Seems like yet another press release from an autonomous driving company that promises to test without humans soon and has vague ambitions of level 4 autonomy in the 3-5 year horizon.
"This year"... kinda... This year there will be the first real tests run with no one in the cab. "By 2024,TuSimple plans to achieve Level 4 autonomy, meaning that its trucks will be able to operate without a human driver under limited conditions that may include time of day, weather, or premapped routes."
Somehow I've managed to miss that trucks are better for this stuff not just because of the type of driving they do, but also because they're trucks! Being bigger gives them some advantages.
"Trucks are also a better platform for autonomy, with their large size providing more power for computers and an improved field of view for sensors, which can be mounted higher off the ground."
I'm the author of this. It was certainly surprising to learn that a lot of what AVs are doing right now is limited by the amount of power available to the computers. It didn't make it into the article in full, but we were told that in general more computing power means detecting and classifying more object types, tracking those objects over longer periods of time, and doing more detailed motion prediction and planning across longer time horizons. They were explicit that the power they can squeeze out of the vehicle's alternator is currently a significant limiting factor to the performance of the system.
I immediately thought about the car stereo guys they run stronger alternators and power massive amount of subs and speakers. But I know nothing about big trucks and their electrical needs. They have refrigeration and other electrical needs I am unsure of.
You made me curious. Big refrigerators and shipping containers have their own diesel engine: https://en.m.wikipedia.org/wiki/Refrigerator_truck
They are independent from big truck’s power system. That is a solution for powering electronics - external diesel generator.
10 kW is almost 10% of the max output of the engine in a honda civic. for another comparison, that would be like running three typical car AC systems at max simultaneously. so not infeasible, but a meaningful hit to fuel economy in a smaller vehicle. it would be even more detrimental to the range of a small EV.
If so, why not use cars without alternators? 48V mild-hybrids have much higher capacity starter-generators instead of alternators, and electric cars can generally pull power from their batteries at the same rate they can charge it. For example, for a Tesla Model 3 that's 250 kW.
Hybrids are absolutely the most effective way to go here in the near term, there is just no question in my mind. I'm honestly very, very surprised we haven't seen that take off yet-- either it's a lot harder than it looks, or owners of large fleets are irrationally gunshy. I've been expecting it to show up and start kicking ass any minute now for at least the last 7 years.
Seems like electric vehicles are not limited by this problem of power availability. One just has to be concerned by how much that power consumption for computer operation affects vehicle range.
People also tend to give trucks a wider berth, or at least notice you more and respond more readily to your actions.
I’ve only ever owned small cars, and in the handful of times I’ve needed to drive a U-haul truck or similar the trip started with trepidation. But I quickly discovered that other drivers have a strong tendency (self-interested, of course) to help big trucks out.
Trucks tend to have bigger alternators, so there is more power available. Even the most basic economy car has more than enough engine power, but the alternator can't deliver that power. The engine power instead mechanically goes through the transmission to the wheels. (in a hybrid car there are two alternators, one for the wheels and when for the rest of the electronics so the same problem exists)
the is solvable. you can buy bigger alternators, and in some cases there is enough room that they will even fit... However it is an additional upgrade that needs to be considered.
I don't disagree with what you're saying, but my point was that I don't know if power consumption is even relevant, considering what Comma are doing from a phone on a windshield. I don't know (I'm not an expert) but I don't think putting TPUs in trucks is a requirement for solving driving.
Do you actually have data to back that up? I read your article, which (no undue disrespect intended) looked to be full of unsubstantiated qualitative data/graphs and opinion. Are you an "expert"? Or are you trying to get medium clicks on your article/ purchases of your book?
That's fair, although credentials != proof. Working for Waymo, Lyft, Uber, Nuro etc is a credential. The engineers you mentioned have verified that what you wrote fits with the method/ approach they think will work. There is a possibility that they're/ you're wrong.
Yeah fair, I wanted to see if you were full of shit, but you probably know more than me, and bickering and points scoring over stupid things is how hacker news turns into reddit, sorry.
I cannot believe that this isn't the foundation of autonomous driving: long hauls, highways, no stop signs, semi-controlled circumstances. The economics are more clear and industrialized.
Even if a driver had to be present when they got off the highway, I think 'long haul' should be naturally the first thing to be tackled.
And I'm surprised Walmart isn't leading the way, they should be.
I've always wondered about what happens when the system really gets off the 'happy path'. Maybe there's an accident, and police are directing traffic? Does the truck understand hand gestures (which aren't necessarily standardized)? Or maybe the police aren't there yet, damaged cars are blocking the road, and the truck may need to get out of the way for emergency services to arrive (possibly requiring going a little off road). These happen less on highways, but they still certainly happen, and not having a human in the loop can make things dramatically worse.
