driving non-zero failure rate self driving car.
The rest of gen AI tools might prove that wrong- Need help with your homework? We have software that will only sometimes give you facts that are incorrect. The res of the time it does great!
> we wouldn't allow someone to be an Uber or Lyft or truck driver if they were candid and said "I'm going to text on my phone, and be drunk driving and sleepy and distracted as often as the average motorist"
Half the taxi and Uber drivers in New York are perpetually on a phone call, and frequently interacting with their phones.
I'll take your word for it (not being an American and only having visited NYC once), but even then most of the rules of the road are… not perfectly enforced. I suspect if the rules were perfectly enforced, the only humans allowed to drive would be those who actually don't.
I suspect — no, I hope — that anyone who admits in advance that they intend to break the rules, won't get a license.
Not sure which side you're arguing. Does the US even have automated speed traps? Every time I hear about them, there's opposition and it doesn't happen. I think the most recent one I heard about in the US, the plan was to issue a ticket unless the speed is at least 11mph over the limit, which seems... both fine but also kinda silly?
Red light cameras were even a big fight to get implemented back when they were a new idea, though they're relatively uncontroversial these days.
To me, the difference is that the speed limit is intentionally poorly enforced, because law enforcement knows that speed limits in many places are set inconsistently and at unreasonable levels (and I believe that's also intentional; I recall reading that limits in many places are set at the 85% percentile of actual measured traffic speed, or something like that). An automated speed trap can't judge whether a particular speed is safe for the current weather and traffic conditions.
On the other hand, it's never safe to run a red light; that's just a binary "you did it"/"you didn't do it". Yes, I know, technically it is safe to run a red light when visibility is plentiful and there's no cross-traffic within sight, but I think we as a society have accepted the idea that you just shouldn't run red lights, and getting ticketed for doing so is fine under nearly any circumstances.
Really the bar for humans to legally operate a vehicle is whatever licensing process is in place in a given state. We don't make them pinky swear to not do anything dangerous or illegal.
If an autonomous system can get a CDL, it's probably gonna be more effective at continuously meeting that standard than a lot of the humans that do the same thing (but are distracted on a given day or have started using substances, or didn't sleep well or whatever).
> Really the bar for humans to legally operate a vehicle is whatever licensing process is in place in a given state. We don't make them pinky swear to not do anything dangerous or illegal.
That is not sufficient for self-driving vehicles. The license is the thing that shows "okay, this person seems safe" but then some of the things that keep them safe are the threat of accidentally killing themselves, or being arrested and put in jail for a crime they commit while driving, or the financial penalty of being sued for an injury / damage they cause, or the risk of having their license taken away for errors. If a human being applying for their license was invincible, incapable of being jailed, sued, or having their license taken away, you might expect that our CDL or other licensing processes would be more stringent.
Wait, why do you think that permission to operate the autonomous vehicles would not be withdrawn? It's happened!
It's totally bizarre to give mediocre humans credit for their self preservation and then demand that an autonomous system be beyond superb because it doesn't have it. Statistically, the mediocrity is going to be the bigger problem.
Maybe, but our driving tests are designed around assessing humans. A system that fundamentally works quite differently likely has very different performance characteristics and failure modes and may need to be assessed differently to demonstrate that it will probably perform adequately in non-test conditions.
> We'd demand a way higher standard because these companies don't really have any skin in the game, except for financial penalties which we now understand is not a deterrent for anything.
I'll add that one of the penalties for accidents caused by human error is a prison term.
Until a Tesla/Uber/Waymo exec responsible for autonomous driving can serve a prison term for accidents caused by their service, the penalty for accidents caused by autonomous driving is orders of magnitude lighter than those applied to human drivers.
We should also demand that these companies are run by engineers, not MBA types. Otherwise, we'd end up with neglect of safety measures as seen at Boeing. And I'm not sure if any kind of external incentive could change that.
> like, we wouldn't allow someone to be an Uber or Lyft or truck driver if they were candid and said "I'm going to text on my phone, and be drunk driving and sleepy and distracted as often as the average motorist".
But we do allow it. Sure its illegal but we don't put in the effort to actually prevent it. Just because we don't like it and would want it to be better doesn't mean we aren't allowing it.
> We'd demand a way higher standard because these companies don't really have any skin in the game, except for financial penalties which we now understand is not a deterrent for anything.
As we have seen with the titan sub, not even death is a full deterrent. A lot of companies will still risk it for the biscuit and take the risk of getting jailed for their shot at leading the market (in fact, a lot of people risk actual jailtime for financial gain). Having strict regulation and not allowing subpar cars on the street really is the only way.
I agree, though I either don't see (or I see and misperceive) many examples of people mistaking the utilitarian result for the politically acceptable one.
The problem is that if you have safer than human systems but you don't deploy them it's basically killing people. If you had autonomous driving that was 90% as deadly as the average human driver, not deploying it would cause 4000 deaths every year in the US.
Yes. There are a lot of people who are driving for things that a good public transit system could handle. Sure you can have your SUV for towing the boat to the lake - but you only do that on weekends (and no every weekend) - why are you driving to work, to church, get groceries - for most people those are tasks that that a good transit system could handle.
Note that by good transit system I mean one where you never check the schedule because the next bus/train is never more than 5 minutes (that is a maximum time). By good transit I also mean a system of transfers to express service so you can get to more distance places quickly. Nobody in the US has a good system (NYC's subways don't even come that often!), but there is no reason we can't have that, and some places in the world do have it (at least for a few).
> the next bus/train is never more than 5 minutes [..] there is no reason we can't have that
There's no logistical reason, but very few municipalities have the density to support that kind of frequency (via fare revenue), and taxpayers (especially in the US, but in other places as well) don't want to pay for all that unoccupied space on trains and buses.
I'm not saying that's a universal truth. There are a few places that do have the density to support that. And there are a few places that don't, but local residents are happy to pay more in taxes to get that kind of frequency. But most places are not like either of those things.
Yeah I’d totally take transit so long as it will pick me up in my driveway, is ready within a minute of me deciding I need to leave, and is a ride for only me and my family/friends.
This attitude is why we can't have nice things. You have probably never experienced a system like I outlined and you won't even consider it without adding other requirements that drive up the cost until we cannot get it.
If instead you would accept that you can wait up to five minutes, and "other people" are nice enough and so you can share a space with them - we could have a nice transit system.
The five minute time I used above is not arbitrary. While there is some debate on the exact number, 5 minutes allows for a lot operational things that are really nice.
The same way as getting out of the house during a pandemic was similar to "killing people" for a big chunk of the educated Western electorate.
You can see how that viewpoint might be received in a lot of places nowadays, especially blue-collar places that rely on that type of work to still be available and not farmed out to some AI bots that will manage to also kill some of us in the process (but in smaller numbers compared to what our fellow humans would have registered, that's a relief).
Many people are safer than average driver, since most deaths are associated with high risk driving (more than a third are dui alone). So forcing everyone to use autonomous vehicles essentially kills safe drivers.
since this is literally the trolley problem, the issue here is whether you want to take authorship of the deaths or not. I personally never move the lever in those puzzles.
That's only if you're able to deploy it in a way that the human drivers you replace have the same risk distribution as the human driver population as a whole. If your deployment instead skews away from replacing the high-risk human drivers, the 90% AV could make the overall situation worse.
Truck or cars are not strictly required to operate in all conditions in the current system: roads, entire freeways close when the conditions demand. Or demand that vehicles stop and put on chains. There is precedent.
I think that if you were reading my comment charitably you could infer that I meant "under all conditions where a car is legally allowed to operate." I don't think you actually believed that I was saying that a truck should have the capability to drive on the road when a roadway is not legal to drive on.
