Another Step Towards the Driverless Car

jtogel writes "At Essex, we have for some time been working on automatically learning how to race cars in simulation. It turns out that a combination of evolutionary algorithms and neural networks can learn how to beat all humans in racing games, and also come up with some quite interesting, novel behaviours, which might one day make their way into commercial racing games. While this is simulation, the race is now on for the real thing — we are setting up a competition for AI developers, where the goal is to win a race between model cars on real tracks. As the cars will be around half a meter long, the cost of participating will be a fraction of that for the famous DARPA Grand Challenge, whereas the challenges will be similar in terms of computer vision and AI."

Not just for older drivers

[rolfwind]

I would love to have a driverless car: let me get some work or reading done while lounging in the back seat (safer) of my car while it is driving me through the daily rush hour. Because I can get work done, I can either drive off later or am in less of a rush to get where I'm going. No more tedious trips of hours upon hours of driving.

Lower insurance premiums - and if the car has an fender bender, I can point to the manufacturer and hopefully won't be branded as an unsafe driver for life if I didn't do the driving. Safer roads for all. A recent study (posted on /. I think) said humans have trouble paying consistent attention to anything for an extended period of time without having our minds wander:

http://www.localnewswatch.com/skyvalley/stories/in dex.php?action=fullnews&id=80157 [localnewswatch.com]

We'd be more safe.

No traffic tickets - the AI can go closer to the speed limit than I have the patience to (now if they didn't consistently set the speed limit too low in a ton of places just to be asshats and be able to write tickets when they need the money....)

Seriously, this isn't just for the elderly. Driving ceased to be fun for me long ago. If I had to do it only once a week on a nice stretch of fast highway, I might feel differently....

Where are the Competition Specifications?

[TempeNerd]

The links go to an AI presentation of virtual cars and a news release saying there will be a competition without any details about the competition. Did I miss something? Is the only news that they have developed neural nets to drive a virtual car?

The competition sounds like a manageable project for academics (versus the DARPA event).

Is the competition still in the vapor-ware or maybe-someday stage?

Anyone have a link (perhaps IEEE) that has details?

I would love one

[icj]

Being legally classed as being blind, I would love to have a driverless car. I would live the independance it would bring.

Re:Forza 2

[jtogel]

I don't know much about the techniques underlying Forza 2, but I went over and talked to the guys who worked on Forza 1, and we compared our approaches. At least for the first game, what they are actually using is recorded trajectories on different track segments which are then spliced together at the junctions of segments, so as to create similar-looking behaviours on unseen tracks. The problem here is of course that the new tracks are constrained to being constructed out of the same segments as the driver has already been tested on - there is no generalization. The track designers for Forza simply had to live with this constraint.

We have ourselves gotten player modelling working fine with evolutionary neural networks, which can generalize, but the Forza team didn't consider these techniques reliable and fast enough in time for the release of the original game. Maybe things have changed with Forza 2.

There is some information on the Forza AI on http://research.microsoft.com/mlp/forza/ [microsoft.com], and our approach to modelling is described in http://julian.togelius.com/Togelius2006Making.pdf [togelius.com].

Note that all this is about modelling behaviour, not about creating new behaviour from scratch; there are some papers on this on my website [togelius.com] as well.

Traffic

[et764]

A few months ago I was thinking about doing something like this, but in heavy traffic situations. What if the course had way more cars than should ever actually fit, and the cars independently tried to minimize their travel time around the course. I wonder if the computer could get a better overall throughput than people seem to do on the crowded highways of, say, Seattle.

Safety vs. Freedom , again.

[Original Replica]

Make it either Opt-In or by Court Order

Now this, to me, is a very important distiction. What if for this to work well, all the cars have to be computer controlled? What if computer control is then mandated? This is a whole new exciting level of "nanny government". Sure this might be safer in that there would be fewer auto accidents, but do you really want all transportation to be centrally controlled? Sure each car might be autonomous at first, but emergency workers need the ability to remotely turn on off, right? It's for everyone's safety.

