In the meantime in Japan we have automatic parking buildings using robots for like... 15 years or something. Lifting the cars up to store it vertically with minimum waste of space.
I assume it knows it is "your car" by the location where it dropped it, not by reading license plates or some other form of id. This would make moving cars around a little trickier, though I guess you could always keep the latest location.
I wonder if it optimizes for your return. This is, if a person is not returning until 5 days from now, maybe that car can be moved farther out in the parking lot to make room for cars of people who are returning "soon."
Basically, all cars would be placed "far away" and then slowly move to the front.
Maybe it's a "small" airport and this is not needed.
At 6MPH and, I'm assuming, a non-trivial time for even "loading" and "unloading" a car, I would not be at all surprised that the total time it would spend reracking cars would end up being significantly longer than just leaving them where they are and taking the extra time to fetch them.
Yeah, particularly if humans also park in the same spaces. I thought after posting that if the robot knows a space is available far away, then maybe by the time it gets there the space is taken by a human! :-)
I can envision tons of opportunities to mess with this robot.
Those exist, however they handle rush hour rather poorly. Also, when the system malfunctions, vehicles are not retrievable. Human-operated valet service is a very inexpensive, very reasonable alternative.
There's still the matter of all the owner's crap in the car, and current EVs would need nearly as much engineering as ICE vehicles to sit vertically anyway.
For example, the Model S still has brake fluid, washer fluid, coolant, and transmission oil, all which are stored in conventional automotive containers and which would absolutely perform abnormally if the car were raised to vertical.
I also suspect the transaxle has open breathers which would drain oil in the vertical position.
EVs do contain far fewer moving parts and reduce drivetrain complexity over ICE cars, but in every production EV design I've seen you only save one fluid (engine oil). Tesla just hide the car's fluids from their owners by putting plastic covers over everything but the washer fluid and then demanding you take the car to an "annual service."
Interestingly enough: there's countermeasures you can take to prevent oil migration! GM actually had some rather interesting experiments in this area. They tried storing cars vertically for transportation[1] in the late 60s and early 70s.
> ... Chevrolet's goal was to deliver Vegas topped with fluids and ready to drive to the dealership.
> To do this Vega engineers had to design a special engine oil baffle to prevent oil from entering the No. 1 cylinder, ...
There are also non-technology reasons you can't store cars tilted 90 degrees. People leave small items, half-finished sodas, etc. in their car and wouldn't want them all on the back windshield when they get back. Not to mention, in a high-density packing, you don't save any space by tilting the car 90 degrees.
Ahhh crap. I was returning a rental car at DUS a few days ago and saw this very setup. My son wanted to hang around and watch a car get lifted and I decided to move on.
Oh I was curious too: the whole area is lit up like Tron and painted with racing stripes all over it, in the bottom floor of this large, dark concrete garage. It was really cool.
However, it was late (22:00) and there were just no cars checking in at the time. I had a mental note to find a YouTube video and then this link popped up... =)
Yea but at one point we need to figure out what to do with our lower-educated part of society. Not everyone can become a programmer. And once we kill all blue collar jobs and replace them by either robots or self-service, then what's left for humans to do?
I don't believe these people should be marginalised to menial or tedious jobs with little appreciation and a pittance as pay. Essentially to render them invisible to the larger operation of society. This might be hopelessly naïve on my part, but I believe they can be just as productive in a different capacity.
Trade schools and apprenticeships can greatly reduce the burden society must endure in order to level the field for non-skilled labour. I don't believe white and blue colour jobs need such a strong line dividing them and so I'm sure new avenues for selling services or exchanging things of value and different definitions of what we value will emerge.
Of course, this is now getting into an area outside the scope of - and much complicated than - a robot parking system ;-)
I believe they can be just as productive in a different capacity.
1. [Cynically] the unemployed are productive, as a discipline/threat to the employed -- "do as we tell you or you will end up unemployed" [with long list of implicit penalties].
2. The whole idea that humans must work to live is a weird cognitive bias in western political discourse. If you look at actual employment rates they max out around 40% -- children and pensioners don't have paid employment, homemakers aren't paid for their labour, many folks are only partially employed. Yet our whole dialog around how to structure society is based on the axiomatic supposition that full employment is some sort of religious directive.
Wouldn't we be better off looking for more culturally nuanced ways to measure human value than mere fiscal utility over time?
If i didn't have to work I would like to run more and become an archeologist. I would have no problem just living of the robots fruit of labor. Like someone born rich basically.
My car (a 2010 VW Tiguan) parallel-parks itself - well, in as much as it operates the steering while I handle the pedals and gear changes - current models also reverse into parking bays.
This is also a standard feature on the top-end model of the Ford Focus, and probably a bunch of other cars. They don't even need reversing cameras for it - just some carefully positioned 'parking sensors'.
The biggest advantage to self-driving cars that I see is that I can drive a car into a city center, drop myself off, then have the car wander aimlessly through the streets until I need to be picked up. The cost of the gas is almost certainly cheaper than the cost to park would be.
You don't want it to drive around, that's pointless. (Also, petrol's not cheap everywhere.) It could just zoom off to an out of/edge of town car park, and presumably plug itself in to charge while it waits.
The biggest advantage to self-driving cars is that you don't have to buy a car, so the car doesn't wander aimlessly, it just picks somebody else up and taxis them around.