There is precedent for either roads or vehicles for choosing not to handle some conditions. In a driving rain storm you will find rest areas stuffed full of trucks and cars that picked that time to nap. In San Francisco Waymo first served some hours and not others, some areas and not others - some of these choices apparently related to fog and wet pavement. In Oregon, you will find roads open, but white-out-ed. Etc, etc. Nobody forces human vehicles to keep moving and most of the ones with some sense make choices. Which are available also to automated vehicles. - Sure it should be expected of the vehicle to find a safer place to move itself than right in the middle of the lane, like I point out in another response here.
> Truck or cars are not strictly required to operate in all conditions in the current system: roads, entire freeways close when the conditions demand. Or demand that vehicles stop and put on chains. There is precedent.
I think you're confusing things. Weather changes midway anyone's drive, and all drivers are required to drive safely and reliably even during sudden extreme meteorological events.
For a road to be closed, it takes an administrative action that reflects a decision that's largely arbitrary. Until a third party makes that decision, any driver is required to drive safely and reliability, regardless of the weather.
This article is certainly interesting, but none of these arguments are new so it could have been written five years ago.
It would be interesting to hear what a company like Aurora and their backers were thinking back in 2020. It seems unlikely that they did not come up with these arguments themselves, so what were the counterarguments?
Aurora backer here (small investor LOL).
Their thinking is that they can make money.
The "driverless bar" presented in the article is pretty meaningless.
I've been observing that AI tools often don't really replace people, but rather amplify the amount of work that a single individual can do. In that way, rather than seeking to replace truck drivers, it might first be more useful to make it possible that one driver can navigate two trucks in tandem, and later perhaps more. The largest issue with autonomous driving seems to be the ~20% edge cases that may occur, and have high variance, and ~80% could be automated. So perhaps at some point a single driver could drive more than one or two trucks.
It seems like there are a lot of 'default routes' that trucks take, and in the beginning, it'd probably just be a small number of highway routes that could get clearance in such a way, and then scale up.
Yeah that would be ideal. Sadly a lot of stuff goes wrong in building rail because governments have to do it and clearly suck at it, and if it weren't so, there wouldn't be nearly as many trucks on the road.
Unlike building rail, the concept I outlined would be feasible to do for a private corporation though, even a startup. The delta in necessary resources between the two is huge.
I'm also not quite convinced that you can pack nearly as much cargo on a rail at the same time (with different destinations) as on the road, but the reason for that believe might be the fact that I'm not a rail expert.
Fortunately not all governments are terrible at rail infra planning/upkeep. Perhaps autonomous trucks as you outline could do the last X miles from the point the train offloads.
The US is number 3 in the world in both total tonne-kilometers and tonne-kilometers per capita shipped by rail, and it’s in the top 10 in percentage of freight moved by rail.
You’re replying to a comment about freight rail, in a thread about freight rail, and you mentioned autonomous trucking. I’m thoroughly confused how you could possibly have been talking about passenger rail.
I don't really see the point in anything long range for autonomous vehicles in the near-term. We should optimize the use-case of them driving people and goods to and from train stations. For this use-case it should be entirely fine if the max speed they can achieve is 20mph. The stopping times and force that can be achieved at those speeds should bring down the danger by quite a bit.
Do roads have to go more places? Are they dramatically cheaper and easier to make and cheaper to maintain than train tracks? Or do we just already have a bunch of them?
Mind you I'm not taking about last-mile dirt road up a mountain to one family, but the massive arteries with enormous throughput that constitute the vast majority of lbs/mile of shipping.
Railroads have much much stricter requirements with respect to maximum allowed curvature and slope. So yes roads can go far more places than rails can.
I think I would be more comfortable with triple trailers with parking assist and stability control than with fully autonomous trucking.
Growing up we had a neighbor who drove UPS trucks and was rated for double trailers, and that earned him extra money. I gotta assume triples make even more. Doubling the payload per driver mile by making it so any trucker can manage doubles and the old double drivers can manage triples seems doable.
The thing with doubles and triples though is that they are shorter trailers. And it’s probably less likely to pack a triple trailer to the gills than a single, so are they getting 70% as much payload per trailer? If so that puts triples around 2x the practical capacity of a single.
I'm a diesel engine mechanic by trade, and having been to a few mid america trucking shows I can definitively say the autonomous trucking scenes pretty dead in the water. Too many "ideal condition" technologies, not enough edge condition focus (lots with no turnaround, lots with poor traction or mud,etc...) And too many companies demanding infrastructure investment that won't come.
The in cab electronic log system is still largely ignored because states want money from inspections, no autonomous system can handle a tandem adjustment, and everyone wants to put sensors all over roads that barely see replacement let alone investment and dont have maintenance factored in.
Edit: ultimately these companies miss the mark. Autonomous cannot cost more to maintain and operate than traditional fleets and owner operator/independents. It can't force shipping lots to rearchitect the entire parking lot, it can't demand world class networking and sensor systems in jeffrey Wyoming, and it can't wave away traditional safe operation practices with "autonomous"
sensors would be passive, thus no maintenance. 10,000$ every 200m is stupid. You use RFID chips or similar things.
I don't know how often signs are replaced, but at some point every sign should come with an RFID chip that encodes information about it, although you can probably just glue a chip onto a sign as well.
We scoff at Tesla's self driving claims and the valley between their "FSD" and actual required products, but they'll keep hammering at it. There's probably a trillion dollars of valuation for solving it. Musk's current managerial incompetence aside, Tesla DOES have the data collection platform to get all the data needed for self-driving, especially the "easier" stuff.
Once enough cargo is moving along a route, if there's sections of that route that are not ideal for automated movement, that will be brought to attention of the transportation agency.
The value proposition is obvious: generally, interstate highways are almost totally unutilized from about 10pm to 5am (outside of cities). 6-7 hours of transit that could be used (and offload the transit from rush hour times when passenger cars demand it more).
Even if autonomous trucks managed to handle digital logging they'll still get stopped for inspection due to weather, traffic, and load size. The rear tandem adjustment takes place on the trailer and is mandatory for safe operation so, youre building autonomous trailers too (no one is).
How difficult would it be to build a semi-autonomous system where e.g. the forklift operator (or whoever else is involved in unloading/loading/connecting the trailer) performs the manual actions on the trailer while the robo-tractor does the parts that require driving?
A robot truck could just follow a smaller autonomous vehicle.
Alternatively a drone flying ahead of a truck with cameras could augment the sensing on the truck and extend the sensor range. Deploy the drone(s) from a station on the truck. 2-3 drones per truck for redundancy and charging. Tho maybe drones can not fly at 65mph for long in bad weather.
That would essentially be a virtual road train. It's a great solution because you can reduce the distance between the trucks, which improves the fuel efficiency - especially for European-style trucks.
The problem is that they are an absolute nightmare when it comes to highway merging. You really don't want third-parties in-between elements of the convoy, so it essentially becomes a single really long vehicle for merging purposes. On most highways that's simply not a realistic option.
Indeed there are multiple such optimizations. One that occurs to me is a hub/spoke model where inter-city/state is driverless, from/to truck "yards" on the outskirts of cities. Then human drivers take over and drive the last few miles. Drivers get to go home every night and lead normal lives instead of living on the road.
Or a variant of it: as a truck driver, I can « latch onto » a truck in front of me and take a 1h nap.
Would require a virtual handshake between the two drivers, but heck why not esp. if you’re guaranteed that the truck behind you won’t try to pass you for an hour :-)
Or more or less automate the highway so that vehicles share information back and forth. Huge task, but goes a long way towards helping big heavy vehicles have information about what is ahead.
I'm wondering why there is so little talk about changing road infrastructure to make self-driving easier to pull off.
E.g. dedicated lanes for self driving or hand-off stations where human drivers can take over etc etc.
Amid the massive potential upsides, any reasonable government would invest in measures that turn out to effectivley improve self-drivability of the respective national grid.
If we’re going to be investing public money and ripping up/modifying the current infrastructure I’d actually prefer we just focus on the usual goals of urbanism and deprecate the car from the picture.