Do they have to stop at weigh stations? Do they understand the signage? I suppose most would have the system where you don't have to stop, but I don't know if all weigh stations support that.
There are lots of instances you can imagine that don't require driving skills, but actually require a general intelligence with knowledge of the world around them.
(I suppose you can have a command center where someone can take control or give commands to the truck. Sounds tricky but possible).
They'll just have remote control and call HQ on every unexpected operation, where an operator will make a decision using cameras. On routes without cell coverage they'll use satellite internet or starlink, etc. It's not reliable enough to make operator drive it all the time, but it would already today be reliable enough to keep safety in extreme situations and make the operator make decisions.
It already is; autonomous vehicles active now are e.g. airport light rails. Short hauls, fixed track and schedules, very controlled environments so not even a need for e.g. lidar or smart cameras or anything. Distance sensors to kick in in case of unexpected obstacles maybe.
There is one minor detail to add: the truck weight and speed makes it much more dangerous than a simple passenger car. The first couple of lawsuits when these statistical engines do some real damage will be brutal.
Walmart doesn't do much long haul. They do the warehouse to store, but that is a lot of local roads so the logistics of getting drivers to/from the truck on highway access is harder than just keeping a driver in the truck.
The big truck companies are interested in this. They mostly highway travel so a team of trucker drivers in each city isn't as logistically difficult. Plus they can promise their drivers home each night instead of twice a month.
Walmart is interested in moving a lot of stuff at scale. Apparently they wanted to help build a massive superhighway from mexico to somewhere in the midwest.
It's a great idea.
Walmart should be taking much more initiative in all of this as one of the only entities that can feasibly take on Amazon.
I wish the President would make a national strategy over this 'Digital Pathways Act 2030' - i.e. make all major highways, railroads, and other 'rights of way' completely autonomous and electric by 2030.
Even for human passenger cargo - long-haul buses may actually compete with trains on price and comfort if they were fully autonomous - especially for shorter routes.
One of the problems with 'trains' in America is that the US is a little spread out, trains can't just go downtown to downtown - but autonomous buses could.
So this could be like quasi-public transport initiative from at very least New England down to DC - that entire corridor.
What we need is someone to 're-brand' bus travel and make it social acceptable for middle/upper class folks to ride instead of the agrarian/lower-class greyhound. Nothing against greyhound but it's just not the experience many people want.
Over the last five years prominent self-driving companies have consistently over promised and under-delivered. I was quite optimistic for the approach that Starsky chose (but unfortunately couldn't bring to market [1]) with autonomous segments combined with occasional remote assistance. It might not be as sexy as FSD but that's tech we can use now.
That said, if this team can make it work, power to them!
Edit: I realize this might sound dismissive. I worked on perception tasks for autonomous driving, and believe FSD can be done. I just don't feel that it's a year away, but I'd love to be wrong!
What are some examples of risk not taken that have lead to falling living standards?
My understanding of the data is that living standards have fallen because wages for most workers have been fairly stagnate since the 70s while the price things like housing and healthcare has skyrocketed.
Nuclear power was set to decrease the cost of power by an order of magnitude. It was stopped due to perceived risks while coal was allowed to continue to be used at known high death toll of millions of people per year.
You cannot legally build a nuclear reactor in a normal way. You have to design it completely and then get it approved before you build it. This costs around $500 million dollars, as NuPower has shown. If you change the design much at all you have to get it approved all over again. Thus only one new reactor design in the US has been approved for 50 years and it still has not been built yet.
If I block a road an autonomous truck is expecting to drive along, will it automatically reroute to another road? If it does, then these trucks are going to be very easy to rob simply by corralling them in to a quiet location..
Is truck robbery at gun point a huge problem currently? Already robbing trucks is pretty simple afaire, but there doesn't seem to be huge boom going on...
But not really. The person never makes a fuss (nor can realistically do so), and netiher does want to make a fuss for a simple job for one company, driving cargo owned by another company, with overall operation insured by a third company that's just going to reimburse everyone.
But it is, they would need to use violence to compel them. That greatly increases the risk of them getting shot (Many truck drivers carry guns) and it goes from larceny to robbery which greatly increases the criminal penalty.
That would probably happen if you had armored autonomous Brinks CIT trucks driving around. But just as likely they’d have thought of this scenario and prepared for it.