A car, as currently used, is not just a transport mechanism. It's also a storage mechanism. (Umbrella, sunglasses, kids' car seats, emergency diapers, emergency cash, emergency feminine products, OTC meds (ibuprofen, antacids, etc), a reasonably secure place to keep my backpack if I don't want to carry it, mints/candy, etc.)
Automating a taxi service doesn't 100% replace the utility of a personal car.
Because of rush hour, a large fraction of self-driving cars will still need to park. However, they can drive themselves out of the city center and park in the cheaper outskirts.
Why would I want to use the same car as other people? It will be dirty, because people will do all sorts of things in there,and I would feel very uncomfortable.
In most cases that's not really an advantage at all. We already have a system like this: public transport.
Public transport seems to be generally rather bad in the US, which might make self-driving cars seem useful but other parts of the world often have better public transport and self-driving cars seem a lot less exciting here.
Annnd reality surpassed your idea already; some cars have a system using cameras and such that automatically parallel-parks it. One of my colleagues has a car like that, it's quite fun to sit in a self-parking car.
> Sensors in the station measure the dimensions of the vehicle.. The robot adjusts its arms to fit the new vehicle
I wonder whether they considered classifying the make/model of the car, perhaps using a simple computer vision approach, and lookup the dimensions in a database of known models.
That approach would require constant database updates as new makes/models come out. Additionally, it wouldn't be able to handle vehicles that have been customized with things like larger tires. Scanning every vehicle makes the system much more set-and-forget.
I was thinking this very thing. Once you decide to support all cars, then you need to build in the smarts to automatically measure and the database of models is no longer needed.
$40 dollars a day or $5.50 an hour! I think will park my car myself. Most valets cost half that with tip included.
How can a 1.2 million dollar be feasible when you can hire or even create a valet company for fractions of the costs? I trust a human with my car over a robot at this point in time.
Parking yourself isn't that much cheaper. Actually can be even more expensive than $40 a day and right now this robot is only cheaper because you get a special deal.
Atlanta, the most I pay for parking would be at a sporting event or concert where it's usually around $20 for the event. The only $40 or more parking would be at the airport.
Chief executive Rupert Koch claims using the Ray can increase a parking structure’s capacity by 40 percent.
I'm a bit doubtful about this, since the apparatus is wider than the (typical) driver and would have to leave plenty of space between the cars it drops.
Maybe they pack cars in five deep and then dig the cars out when required. Also there doesn't need to be a gap for people to get in and out between cars.
Parking spaces generally have about 2.5ft of extra space for door opening. Airports often a bit more due to people carrying bags.
It's also the fact that a Smart car takes up the same space in a parking garage as an original Hummer. I've been in a Costco parking lot where someone in a Mini has parked over the line into the opposing space (overlapping on the front not the sides). I drive an F150 so I pulled tight on him and bumped him and the parking gear rolled me an inch away. I didn't grab much (I think it was pop, water and hotdogs/burgers for a party) and got out while he was still loading his car. I dropped my tailgate and started throwing the cases in (and pushing against the tailgate) so that my truck rocked and hit his car. The guy said "Hey buddy, you're hitting my car." I walk around, look at the fronts of the vehicle, take a picture with my phone (my trucks a company vehicle so we get targeted by all sorts of morons who think we punctured their tire) and I said "I don't know, it looks like you hit me. I'm well within my space."
The TL;DR: to that was "People are idiots who can't park small cars in big spaces" so I'm quite confident parking efficiency can be increased dramatically by decreasing wasted parking space.
You can also presumably halve the amount of driving lanes required. If only 3 robots serve the entire complex there should be enough routes to never run into them having to travel the same lane at once. That alone could likely increase efficiency dramatically; if a lane is wide enough to drive in, it's wide enough to park a car in, and by a guesstimate with no room between vehicles you could likely pack 3 vehicles wide in the width of 2 driving lanes.
You also mention packing the cars in. If the software has access to boarding times, and return times then they could presumably pack vehicles in 5 deep or more. You could have a parking space 10 vehicles deep, if you've got lane access on either side you put the cars in ordered by arrival time. Row 1 gets in at 3pm, row 2 gets in at 3:30pm, etc. and you just unload from the opposing side.
If you made an entire parking lot structure accessible this way, you could presumably increase parking efficiency well beyond a 50% increase with the reduction of all driving lanes.
Yeah... I vandalized their vehicle in that no damage whatsoever occurred because it was stationary vehicles and these vehicles come with these contraptions called "BUMPERS".
"collided" is a gross exaggeration BTW. This isn't the 1920's cars come factory installed with bumpers.
As for fault it's a parking lot. Under the law in Canada it's the fault of whomever didn't obey the posted signage. If you can't park inside the lines and an accident happens, you're to blame.
I dont think the sensors would detect any movement unless a problem happened. The wheels don't rotate and the movement does not seem to have acceleration high enough to trigger the alarm. Many alarms today don't even get triggered when the car is hit by a ball, for example.
This is interesting (never had an annoying alarm)! Seems sound waves create more disturbance for the sensor than an actual blunt impact which is localized to a point on the body and possibly not transferred through the chassis to the sensor.
I don't see how it would be an issue. A parking garage is out of the way anyway so wouldn't annoy anybody nearby. Nobody should be in a giant automated garage so nobody to annoy, and any security threat would be monitored for visually not auditorially.
Not really a concern unless the cars are being shuffled constantly. It takes a long time for an alarm to drain a battery, believe me I've had some awful neighbors.
[0] https://www.youtube.com/watch?v=ChOHnSL7ZCg (warning: NSFW, gore, German humour)