Only a small fraction of people will give up their cars, but I do hope the no-car evangelism takes off. Y'all can ride the trolley with stops every quarter mile while I just cruise on the now empty roads in my own personal car, parking right next to my jobs front door. Thanks.
This is my thinking as well. We need to build for the task at hand not try and integrate into what we already have. The first sentence in this article is where the failed assumptions are.
"Trucking was supposed to be the ideal first application of autonomous driving. Freeways contain predictable, highly structured
driving scenarios.."
When sharing the road with human drivers this statement makes no sense as all. Vehicle's are their own entities with no connection to each other outside of the road, signage and defined lanes. Instead of trying to build sensors for a existing 8-lane highway just do it for (2) isolated lanes. You don't have to plan for 100% of human scenarios if they are mostly removed. We don't fly airplanes adjacent, behind, ahead, etc of each other why is the assumption that autonomous trucks need to be on the same road as everyone else.
Well because if we're going that way, we might as well use trains, as lots of other comments unhelpfully mentioned. The strength of a car is supposed to be that they are more flexible in how to go from A to B and may even stomach less than ideal road conditions. Dedicated self driving infrastructure may negate those advantages.
I don’t t suspect it’s a technical issue at all, it’s economic.
Trucking and truck routes do have a strong incentive to switch eventually. But the examples given of Cruise and Waymo are losing tons of money.
Trucking companies would need to rework their business model around these hubs. So they won’t invest in this yet, not until it’s closer.
And for parties trying to make self driving work, it makes sense to try it on smaller, less expensive vehicles first. It’s the number of miles driven that drives learning rates, and if your cost per mile driven is lower, you get more learning per dollar.
And achieving true autonomy also needs a variety of scenarios, not just freeway. So being freeway only isn’t quite the advantage you would want. A million miles of city and highway driving helps you learn more than a million miles of highway driving. I think regulators know this intuitively too, and are more prone to approve.
DALL-E prompt produced a road with a semi on the wrong side of a no-pass divider line. Love it.
Anyway, the speed doesn't have to be 50mph. Do 40mph on late night trips. More EV range/gas efficient anyway.
The deceleration would likely be helped by an EV drivetrain where regen and braking can contribute to stopping. Likewise the high-torque EV motors can accelerate a semi far more manageably.
Emergency/Fault stops should likely be handled by convergent infrastructure, or only using routes with sufficient shoulders.
How many scene understandings can be done with on-demand manual takeover? Highways generally have some of the best cell/data networks, and again, convergent evolution of infrastructure.
Sensor distance can probably be improved with the concept of a "scout car". Likely automated trucks would drive in formation (to draft to get better efficiency), a lead car can scout and provide additional effective distance. This car doesn't even need to be a full size car, it could be a smaller teardrop-optimized drone. And if the drone crashes or faults, the following train automatically pulls to the side.
Trucking routes can benefit from neural nets tailored to repetitive navigations of the route. No handling general queries, the tractors (or the control system utilized) isn't driving "a general truck". It is driving from Minneapolis to Chicago.
I think I disagree with the article. It assumes too fast for the vehicle, too general of routes.
Oh, my final disagreement may be the total size of the vehicle. Once automated vehicles can drive at distance ... do you need a massive tractor trailer, or can you instead use more capable (and more cheaply mass produced) vans or smaller form factors? The industry is what it is in current state and a trailer ~= a container, so I guess I see why that wouldn't be a go.
"Now let’s compare these distances with the capabilities of various sensors:"
Many cars monitor rotational speed of the tires in order to determine whether a wheel is slipping. A Model 3 can detect whether a tire is new or old based on tread depth, something I learned the hard way.
If trucks were outfitted with rotational encoders on all tires, 'slipperiness' could be monitored periodically by, for example, tapping brakes and checking whether all wheels continue to rotate at the same speed. In emergency braking or maneuvering, the rotational speed of all tires can be input to the recovery algorithm to perform braking. If the brakes are independently controlled, this can be performed per wheel.
More generally, trucks run in fleets and on routes. Similar to aircraft reporting turbulence so the next craft can adjust, trucks can report road conditions so that approaching trucks can adjust their speed or pull over completely if conditions are so bad.
Makes me wonder if it would be more efficient to run trucks as "convoys", maybe consisting of competing companies along predetermined routes. And you could have an ultralight canary car go an appropriate distance along the front to ensure that in the worse case, it is the one that crashes.
Unlike a railway train, trucks operating in fleets assemble and disassemble easily, without stopping, when their routes converge or diverge. They also can operate with a number of other vehicles between them.
Rails are a more efficient interface to wheels, but also more demanding, and less flexible. 10-lane freeways are widespread, but most railways have two pairs of rails at best, and one pair in many places. You can't effectively use them without strict scheduling.
If he thinks driving cars in cities is the easiest problem, he does not have experience driving a car in a European city with lots of cyclists and pedestrians.
Driving a car in inner Copenhagen is a stressful situation due to the insane number of cyclist you have to watch out for, see https://youtu.be/FaySp9i2zMA?t=113
The point I gleamed from the article was not that city driving was an easy problem, but rather that freeway driving has unique issues when it comes to reliability and safety. Simply stopping the car could be an acceptable minimal risk condition for a crowd of cyclists and pedestrians, but that's no longer the case in a freeway. On top of having to deal with sensing range + on-board offline decision making + truck stopping distances, having no safe fallback like "just brake" seems like a pretty difficult problem.
There's also the "interesting events + training" argument in the article, which I'd love to see points for or against.
You're describing driving from a human perspective. The delta from no cyclists or pedestrians to 100 cyclists or pedestrians is a lot easier for a computer than a human because it can pay attention to all of them simultaneously and plan a trajectory accordingly. Waymo handles this kind of complexity in SF all the time.
This observation holds a lesson for cycling advocates in other countries, I think. My happy experiences of cycling in Copenhagen are only partially due to the reasonably good infrastructure. The care that drivers take (thus producing the stress S above notes) is also a big part. Good infrastructure together with careless drivers doesn't deliver a good cycling experience.
Also one of the things autonomy tech could deliver is better driving aids to ease some of that stress (it's not a slam dunk because it is easy to create driving aids that present a false sense of security).
Thanks! Pretty alert pedestrian handling in that 1st video. Including one threatening to dash in between cars, one actually doing so, and one (apparently) making to cross and then not crossing. Also traffic cones are now explicitly shown on the monitor.
Now imagine all those cyclists and pedestrians would also drive a car, would also be on the road at the same time. You think driving would be less stressfull then?
In my expeirence, it is far more common for cyclists to break traffic rules than it is for cars to do so. Even pedestrians are far more likely to cross on a red light than cars, for example.
The vast majority of humans have some respect for the fact that they are in charge of a ton or more of hulking metal when driving a car, and thus take on more responsibility. Drinking and driving is far less common than drinking and cycling too.
It depends on which traffic rules you are watching for. Car drivers are vastly more likely to break speed limit rules than bike riders are; in fact it's a minority of drivers I see who don't break speed limit rules (which leaves me wondering why they are still considered to be rules at all, in a supposedly democratic society; but that's another conversation) while it's a rare bike rider who is even capable of the athletics necessary to do so.
If a car makes a misstake - that is a 2 ton fast projectile. If a cyclist does, it is a 150 kg and way slower one - that's the difference and the post I replied to did not say anyone was breaking any rules. Just that there were so many of them. But if they all would be driving cars instead - it would just mean traffic jam.
So maybe be thankful for those people choosing to not contribute to more traffic jam. And noise. And pollution.
This is one of the biggest reasons I greatly prefer public transportation in cities. I feel stressed knowing there's a 100% chance someone on a bike is going to actively zoom out of nowhere right in front of me and pretend they're the only person around for miles. So many bicyclists and scooter riders seem actively suicidal, which is okay and I respect their right to end their life on their own terms, except the law regards car drivers as criminally negligent.