They'll just insure it and be done with, like they currently do with regular trucks (which BTW have to stop every 9 hours and let the driver sleep, thus creating a very vulnerable situation wherre the truck is parked and driver is helpless). Truck robberies are just not a big problem already.
I think truck stops and middle of nowhere are different though. Truck stops have other truckers. Some have facilities and cameras.
But yeah, I think most truck jackings are targeted and often involve insiders, or occasionally a driver leaves the truck unsecured and someone drives off but that’s rare.
Granted, I am no truck driver myself, but from my vast experience hitchhiking in the EU and former USSR and hearing stories from drivers, robberies or theft of cargo happen quite often at truck stops along the roads, most of which lack cameras or even much lighting. Other trucks may be present, but their drivers are usually sleeping and wouldn't intervene anyway, because who wants to risk their life for a stranger when all cargo is insured anyway?
One with a large array of cameras. So the dumbest possible target.
But people rob banks, where they're also 100% guaranteed to be on camera and automatically involves an FBI investigation. As well as having a surprisingly low average haul (a few thousand dollars).
If I'm actively robbing a truck, I will put on sun glasses. What I'm saying is that it's harder to know a masked man is robbing or just inspecting a truck.
Pray tell what are you going to gain by robbing a truck? A bunch of plastic crap you're now going to have to fence? At least in the 80s it would be TVs or cameras you could liquidate on the black market.
A few months ago I listed an old LCD TV on Craigslist. A bunch of offers fussed about coming to get it, and I ended up picking the most polite one and delivered it. I could barely even give it away!
I had the realization the other day that what I want is a self-driving truck bed.
Just a truck bed. Maybe one with a cab on it to stop the weather, but 8-10ft would be enough. I could send it to get my groceries, have it follow me places if I need to take something with me, and send it to my family if they need to give me something.
Rental truck beds would be even better. I could order one, have it go pick up my curbside order from Home Depot and the Grocery Store, and then drop it off at my house. Then I could send it back to the rental facility. Likewise, companies could buy them to deliver for the last mile.
That is not the policy of drivers who wish to retain their CDL. Gradually disintegrating radials are a hazard to everyone on the road. It isn't as though blowouts are uncorrelated, anyway.
That’s the joke. The obvious choice for the AI is to slow down without applying full breaking power, find the nearest safe spot to pull over, and then call for assistance. But, boring yet practical solutions are hardly worth discussing.
It was a lone comment that actually has several compelling implications that we would hope the TuSimple company thinks about. Each step of that has to have functions related to those actions backed up by many neural nets. And if it's hardly worth discussing, why is it the only comment you replied to in this entire thread?
What parts of "pull over" related to an autonomous semi driving on a freeway looking for a spot that's not sheer, a bridge, unstable rock, mountainous, long enough and doesn't leave the ass in the freeway, not blocking an exit, etc.. is not harder than actual rocket science?
I wonder if they would respond better or worse than a human. Do tires blow often on large trucks? If not a trained system could have more fed data on what to do in that situation than a human who has never or rarely experienced it.
Standard response is to explode the entire truck in place on the highway. You can upgrade for a nominal fee which will cause truck to pull to the next available side of the road in case of tire failure.
What are 3.5 million people going to do if they're suddenly out of a job and have to change career completely?
I am far from being a Trump supporter, but it would be easy for authoritarian politicians to ride to the White House on the grievances of millions of voters who had their livelihoods taken away from them by automation. Trump's promises to bring back manufacturing jobs were fueled by the automation and outsourcing that have already eliminated millions of positions. So Trump's victory in 2016 was already due in part to a backlash against automation.
No society where an average person - of average intelligence, average skills, average potential - cannot make a good living for themselves, can survive as a civilized society.
We can't all be developers, engineers, or doctors.
The reason we don't live in sub third world conditions like our ancestors of the 1700s is that with the Industrial Revolution, a few % of jobs have been automated each year.
Through this history people have always been worried about those whose jobs are automated, but they've always, on average at least, managed to find new and better paying jobs.
It's important to understand that 3.5m jobs won't disappear in a month. It will take at least a decade to switch around an industry this big, and people will have time to adjust.
Also, this means transportation will be cheaper and faster (computers don't need sleep), which benefits everyone.
> No society where an average person - of average intelligence, average skills, average potential - cannot make a good living for themselves, can survive as a civilized society.
I'm actually also worried about this, but I think it's a separate issue.
If fully autonomous long haul trucking was buyable today, then it'd still take 10+ years for it to be phased in. And even then, your 3.5M people number likely includes both long haul and short haul trucking. Short haul is entirely different ballgame. So, a more likely scenario than 3.5M people losing their jobs suddenly is that the trucking industry stops hiring, and mix of short/long haul jobs shifts from its current mix to entirely short haul.