The problem isn't in the technology

[RandomWordGenerator]

The problem with driverless cars isn't the technology but insurance.
Many manufacturers have been dissuaded from pursuing the technology and installing in their vehicles because in the case of any accident the corporation would be liable. Obviously the 'driver' wouldn't be at fault because they wouldn't be driving.
No large corporation is going to put itself in line to pay out on every bump, scrape and minor slaying caused when their killer robo-cars Attack!

Re:And try this against a real racer

[tarball]

Yeah, except the boring oval tracks, which I probably dislike more than you, would be the easiest place to kick the butt of a program controlled car. Because the program doesn't feel the dynamics of the air or the track like a human does.

A boringly smooth road track with an entirely predictable car, like most F1 tracks and cars are becoming, would be the ideal combination for these algorithms. No one passes anyway, so maintaining position until a perfect pit stop occurs is the way to get ahead. Good pit in and out laps win races nowadays. This is where perfection in an algorithm could possibly win, whereas NASCAR is too noisy an environment for it to work.

That being said, I don't watch NASCAR (much), and I am an F1 junkie.

Hamilton is very likely the F1 version of Tiger Woods.

tom

Re:Traffic

[et764]

I think the reason why we don't all start accelerating at the same time when the light turns green is that we normally stop closer to the car in front of us than we are comfortable driving. I frequently stop less than a car length behind the one in front of me at a light, but I don't feel comfortable driving that close behind a car. The reason for the lag is to leave a comfortable distance between you and the person in front of you.

Re:Safety vs. Freedom , again.

[sumdumass]

The other posters fears about unaccounted for circumstances are unfounded. If there was some problem with the program it would default to safe mode (eg car would stop)

That would be great if there was a way to make sure every other car would stop too.

We have autopilot on planes (in fact even planes that can land and take off by themselves), we have satellite navigation, we have remote driving of cars - so why not go the whole way and allow an artificial intelligence to do it?

There are some differences here. First that aren't near as many plane with autopilot and those capabilities compared to cars. Second, the planes flight is planned and all the other planes are aware of possible close calls. You simply couldn't do this with cars and have the level of communications you would have with planes. Planes have control towers that follow the planes and inform others of them. Not only can they go around the traffic, they have the ability to go above or below. We don't have that with current cars. And finally, planes use expensive radars that look not only for objects but weather hazards that let the autopilot or pilot adjusting the outopilot to make adjustments.

The problem is programs thrive on predictibility, even with built in radar, two camera so it can judge distance etc the drivers on the road are always going to act in a way the computer can't predict (foreigners trying to drive on the wrong side, drunk drivers, drivers talking on their cellphone or listening to the radio distracting them from driving). Like TECAS in planes I'm sure AI could now be used as a collision warning/automatic avoidance system in cars which would massively reduce accidents (even if only one car in a potential collision was fitted).

There are some cars, And I'm sorry i forget their brand, that use sonar to gauge the distance of the vehicle in front of you while driving. It you get too close for your speed, it give an audible warning followed with cutting of engine power and finally applying slight pressure the the brakes as you get closer. You can override this in some situations by depressing the gas peddle. In California, Going across the hills in hte snow prone areas, there are sensors imbeded in the roadway that allow the snowplows to check if they are in their lanes during whiteout and low visibility conditions. So, yes, they are being used as tools to assist the driver in being safe. I would hope it becomes more usable in the future with more applications being available.

Some of the challenges to having this would be making sure all the systems can inter operate with every other manufacturer, Putting some sort of traffic control system in place other then regular streetlight and stop signs, and embedding some sort of positioning system that is more accurate then GPS. (remember that darpa challenge entry who was disabled by a drop in the GPS signaling over some sort of correction)

I think we are a long way from using them in mass on the open road.

wow, what country are you writing from?

[fantomas]

No option to walk? you should speak to your political representatives and get pedestrian walkways put in either side of the road. Even sub-Saharan African countries can afford to do this (see page 27) [eco-logica.co.uk] - it doesn't have to be expensive - so I guess your country needs to think about its priorities on spending. Which country are you writing from?