I'd rather take a train where all but the craziest bicyclists venture or let a bus driver handle the consequences of hitting someone. When I'm traveling around the countryside, I take public transit to the edge of the city then rent a car.
City driving is obviously more difficult than highway driving. I don't know how in the world he came to the conclusion highway driving is more difficult. Especially if you consider the truck form factor in a city.
I'm pretty sure "More difficult" here is a comparison of vectors in a non-Euclidean space of high dimensionality. Depending on how you measure across those dimensions you can come up with wildly different answers. The author has explained some of how they assess the dimensions.
In my (European) experience the rules cyclist break most often involve using pedestrian infrastructure because they don't feel safe on car infrastructure (or because provided bicycle infrastructure is poorly thought out). Which is understandable, but makes being a pedestrian somewhat more stressful.
They also occasionally run red lights, but less that pedestrians crossing on red.
The "rules of the road" in the US are completely slanted towards cars and against all other road users. That makes complying with them inconvenient and dangerous for pedestrians and cyclists to the advantage of cars.
For example very few countries have "right turn on red" laws like the US has. That law alone is responsible for hundreds of pedestrian deaths. Jaywalking laws were introduced to benefit cars irrespective the fact that they make life insufferable for pedestrians when legal crossings are so few and far between in the US.
How much of this is just money? Waymo has been operating for a decade and a half and has spent billions upon billions. Probably it's spent more than all the trucking startups listed put together. So it'd not be surprising if they are doing much better.
And an arbitrary one, in my opinion. We will be racking up millions of driven miles by hundreds of autonomous trucks before that bar is reached. We will have drivers in the trucks. We will have connections to remote operators.
The article calls out the challenge of stopping distance and sensing range. Drive Hwy80 over the Sierra Nevada mountains in winter, stopping, predictability, many other factors come up as well.
Yes, some routes are dangerous, but the trucks don't need to take those initially. There's a lot of scope for a phased ramp up.
A lot of I80 is really straight, same with other highways. I70 from Denver to St. Louis is incredibly boring all the way through Kansas. I90 from Wisconsin through Minnesota and South Dakota is the same. Just dead straight easy roads. Down south I10 was pretty easy too.
Just because there are a few tough stretches, that doesn't mean the auto-pilot can't drive the other thousand miles on the route, with human support where necessary.
Pick the MVP routes that are easiest. I95 is trivial from Florida, all the way up to DC. Then it remains mostly easy with some really annoying bits around DC, Philly, and NYC. I wouldn't want it trying to get through the traffic over the GW and through the Bronx, but after that it's mostly smooth all the way through Maine.
For the East / West routes between California and Nevada / Utah / etc. I80 is that corridor. I5 would be an interesting one in CA, but you have the grapevine in the south and Siskiyous in the North (again, Siskiyous not fun in weather).
I know that “just put it on a freight train” doesn’t solve every logistics problem but it does kind of address this one. Maybe their solution should be focused on improving transloading between truck and train and using autonomous trucking to solve the “last mile(s)” problem for trains.
Oh I think the major rail lines should be used for long haul cargo traffic Salt Lake -> Reno -> Sacramento -> Bay Area would be one example an keep it off mountain roads.
The US have a well-developed and really busy network of freight trains. Trains already carry a lot of stuff that's economical to carry by train, especially bulk goods.
Yes trucks slow down slower than cars but that's just physics. You've got the exact same problem with humans driving under say rainy or foggy conditions.
If anything driverless helps dramatically on this front. If you can run trucks 24/7 without stopping then you can cut their speed say 33% or whatever and still come out on top.
You can also adjust the cargo to reduce stopping distance - use autonomous truckers for things that are large but not heavy.
Or hell even just add more wheels for more stopping traction. Sure gets you more wear & tear but that's an infinitely easier problem than trying to teach an AI to recognise a kid running into the road in a busy urban steet or navigate construction site or whatever other surprises come up.
Author makes some valid points - stopping distance vs sensor range - but to me the conclusion that trucking on highways is hardest doesn't follow at all.
> If anything driverless helps dramatically on this front. If you can run trucks 24/7 without stopping then you can cut their speed say 33% or whatever and still come out on top.
I think that this line of reasoning misses the forest for the trees. If the promise of a silver bullet is a mode of transportation that can safely and reliably work 24/7 day and night with limited to no human interaction and be able to move large volumes of goods, then we already have that: it's called railway.
> The problem with rail is that most businesses don't want a whole train worth, and removing just a single car from the train is hard.
You don't need to remove cars from trains. The transportation world has been moving around shipping containers for almost a century. Shipping containers, also known as intermodal containers, were designed specifically to simplify the task of moving arbitrary units of cargo between modes of transportation, which includes to/from trains. You just need to drop them off on a flat car, take them from A to B, and then drop them off to handle the last mile.
Why don't containers already solve that problem? You can unload individual containers just fine, even put them on a truck for last-mile delivery. And then there's this American idea of putting entire trailers on flatbed cars, so you can just roll them on and off.
In the US there are lots of long 3-4 lane highway stretches where this is perfectly fine. Also we already see this on mountain ascents. Plenty of times I have seen trucks climbing a pass at 30-45mph in the right lane and cars zipping up it at 80mph in the left lane.
Isn’t dual speed limits (lower for trailers) fairly normal in Western Europe? Pretty sure that’s true in the UK for caravans/personal trailers, though I don’t recall if that lower limit applies to commercial trucks.
Germany is much more disciplined than the US about being in the right lane and only overtaking on the left. Most of the US has the same rules in theory, but people following the rules less strictly combined with the low speed limits means they don't get the clearly separated speeds for each lane you get on a German Autobahn.
I can understand where you're coming from but driving on a road where everyone has the same speed on average is less stressful than driving on a road with large differences in speed.
The latter requires a good evaluation of the speed of the vehicles behind / in your mirror rather than just checking for whether a vehicle is there. On the former, you can simply match speed with the person in front and you're good to go.
Due to this, ADAS is also broken on roads with high speed differences. It won't alert you of a vehicle is coming 100+ kmh than you in your blind spot.
But then pressure (force) on each tire is lower now, so the friction is also lower.
Effective friction is a function of weight and coefficient of friction between the materials. Interestingly the surface area of the contact is not part of the equation.
On a flat surface a truck and a car using the same tire compound should have identical stopping distances. The issue comes when you are going downhill.
That equation works for friction between idealized solids. Tires are not even close to an idealized solid, they deform as more weight is added to them, so the contact surface area increases as more pressure is added.
That is why you see racing cars with big fat tires, because it's more grip and the relationship is not linear.
The equation still holds. As the tire deforms to have more area the pressure on any particular part of the tire is lower. The total friction is unchanged.
> That is why you see racing cars with big fat tires
If you mean drag racers those tires are glued to the ground, and do not work off of friction. Other than that, what race cars have big fat tires?
Tires are usually sized so that the ground pressure of the tire is in the 40 to 100 psi range (which is also the tire inflation pressure). That's simply to keep the ground from deforming and the tire from popping. It does not affect friction.
This is also why off-road tires have lower pressure - the sand and mud they work on is not strong enough for a higher pressure. It is NOT for increased friction, it's because the surface they drive on can't handle it, so you spread out the weight, and reduce the pressure.
They have the fat tires because they're purposefully melting tire rubber into the pavement for a bit more traction. Rubber tires have maximum grip at 7% slip; just about the only material where sliding friction is greater than static friction.
Road-going tires won't get hot enough unless you're skidding far below 7% slip.
That friction equation comes originally from empirical experiments, not from idealised models.
The explanation in fact relies on deformable surfaces. If you imagine keeping the normal force constant and reducing the apparent contact area, the pressure on the surfaces will be increased which will cause the surfaces to deform more, resulting in more microscopic points of adhesion.
> On a flat surface a truck and a car using the same tire compound should have identical stopping distances.
