Filed under: Emerging Technologies, GM
Autonomous vehicles could debut with automatic valets
The Tartan Racing "Boss" Tahoe that won the recent DARPA Urban Challenge
We've all been there. Rolling up and down the aisles at the mall or the movie theater trying to find an empty spot, or waiting anxiously for someone to finish loading their car only to find them going back to another store. Or else you see what looks like an empty spot beyond the Suburban only to find a Honda Fit tucked away behind it. Then of course there are the times that you've raced through the rain from your car to the store. According to General Motors VP of Research and Development Larry Burns, studies have shown that twenty percent of the miles accumulated by drivers in Manhattan are spent trolling for parking spaces. That stop and go driving wastes a lot of fuel.
Recently the development of autonomous vehicles has seen a lot of progress, particularly as a result of the DARPA Urban Challenge. While real, fully-autonomous vehicles are probably still a ways out, some of that technology may appear sooner than you might expect. The first real implementation of fully autonomous vehicles could be in the realm of parking. A parking lot or garage could easily be equipped with sensors that detect the presence of cars in each space so that a computer knows where room is available. A vehicle with a communications system that could talk to the parking system could search for an empty space, reserve it, and then automatically park itself after you get out right at the door. When you come out, you can automatically retrieve your car and never have to worry about remembering where you parked. Imagine it, no more idling in the aisles. According to Burns, we could see this start to appear in about five years.
[Source: General Motors]
Reader Comments (Page 1 of 1)
Ron Fischer 12:52AM (1/08/2008)
Autonomous vehicles trump so many other technologies. Robotic parking deck? Personal Rapid Transit Pods? Don't need 'em! The question is how will they be introduced? Adaptive speed control and lane-holding are entering the market, as well as Toyota's automated backup parking. Question is what are the hard advantages to consumers and car companies? For instance, what if Herz and Avis start renting autonomous car time? They become large, responsive, versatile taxi companies. Would that cause enough consumers to give up their cars? Does that drive profit down or up for major automakers? Does that increase average MPG enough that we don't need plug-in hybrids? Diesel? Or perhaps short range electrics start will be enough...
Reply
GenWaylaid 3:59AM (1/08/2008)
It should be pointed out that the vehicles in the DARPA Urban Challenge moved at only a few miles an hour and were literally stuffed with computers, most of which were dedicated to processing sensor information. Even at those speeds they ran into each other with distressing regularity. Autonomous automobiles are quite a bit further away than press reports make it seem.
An automatic valet would be possible if the parking lot/structure had some way to indicate which routes were safe. This could be as simple as a line painted on the ground, a strategy widely employed by automated parts-delivery carts in factories and even a few city buses in Europe. Today's technology definitely could handle the rest of the job, which mainly would be detecting unexpected obstacles.
For the farther future, the prospect of driverless taxi fleets has tremendous potential to change transportation since they would come nearer to replacing all the handy aspects of a personal car than any other public transport. Not just Avis and Hertz but Zipcar and the local taxi companies could find themselves offering very similar services.
Perhaps the driverless taxis (or rental cars, or shared cars) would be something like the "Johnnycabs" in the movie "Total Recall." Schwarzenegger's exchanges with the robotic cab drivers also perfectly demonstrate why "artificial intelligence" is still a contradiction in terms.
Reply
TX CHL Instructor 11:53AM (1/08/2008)
Too bad our government has deliberately crippled the current GPS. That could provide another enabling technology.
I really like Ron's idea about the car rental business.
I wonder if I will live long enough to get into my car, specify the destination, and read or nap until I get there. Now *here's* an idea -- can you imagine the end of drunk-driver-caused deaths?
Reply
sila332 3:52PM (1/08/2008)
Great Comments!
I think Ron asks the most important question of all; How will an automated transport system find its way into our present transport infrastructure? Part of the answer is that many of the required technologies (telematics, car to car communications, and computational abilities)are already being introduced in high-end automobiles. Perhaps an even more important step, though less intuitive one, would be to severely constrain this technology's footprint. Automated vehicles must be segregated--physically-- from other traffic. With even 20% of the freeway's overall width (perhaps the shoulder and the carpool lane)a well designed automated system could move well over 50% of a typical freeway's capacity.
This constraint of the technology would focus the capacities that were demonstrated at DARPA, requiring navigation of a two rather than three dimensional space. Of course, the devil is still in the details!
Feel free to e-mail me at sila332@gmail.com
Reply
fnc 4:58PM (1/08/2008)
It would be incredible if the automobile manufacturers were able to mandate some sort of wireless communication standard for automobiles, and put it in all new cars. There are applications we can't even yet fathom for cars that have the ability to get information from the surrounding environment. It's one of those enabling technologies, like the internet, that allow for the creation of tools and uses not even imagined by the people who designed the underlying structure.
Parking garages that guide you to an empty space would be one small area in which a communications capable car would be of immediate benefit.
Reply
Ron Fischer 6:45PM (1/08/2008)
One additional comment: such a system does not need direct connection to steering, brakes and accelerator. Nav systems could say "accelerate to 45 mph" or "slow down for safer following distance" or "recommend using open left lane". In parking decks the initial automation cost is a sensor at every stall (cameras can cover many). Stalls could be assigned at entry. Extra fancy: locally augmented GPS guides to a spot. The point: many telematics features could be introduced without directly controlling the car. In this way infrastructure can be built incrementally, without one huge project's cost.
Reply