I'd also be interested in what the median age range is for truckers. If adoption is sufficiently long term (20 years wouldn't surprise me), the economic pain might be surprisingly minimal due to people aging out of the workforce.
What are 3.5 million people going to do if they're suddenly out of a job and have to change career completely?
There's a precedent for this. Back in the early part of the 20th century the number of people working in the horse trade fell dramatically as horses were replaced with automobiles. Society survived.
My grandfather grew up in an orphanage and was later sent to train to become a wheel maker, that is, someone who makes large wooden wheels with spokes for horse carriages. This must have been around 1925. Luckily for him, the skills he learned were quite easily transferred to general carpentry and cabinet making.
Sure, society survived, but the decades after the industrial revolution/jobs realignment were wracked by insurgency, terrorism, the rise of the communist, fascist and anarchist movements, and culminated in two ridiculously destructive wars and a catastrophic depression.
I'm not saying those things will happen now too (we would never see extremist violence in this day and age, would we?) but given that history tells us that when large numbers of young-middle aged men are underemployed, you get violence, crime, and deaths of despair, perhaps we would be wise to spend some time thinking through how to help those people reestablish their connections to society.
For that to happen, you actually need working autonomous trucks.
Sadly, press releases, PowerPoint decks, heavily edited videos on YouTube, and even artificial demos in tightly controlled environments are not enough.
> What are 3.5 million people going to do if they're suddenly out of a job and have to change career completely?
Retire, given the average age of a truck driver is between 45 and 55 depending on which article bemoaning a shortage of truck drivers you care to read. In practice, of course, 'under limited conditions, maybe' isn't going to automate away 3.5 million jobs, and we still hire people to drive trains despite full autonomy for trains being well-established reliable tech.
You're getting downvoted to oblivion, but it's an important question - though perhaps you could have at least explored some possible answers a bit in your comment.
Inequality in society, driven by poor early family life and environment for many (leading to many issues including poor education) is a real problem for 'the West', and especially the US.
I agree. Liberals and most people in the tech space believe Trump got elected due to racism. It just is not the case. He got elected by being an opportunist and latching on to the real biggest issue in America, the decades of downward pressure and degradation of lower and middle class families.
There just are not enough opportunities for average or poor people to find meaningful quality work. Going to work as a cog at some global chain store will keep you alive barely but good luck being able to buy a home or have children.
Housing is through the roof, the education system is a scam using undischargable debt to turn our youth into indentured servants. Almost every American is 1 health care crisis away from bankruptcy. We are living in a powder keg.
I am seeing a lot of evictions where I live. Food bank lines are miles long.
It feels like identity politics are being used to distract us from how dire things are becoming for literally everyone not in the 1%
Let's just hope autonomous trucks displace the dangerous "king of the road" drivers with little regard to safety of those around them out of jobs first. Getting that type of trucker off the road can't come fast enough.
A common issue with automation, capitalism, etc. People may lose their job and will have to find something else to do, but if it's not a sudden transition, the amount of drivers will be reduced organically; it'd be a matter of less new vacancies opening up, and the career of driver becoming less attractive, so people will find something else.
It's a valid remark to make, since people lose their jobs to automation or outsourcing or whatnot all the time. But overall, global unemployment rates don't seem to be affected by new inventions, instead new jobs are invented all the time.
I mean a shit job that nobody should pursue as a career now is gig economy stuff; underpaid taxi driver / delivery person, there have to be millions doing these kinds of jobs now.
Anyway I do agree, not everyone can go for the high tech / high paid jobs. But the country can accommodate for it; give everyone equal opportunities by funding education (instead of relying on meritocratic scholarships and / or lifelong loans (the phrase indentured servitude comes to mind), do a state fund and set price limits on education). Make more jobs viable by raising the minimum wage, restricting the gig economy, setting work hours and day restrictions (e.g. 0 hour on-call contracts), basic worker protection laws and stuff. And finally, applying downward pressure on the cost of living so everyone can live comfortably regardless of job.
Everyone should be able to do a 40 hour work week if they choose to, no more than that unless they get paid 150%, and less if they choose to. And people working like that should be able to live in a reasonable distance from where they work.
The alternative is the "working poor", homelessness, working multiple jobs, working 80 hour weeks just to meet bare necessities, etc. This is already the reality for a lot of people.
Add it to the pile! http://rodneybrooks.com/wp-content/uploads/2020/12/SelfDrivi...
(from recent HN post https://news.ycombinator.com/item?id=25706436)