Well according to the article they don't have identical stopping distances, so let's figure out what would fix it.
> The issue comes when you are going downhill.
That just changes the angle of gravity. Why would it affect trucks and cars differently? I thought the point was that going downhill exacerbates differences in stopping distance, not that is causes them.
For one thing an 18 wheeler has 10 brakes, not 18. So that's an obvious thing to fix.
> Why would it affect trucks and cars differently?
The weight of the truck acts as an acceleration force when going downhill. The tire friction of a car and truck is the same - but then the truck has extra force trying to accelerate the vehicle when it's downhill.
> For one thing an 18 wheeler has 10 brakes, not 18. So that's an obvious thing to fix.
That sounds like it would make a huge difference, yes.
> The weight of the truck acts as an acceleration force when going downhill. The tire friction of a car and truck is the same - but then the truck has extra force trying to accelerate the vehicle when it's downhill.
So does the weight of the car. I don't understand this logic. Every pound that's giving you acceleration is also giving you extra tire friction.
A car and a truck should theoretically lose the same number of meters per second squared of stopping force when going down a hill. If the truck started with weaker brakes then it loses a bigger percentage of stopping force, exacerbating the already-existing problem, but if you find a truck and a car with equal stopping distance on flat ground then they should have equal stopping distance on a slope, shouldn't they?
> If you can run trucks 24/7 without stopping then you can cut their speed say 33% or whatever and still come out on top.
Well, they should still be driving no slower than the rest of the traffic in the slow lane (modulo what is safe for a larger vehicle under whatever the prevailing road/weather conditions are).
I think a bigger benefit to not needing drivers that need to rest/sleep is that you can allow for self-driving software that can't handle all the weather and road conditions that humans can handle. The self-driving software can pull the truck over for a few hours in a snowstorm that a human driver would be able to handle, and still be ahead of a similar human-driver schedule.
I didn't expect fully autonomous trucking to be here yet. What I did expect, however, is semi-autonomous many-trailer trucks/convoys, because it's a much simpler technical problem that still offers enormous efficiency gains.
Let's get a little bit crazy and imagine something very different from what we have today, but still only requiring a single driver. One truck cab at the front. Individual trailers have their own wheel motors, their own steering, and a small battery for short range manoeuvring (e.g. at a loading dock/port or). Trailers follow each other at a distance that allows for emergency braking, but close enough that they can be connected by high voltage cables for receiving power from a "battery car". Both the cab and individual trailers constantly communicate, and the moment any participant suspects that all is not right, the whole train tries to slow down and pull over.
If one trailer has a blowout or other issue, it can be left behind to be picked up by another convoy.
At a port or loading dock, the convoy disassembles and individual trailers are driven by remote control, either directly by a human or eventually by the "site AI", so that large articulated vehicles never have to enter the area.
I'm imagining ~12 trailer convoys to balance efficiency gains with disruption to local traffic.
Largely copied from my reply to a similar comment:
Trains requires long-term planning and alignment of a lot more people and organisations, all of whom have their own messy incentives. They're a political quagmire that you might never emerge from.
I'm trying to think of solutions that can be mostly built by the private sector and deployed with only modest legislative changes.
For the same reasons, a friend who works in mass transit talked me around to favouring better "smart bus" infrastructure over lobbying for more rail — at least in most cases where actually getting more rail is an unrealistic outcome.
Doesn't make sense. Switching the road to steel increases cost when the road is not in use because you're building a single-use roadway. Predominant costs here are land acquisition costs. If you can't split land use among multiple users, you're hosed.
Just think about it. You're starting a company and the first thing you need to do is spend a few billion dollars before you move your first ton? Doomed company.
This is the classic problem with rail: you need to plan the route so you have sufficient participants willing to fund development, you need to acquire land (which you can't safely do piecemeal because the guy down the line will extort you).
I take your point, but that requires long-term planning and alignment of a lot more people and organisations, all of whom have their own messy incentives.
I'm trying to think of solutions that can be mostly built by the private sector and deployed with only modest legislative changes.
Your average road train which has restrictions on which roads it can drive on is 3 trailers. 4 trailer trains exist with far stricter limitations.
I just don’t see a 12 trailer 140m long road train with wires strung between the bits being feasible in any way on a road that has other, untrained users on it.
Okay, let's ditch the wires and put a battery pack in each trailer skate. If that limits range too much, we can figure out some kind of clever charging or fast battery swap solution. (None of the reasons for the failure of battery swap in passenger cars apply to trucking, AFAIK.)
The wires aren’t the main blocker to my mind - the sheer length of them on existing roads is. There is a reason we don’t do super long trucks already. They would for example block 2/3rds of an onramp merge zone as they passed.
IF we did gigantic trucks they’d be so severely limited in what roads they could use before being ‘broken down’ in to smaller units for the final delivery that we might as well just make trains.
I wonder why we don't lower the ambitions: chain five trucks together, so that for long stretches of highway (hundreds of miles) one driver can drive the "train of trucks" while the other truck drivers sleep. The trucks could be really close together so no cars can come between them.
What problems am I overlooking that are not easily solvable?
Interesting, but does every unit have a driver? I was thinking, it could be interesting that they all have their drivers, because finally they won't all go to the same destination.
So it would be great that as a truck driver, you could attach yourself to one of these trains, sleep for a couple of hundred miles, put the alarm clock, wake up, then detach at the right time and go to your final individual destination. Maybe even take another "road train" for another couple of hundred miles...
There aren't any logistic piplines that make this favorable. If you actually wanted to do this you'd load it on intermodal freight cars and push it down rail to the destination.
People seem to think trucks go from one "transfer hub" to another. That hardly ever happens, and when it does, the trucker usually drops off one trailer and picks up a different one.
Most truck routes are actual deliveries or pickups. You go from the distribution center for a supplier and then you do a chain of stops at 4 or 5 stores where parts of the load are taken off the truck and the rest goes to the next stop. Or you pick up an empty and go to 2 or 3 locations to pick up finished product to either be delivered or brought to distribution.
I mean.. there's _thousands_ of hours of truckers on Youtube just doing their job and discussing it. You could watch any of those channels for a few hours and be _lightyears_ ahead of the typical "AV programmers" thinking on the subject.
It's kind of depressing really, how so many silicon valley type people never take the time to learn how the industry works.
Here's a place where automation and even EV would be _highly_ useful: "yard dogs."
Like I said, there's lots of "drop and hooks" in the industry, where your job is to leave a trailer behind somewhere and pick a new one up at that same place. That hub then needs a truck to go move the trailer from the spot, to the warehouse, then back again.
Trailers are numbered in a serialized fashion. Parking lots could be mapped with insane precision. Loading docks would need minimal upgrades. You could have a mostly automated warehouse freight yard logistics system. You generally have lots of power and only so many moves in a day so EV would be a big winner in any non-wintering lot, for a good reduction in diesel usage.
Okay.. now you can start thinking about tight integration between the trucks electronic communication system sending gps position updates to the warehouse enabling just in time trailer movements on the lot so a trucker might be able to breeze into and out of the lot in 5 minutes with everything waiting and queued up at the front gate for him. Now he doesn't need to navigate a strange lot at 5mph trying to find the right place to drop and the right trailer to grab, a good source of errors.
Now you can get into moving the EV automated yard dog off the lot and into the hands of larger customers like big box construction stores. A truck could be dropped off at a convenient place in the lot and a small mover could go grab it and maneuver it into the local dock. You could tighten up the spacing on those docks and put them in places a full tractor couldn't possibly pull them into. This also prevents the safety issue where the trucker can move the trailer before the business has finished with it, a good source of injuries.
Anyways.. silicon valley is focused completely on the wrong side and scale of the problems. There's so many "little things" where EV and automation could do wonders. Instead, everybody with a touching back story and a high valuation imagines themselves being able to become the next monopolistic transport baron at the cost of everyone else.
I’m curious about whether the “autopilot+” model could be a useful stepping stone. If you always have a human driver on board, but they can take breaks while the autopilot drives, maybe this covers a lot of the problematic scenarios? A human driver could look at traffic reports and plan to take the wheel on areas with congestion, road works, and the hilly sections that are technical. The boring straight shots they can sleep. Maybe this could get the duty cycle up to 100% and keep the truck moving non-stop? Really any material increase in the duty cycle would be incredibly valuable.
This approach would help data collection (since you have a ground truth from the drivers) and would also mean you have a mechanic on hand to fix the out-of-cab issues that inevitably come up.
> A human driver could look at traffic reports and plan to take the wheel on areas with congestion, road works, and the hilly sections that are technical.
Yes, but honestly, you would have the computers look at the traffic report and at the map.
> The boring straight shots they can sleep. Maybe this could get the duty cycle up to 100% and keep the truck moving non-stop? Really any material increase in the duty cycle would be incredibly valuable.
I think technically that's probably possible on many routes, but I wonder how regulations interfere.
They claim that stoping distance and non-normal driving conditions are an issue but these are precisely the same problems that AVs meant for ride share face but with a difference performance profile. The computer doesn’t care if it’s hauling 20 tons or 2 tons, what matters is we give it the correct performance profile and install adequate equipment to operate the vehicle safely.
The only real argument I see here is the limp mode case where it detects an error and must pull off. I can see the actual freeway being a problem here but then we just make sure the routes they initially travel have been checked prior to deployment - it’s a finite road that can be mapped and tested against. This can be further mitigated by having redundant computer systems and redundant environment sensors than it can fall back on to be able to safely pull off the side of the road or even make it to its next stop. Computers can make faster, more-informed decisions about nearly every driving situation and the challenges with operating heavy machinery are well understood.
The reason autonomous trucking has not been developed is not because of R&D, it all has to do with scale and what the funding is chasing. Running a truck is expensive, especially with a large load. Financing a project would have a lot of money dumped into the just the construction or retrofitting of a truck and it’s operation for roughly the same computer as the smaller vehicle - just with a different performance profile. When the vehicle you are testing has a kerb weight of 1100lbs and is electric, your maintenance costs plummet to almost nothing with fuel being electricity. We don’t have the same luxury with trucking. Additionally, Uber, Lyft, et al, are bullish on this technology because they would rather collect fares and pay a fixed cost for vehicles than having to deal with drivers that want silly things like compensation. The gig companies are also looking down a barrel of a loaded, regularly-shotgun that could drastically cut into their revenue. The dollars they are investing into AVs only care about ride share now.
> While MRC behaviors are annoying for other road users and embarrassing for the AV developer, they do not add undue risk on surface streets given the low speeds and already chaotic nature of city driving.
Except in the case of Cruise.
MRC could be handled somewhat successfully for trucks on the highway on sections of highway with good cellular data connection. You could have someone making minimal money play a driving game on the computer all day to keep them paying attention. If an autonomous truck encounters an unusual situation, their screen switches over to a camera feed for the truck, and they can take over the controls of the truck to get it past the obstacle.
Not sure what fraction of highway has sufficient connection to give them a sufficient live feed. Probably a lot though.
> This is a really, really high bar. For example, on surface streets, this means the system on its own is capable of driving at least 100k miles without property damage and 40M miles without fatality. The system can still have flaws, but virtually all of those problems must result in a lack of progress, rather than collision or injury
This is a weird way to think of it, I don't actually think of it as statistics. The failure cases will be different (between human and AI). Further, we want to be able to sue those responsible (not some nameless company).
Final thought, I wonder if this is the metric, when we'll start seeing insurance remove humans or increase the premiums. Basically, exponentially driving everyone into AI driving
- Sensing distance is a good one for sure. At the same time, it's for now a technical issue. So that a viable temporary progress-making solution might be to have an additional vehicle in the autonomous system: a sensing car is in front, and one or more semi-trucks follow. Then the autonomous system far out-senses a human driver. And we get progress even if the cost-proposition becomes a little harder.
But otherwise:
- Nominal stopping distance as shown include 2.5 seconds of reaction time - fair for a human driver perhaps but hopefully the machine gains a lot of space there.
- "Stopping in lane" is as insane in the city as it is on a freeway. We can't seriously consider that a long term solution in the city? Less risk temporarily okay but not a serious solution. (But fair point on this one: if all else fails it is, for now, an option in the city).
- Even for humans, driving on the freeway is not a question of perfect judgement. It is about quickly but calmly picking a reasonable solution. Including, in doubt, moving to a safer lane and slowing down while measuring the risk of rear-ending. Nobody asks a human to completely eliminate the risk of rear-ending. That sounds within the range of what software can do?
- Jack-knifing and other complex behaviors are probably far better suited to computing solutions - which can sense far more measurements than the humans, independently actuate more control points, and compute actually more or less complete solutions when the human is seat-of-the-pants -ing in real time (with dubious real life record too - see brutal and fatal). Simple no, but all that hard to beat the human sounds absurd (anyone with more input on this?) See for example the drift-parallel-parking stunts achieved with cars. We may not yet be at drift-parallel-parking semi-trucks but anyone here doubts that this is achievable?
- Ease and quantity of samples might be a better point. Waymo really got on it and accumulated a lot of driving data. But then also, most of that data by now is city driving.
- Will trucks happen before cars? Well, cars are already there so... already answered.
With new markets you often want to target a small niche, I've always thought that relatively slow moving road trains travelling at night would be a good place to start.
Air resistance means slower speeds are more efficient (you don't need to worry about human driver needing paid per hour or limited by regulations).
Less other traffic at night, but (presumably) sensors can be designed to work well in these situations.
If you split a truck in 3 it can travel together as a pack for aero yet reduce stopping times as each part brakes seperately.
You could probably even section off an entire lane in certain areas.
Some of these possibly needed EVs to mature more before they become practical.
Trucks pay for more than air resistance though. Last I talked to a trucker their computer was set at 68mph because the company had data showing that was the most efficient speed. Engines get more fuel efficient the more load they are under. There is also the constant load of the lights. With autonomous trucks the computer/sensors is source of constant load.
A train that will take a smaller number of cars ONLY between point A and B. Then at each end offload to a truck.
Not the same as current rail frieght, I am saying a dedicated track between say port of oakland and port of long beach - with some intermediary way points along the way - it just has two lines - there, and back with a dedicated automated train that simply goes back and forth - with gantry ports the train drives beneath at certain locations to lift off cars quickly for that node.
We have done a poor job of leveraging rail in a future-proof manner...
I just think that rail, freight and passenger, is the place to start. Automate that first, get 10 years of that under our belt, then figure out how to add steering and traffic.
Autonomous driving vehicles advocates are as ignorant of reality as are those thinking ai will replace competent programmers.
The trick in driving is not straight lines and clear weather. It’s rainy, snowy, foggy, conditions that make it challenging. Edges cases are the real difficulty, as with everything.
I suppose when your knowledge of the world is limited by a few multiplayer notepad interactions everything is simple.
tldr; Why not slow down when sensor range is shorter than stopping distance?
It seems the basic argument is that sensor range is not always long enough to fully encompass stopping distance, but the data presented in the article shows that this is only the case at 70mph.
At 50mph, it seems that sensor tech is well within requirements in all published conditions.
Perhaps an autonomous truck could simply slow down when grade or surface conditions reduce its stopping distance?
While it is nice to do 70mph all the time, slowing down when you can't see or stop well is exactly what is expected of human drivers.
In many places in the US, the road system is already insufficient. Certainly completely insufficient in California. There is no space for self-truck traffic to move significantly slower than driven-truck traffic. Either self-trucks can safely reach speed parity or there is much less space for them.
Another response mentions there are easy routes that could be self-trucked first. That would work.
Does driving at 50mph decrease throughput compared to 65mph? By how much? Can we overcome that with a somewhat reduced following distance between self-driving trucks, since they can coordinate braking? (or at the very least they can respond to brake lights in front of them near-instantly)
When I think of "completely insufficient California roads" I think of traffic that's already going a lot slower than 50mph.
But no, I was thinking of places like interstate 5 (the main north south freeway) which is often a struggle of trucks passing trucks and cars passing trucks and cars passing cars in a completely absurd 2 lanes of traffic in each direction. Nominally 70 / 55 speed limit but nobody drives at these speeds.
And this is one place where a few trucks driving slower than the other trucks would certainly cause extra mayhem.
The article assumes that the reason autonomous trucks haven't happened yet is that it's harder. I think the real reason is probably simply that #1 it's less glamorous for investors and #2 most of the companies producing trucks are probably not great at doing anything with software. So, the investment levels and effort that has gone into autonomous trucks so far just isn't at the same level and the companies that have gotten the furthest with autonomous driving just haven't focused much on trucks so far. Not because it's harder but because cars are hard enough, more lucrative, and they want to focus on those first. That could change. Once they nail cars, trucks are an obvious next target.
There are a few other possible reasons; such as electrical trucks still lagging a bit behind electrical cars in terms of volume production and a few other things. Electrical trucks are a lot easier to drive for humans because they don't have a lot of complicated gears, hydraulics, and other things to worry about. That makes them easier to control autonomously as well.
Other than being bigger, the job of autonomously controlling a big truck isn't that different from that of controlling a car. If you can do one, you should be able to do the other. Ballpark it's the same kind of problem. Same roads, obstacles, problems, traffic situations, challenges, etc. Same everything, basically. I don't think the speed or the size of the vehicles is that much of a factor for how hard this is to do.
As for safety, the barrier is actually pretty low. There os a high number of accidents that involve truck drivers asleep or unfocused behind the wheel because they've been driving for way too long. Or getting distracted by their phones. Happens more in the US than in Europe because some differences in rules related to mandatory breaks. But it happens on both sides and it's a bit of a challenge.
We're measuring with very different standards for humans and autonomous vehicles. With autonomous driving it's all about hypothetical things that may or may not go wrong and therefore we should wait until it's so perfect that it can never happen, 100% guaranteed. With human drivers, it's just an endless stream of never ending accidents, fatalities, and misery where we just go "ah well, that's just life". Never mind all the obvious issues with aging truck drivers that are severely out of shape due to life style issues associated with trucking after they've been on the road for ten hours. This attitude is more than a little bit irrational.
Australia has there system of 'road trains' https://www.youtube.com/watch?v=PXL9UfP6FMk. They use these because they have very long sections of highway that are suitable. The USA has a smaller length version used on the controlled access interstates, with the same hookup as tractor trailers, with the trains broken down to 1 or 2 trailers for local delivery(usually the third one is shorter and is called a 'pup' trailer. Very common in gasoline/oil delivery.
On larger flat areas, with automobiles forbidden, it would be quite doable to have longer trains, with the hauling force/strength suited to the grade. We already do this with piggybacked trailers on flatbedded rail cars, some of which can have side load/unload of trailers for granular load access at stretches to avoid having to unload them all to get at one. Railyards do this via switching 'hump-yards' where long trains are distributed car by car. Here is one in operation, usually the hump has enough height to allow gravity sorting and there are numerous remotely operated switches where a long load of 100-200 cars is sent over the hump one at a time and assembled into trains of cars on diverse lines for other geographic destinations, with some send directly for local delivery, both as unit containers to be picked of by special top/bottom/side lift fork lifts(locked load type).
https://www.youtube.com/watch?v=e0CUtE5-kKQ
I can see that this would allow integration with road trains, as methods evolve.
I saw several posts that spoke about the ground level interaction of bicycle traffic and car traffic being dangerous to bicycle traffic. Years ago we had elevated railways, that came to be unsightly, but functional ways to mix subways(elways) - it made me think that elevated bike-ways might work to reduce bike/car interactions. Bikes are so much lighter in weight that a suspended bike 'mono-road' would work well, it could even be roofed against weather. It has been done in a few places, and seems to work.
https://www.google.com/search?q=bicycle+elevated+path&sca_es...
We need smart roads. Not just smart cars. Until that happens, autonomous is impractical. You don’t want to build an autonomous system to match human error rate. That basically got us nowhere in the past seven years since Elon bs’d his way to selling vaporware.
Aaaaand we’re back to reinventing trains. There should be a name for this phenomenon. How long do tech folks discuss how to improve transportation until they attempt to reinvent trains?
Stop being moronic. Different transportation mediums exist for a reason. Trains in the US are currently suited for long haul transportation at which they excel at. We already tried running rail lines to warehouses and distribution centers and determined trucks were a better form of last mile transportation.
Do you think that no one at UPS, FedEx, Walmart or Amazon has ever looked at how much more efficient or what cost saving they could get if "only they had a rail line to their warehouses verse a road?"
What the discussion is looking at is making a micro improvement to what we have already tested as the better method of delivery, trucks in this case. Where the issue is coming about is we are trying to build autonomous actions into a hybrid roadway verse dedicating a portion of a roadway to autonomous driving. The article states the following:
"Trucking was supposed to be the ideal first application of autonomous driving. Freeways contain predictable, highly structured driving
scenarios"
When sharing the road with human drivers this statement makes no sense as all. Vehicle's are their own entities with no connection to each other outside of the road, signage and defined lanes.
The term you’re looking for is armchair logistics.
Trains and trucks have been competing for a long time. Autonomous trucks, even in convoys, even where freeways have RFID signs, are vastly different from trains. You don’t need to lay and maintain tracks. You’re not limited to where tracks go. You can assemble and disassemble on demand.
Yet you need to lay and maintain roads, you're limited to where roads and RFID signs go. You also have the added inefficiency and environmental problems of rubber-to-asphalt contact. In addition, you need to carry batteries which means even more inefficient transportation.
Correct this is the best path forward. Night time is typically when there is less traffic and we should be able to at least utilize that for autonomous transportation between distribution centers. Then slowing keep upping the advancement as roads are repaired, replaced, etc. If you build a smart road for lets say just trucking, now you can map out the best places for distribution centers, similar to what UPS, Fedex figured out for air transportation and packages.
Its a bit ridiculous that in 2024 we have all these apps that tell you about accidents and obstructions but none that actually help re-direct traffic in a meaningful way. You have to hedge your bets on any "suggested" better route.
With smart roads department of transportation could get real analyze on crowded intersections, best place for future on/off ramps, access roads, etc. Traffic congestion and planning is usually an after thought when housing developments go up and then its a catch-up game.
One could argue traffic light was the first “smart” feature added to roads. Why do we continue to have costly high way accidents in 2024? Massive pile ups can be minimized and even avoided.
Unfortunately we didn't expand upon thinking about multiple intersections and their traffic lights being a network. We have fixed programming of lighting schedules that don't change unless a traffic survey is done. In most cases these are limited in their occurrences which doesn't reflect the dynamics of driving.
As an example I'm sure most people don't mind moving as long as they feel like they are getting somewhere. With the dynamics of traffic you could re-direct people to a different on/off ramp even though its further away to keep the same rate of speed for everyone. Maybe you send people down a different or parallel road for longer because it will get them there at the same time verses stop-and-go traffic because everyone wants to use the same or well-known road.
I think rail like truck fleet where a single driver controls a fleet of 10 something autonomous trucks for local time low latency backup control and Starlink for remote backup control could be something that could be useful.
Is this not the most obvious statement ever? At the most basic level, trucking requires loading and unloading freight, whereas rideshare passengers load and unload themselves.
> The trucks could also be commercially viable with only freeway driving capability, or freeways plus a short segment of surface streets needed to reach a transfer hub.
The question is... isn't the problem being approached from an entirely bad premise in the first place? Why would it even make sense to develop autonomous trucks for that part of any freight's journey? Laying train tracks might incur a large initial expense since the tracks, unlike highways, aren't there anyway, but a train is way more energy efficient as only the first locomotive has to deal with air resistance and steel-on-steel is less friction than tires.
To a pretty good approximation, everything that feasibly can get shipped by rail already is. Railroad companies operate fairly close to capacity, and they have no desire to reduce their massive profit margins building new rails en masse and handling the maintenance burden. That leaves only the shippers themselves, who are often much smaller companies that couldn't reasonably build a new rail line or buy cars for every customer, but can easily afford a few trucks to be relocated anywhere in the country they have demand.
> Railroad companies operate fairly close to capacity, and they have no desire to reduce their massive profit margins building new rails en masse and handling the maintenance burden.
I'm talking from a "looking at the wide, societal picture" viewpoint. And from that, it makes so much more sense to shift as much long haul traffic to railways as possible. I'm aware that the mechanics of modern day capitalism are not aligned with what's best for society at large, long-term - mostly that railway infrastructure is orders of magnitude better regarding emissions and energy consumptions, but that it takes many more years for investments to pay off, which is why there is so few railway capacity in the US and what is there, is often enough in shabby condition.
Great post with good details. I wonder whether a "truck only" roadway might be a reasonable simplifier here (yes building a 1000 miles of road for trucks only is billions of $ but there are perhaps segments between say ports and warehouses where that might make sense. And as someone who has been fooling around with robotics since the 80's the whole "sensors have to see far enough ahead to respond" is well known. The "workaround" is to go slower, and granted 40 mph trucks on the freeway would be annoying it gives you a way to test responses while sensors catch up.
Freight trains have a much lower fraction of human labor as cost vs. benefits relative to trucking or cab driving. 1 truck and driver carries 20 tonnes. 2 engineers on a freight train carry close to 20,000 tonnes, so there is only 1/50th (ish) the benefit.
No. One form of transportation doesn't have to exist for everything. Rail is the long haul and trucking is the last mile.
Autonomous trucking should start out with its own dedicated part of the highway between distribution centers. We are trying to build autonomous driving to accommodate our current road structure which humans currently drive on. A dedicated / isolated lane with sensors, etc should be the start. Start small and then you can expand after getting feedback data. Everything is always so costly and unattainable because everyone wants to implement it with the current infrastructure but its not needed.
For reference we already do these types of dedicated lanes with ezpass express or HOV lanes.
Caught that did you? :-) The one interesting disruptive thing might be paved roads vs rail infrastructure. It is easier to have a steerable vehicle turn on different paths than doing railroad switches. So the underlying physical plant of building things might allow for rapid deployment vs rails.
we do have semi-autonomous trains, in the US at least. i'm not sure about elsewhere. PTC stops a large class of disastrous operator errors (overspeed, missed/ignored signal, etc.). it is almost entirely complete on all of the class 1 railroads in the US by now.
I still don't believe that analysis that Av city driving is easier than highway. AFAIK the only company that guarantees "self driving" on unknown roads is Mercedes and that is for highways only. I think waymo/cruise are somewhat of a red herring because they only drive on known roads, which I don't really consider AV (otherwise we could call self driving trains AV as well).
That said the conclusion from the article is really that self driving is hard and we likely will not see it for many years.
But an autonomous truck can just go slower, no? It never gets tired, so unless there is a logistic pressure, it would be reasonable to trade speed for safety.
No one will tolerate self-driving trucks that are limited to lower speed than human-driven trucks. Since human drivers have time limits by law, they need to be able to move relatively quickly. Human drivers attempting to move around super slow autonomous trucks is a guarantee for a lot more accidents, more than erasing whatever safety benefits self driving would get in the first place. Also, slow trucks would not be preferred by the logistics managers that arrange truck delivery in the first place. So self driving trucks would have to be a lot cheaper to be competitive. I don't think labor is 90% of the cost of a truck delivery.
50mph or whatever speed you'd still see on the freeways is not that slow.
> Also, slow trucks would not be preferred by the logistics managers that arrange truck delivery in the first place. So self driving trucks would have to be a lot cheaper to be competitive.
Why would they "not be preferred" to such an extent? For short trips it might be an hour difference, is that such a big deal? As you approach a full shift of driving, a human driver spends up to 20% of that shift not driving, so the difference isn't as much as you'd expect. And as soon as you go past 11 hours at human driving speed, the self-driving truck gets there much sooner.
> I don't think labor is 90% of the cost of a truck delivery.
I've begun thinking that human wages are a good indication of how hard a job will be to automate.
It turns out we already have general intelligence (producible in about 16 years) available on demand. And so, the market is already a good indication of which jobs are easy and hard to automate for general intelligence.
Trains are faster and easier than all of the above, and they have about a century more track record. This case was closed before it was even opened. Autonomous trucking has been an idiotic pursuit since long before it was ever pursued.
Or you know, instead of trying to eke out tiny efficiencies from trucking by eliminating the human driver who can solve all these local issues themselves, we could invest in a proven, highly efficient, low staff requirement, long distance freight system...
Stop the truck using a couple of vehicles, cut open the lock on the rear doors, and loot. Pick an isolated stretch of toad, and the pirates will be gone before anyone can respond.
If there's no driver, there's no need to be armed, so instead of an armed robbery charge, the risk is just a theft charge. In a world where people defend shoplifting.
but the driver is a witness and a potential chaos agent so must be at least monitored and possibly killed. Robbing a driverless truck would convey less risk or potential death to the robber and less risk of catching a murder charge to go with your grand theft.
There's a pretty big jump in perceived criminality between violent crime and property crime; humans are social creatures, after all. Far more people are willing to steal a car that's parked on the street than to hijack one, far more people are willing to commit burglary than home invasion, etc.
Even for the people willing to use violence, having a human involved opens up unpredictablity and risk. You can't know what's going on in the head of a random trucker, whether they're armed, whether they fancy themselves a cowboy, whatever. It's a high-risk, moderate reward scenario.
In contrast, if you have a system for consistently hitting autonomous trucks, it's a (relatively) low risk, moderate reward crime.
Just spitballing but it seems a unique aspect of a human driver is their fear for loss of life. We're all subtly terrified we are about to die and that keeps us sharp and responsive and thoughtful while driving.
Is this satire? That’s not how humans work at all. Even a very rational player will make a relative trade off call of the probability of an action having an outcome and the size of the outcome. Case in point: driving with almost no sleep, driving while drinking, etc. and humans are quite often irrational and take risks just because they’re seeking the adrenaline rush. And that’s for their own life. We can easily also discount the life of strangers behind an abstract “I don’t know them” mask.
People deciding to drive while sleepy doesn't mean it isn't scary to do so (in fact it's horrifying). People definitely make illogical decisions. Being afraid while driving isn't even necessarily a logical decision. It's just an emotion we feel. But I know I feel it and a lot of others do too.
But you get that not everyone has this fear and therefore extrapolating that this makes humans innately better drivers, right? Also fear and anxiety can make you more alert on what you’re focusing on, but it narrows your focus and can destroys your situational awareness that your peripheral senses give you. There’s no clear link that fearful drivers are better drivers.
> People deciding to drive while sleepy doesn't mean it isn't scary to do so (in fact it's horrifying)
You’re generalizing your personal emotional response and opinion and assuming everyone or a substantial majority feels that way. That doesn’t seem like a safe assumption given that fear driving tends to not be so common because that tends to be more common in people who learn as adults.
(I used to work for an autonomous truck company, and when you factor in the cost of roads in addition to the development of the trucks, it makes absolutely no sense to do autonomous trucking when you could do trains. As a culture, we've been brainwashed not to fund trains. We collectively spend billions and billions on roads but would not dare spend money to build more tracks. It is shocking and ludicrous, but that's what happens when you suck up a century of propaganda from the fossil fuel and automotive industry).