Self-Driving Car Faces Off Against Pro On Thunderhill Racetrack 151
Hugh Pickens writes "Rachel Swaby reports that a self-driving car and a seasoned race-car driver recently faced off at Northern California's three-mile Thunderhill Raceway loop. The autonomous vehicle is a creation from the Center for Automotive Research at Stanford (CARS). 'We tried to model [the self-driving car] after what we've learned from the best race-car drivers,' says Chris Gerdes (who talks more about the development of autonomous cars in this TED talk). So who won? Humans, of course. But only by a few measly seconds. 'What the human drivers do is consistently feel out the limits of the car and push it just a little bit farther,' explained Gerdes. 'When you look at what the car is capable of and what humans achieve, that gap is really actually small.' Because the self-driving car reacts to the track as if it were controlled in real time by a human, a funny thing happens to passengers along for the ride. Initially, when the car accelerates to 115 miles per hour and then brakes just in time to make it around a curve, the person riding shotgun freaks out. But a second lap looks very different. Passengers tend to relax, putting their faith in the automatically spinning wheel. 'We might have a tendency to put too much confidence in it,' cautioned Gerdes. 'Watching people experience it, they'll say, oh, that was flawless.' Gerdes reaction: 'Wait wait! This was developed by a crazy professor and graduate students!'"
Seconds? (Score:4, Insightful)
Seconds aren't "measly" in motorsports. They can decide an entire season championship.
Re:Seconds? (Score:4, Interesting)
Outside the race track, who cares? It is like saying my processor is 1 Mhz better than yours.
Re:Seconds? (Score:5, Funny)
And that would be technically correct. The best kind of correct.
Re:Seconds? (Score:4, Insightful)
Similar to ABS, this has potential for saving John Driver's butt when he gets in over his head.
Firstly, if all robot cars can successfully drive at the limit of performance, avoidable accidents will cease to occur. Race drivers might be able to dodge a spinning car or deer in the blink of an eye, but John Driver can't.
Secondly, taking a page right out of ABS's playbook, one of the places human drivers have the most difficulty is variable or unusual conditions, like rain or loose surface (gravel, dirt, etc). Robots, as ABS has proved, are quite valuable in this sort of situation. While they cannot perform at the absolute limit, they can consistently perform close to that limit, while even the most skilled humans can have difficulty performing at the peak in variable or unusual conditions. Rally drivers may do it for a career, but John Driver isn't a rally driver.
Re:Seconds? (Score:5, Funny)
But clearly autonomous cars are bad for everyone because of (insert fringe case) where the car can't perform above 98.5%. Also, humans are better at (thing that rarely happens,) so since I think I'm a better than average driver (just like 90% of people think they're above the median,) then this is clearly a failed technology that shouldn't be allowed to be used by anyone.
Also, I don't use a seatbelt, airbag, or ABS because I'm not fooled into thinking that researchers, engineers, empirical evidence, and years of track record prove that these things make driving more safe. Obviously me, an IT professional, knows better.
Re:Seconds? (Score:5, Interesting)
I've done rally racing and I'm pretty damn good at it, and I'd still welcome an AI assist for my daily driving. Would I want it in a race? Probably not just yet, but when I'm driving to a job site I am not racing and would love to have the extra protection.
Re: (Score:3)
Screw the assist. I love to drive, and I'll never want to give up the option of manual control completely, but I would MUCH rather spend my commute reading, playing music, writing, watching movies, or fucking while my car automatically gets me to work. Give me the whole enchilada and not just some driver augmentation.
Re: (Score:2)
I am Lashat and I approve of this message.
This is right. Semi-Pro Racer here. The last thing I want is my vehicle second guessing my input in white knuckle manuevers. However, I would enjoy kicking back and playing with my kid while during the commute.
Re: (Score:3)
Also, I don't use a seatbelt, airbag, or ABS because I'm not fooled into thinking that researchers, engineers, empirical evidence, and years of track record prove that these things make driving more safe.
Ok, you're joking, but many people actually believe that seatbelts and air bags are more dangerous than not having them, and I even doubted ABS. I was trained as a driver in the USAF and knew that in a skid the vehicle will travel farther than when the brakes are almost at skid level, and that when in a ski
Re: (Score:2)
Re: (Score:3)
Seconds depends on the length of the track. There are tracks that can be cleared in less than a minute, and a few seconds is a significant difference. It also depends on the type of car being used... track records at Thunderhill vary from 1:37 for a lap to 2:15 depending on the type of car, and TFA doesn't provide any information as to the type of vehicle being driven. ( http://www.sfrscca.org/index.php?option=com_content&task=view&id=4448&Itemid=93&selTrack=Thunderhill+Raceway+Park&selL [sfrscca.org]
Re: (Score:2)
Re: (Score:2)
Outside the race track, who cares? It is like saying my processor is 1 Mhz better than yours.
Well this was on a race track so it matters quite a lot. If this was about driving to work I could understand you point, but the experiment was to exactly to demonstrate the minute detail that differentiates the winner and loser on the race track.
Re:Seconds? (Score:4, Informative)
AC is correct.
In motorsports few seconds is a very long time. The lap times are not mentioned.
The raceway in question is probably this: http://thunderhill.com/staticpages/index.php?page=TrackMap [thunderhill.com]
But which variation? Long version 2.866miles record times tend to be just over the 2minute mark for somewhat normal cars.
Short version is 1.769miles for which SCCA website is missing the record times, the medium version is 1.814miles and record times tend to be close to 1:30 mark with somewhat regular cars.
Also they don't say how good race car driver was the AI against, there is a huge variety of race car driver skill levels.
Few seconds? They are being vague, i bet it was more than just 2 seconds because they are being vague. :)
Some racing series have 3% rule to qualify, ie. within 3% of the best time, for 1:30 lap time that is 2.7seconds, in other words this AI wouldn't even qualify.
All that being said, great work! Got to start from somewhere.
In theory AI could become better than humans, but then again AI will most likely always lack intuition, so could well be that a human will always surpass AI.
Nevermind that a very highly skilled human with very high motivation can do some insane reaction and completely remove the guesswork some of the time when surpassing the limits, ie. see Ayrton Senna. For AI we'd need sensor capable of few ms polling rates with data returned, then compute all the data within few milliseconds and then some insane fast and accurate servos to achieve that level.
Few millisecond polling rate doesn't sound like much until you realize that for example USB has 90ms, PS/2 is in theory capable of 5ms, and serial port even faster, but that doesn't account for data transfer rates.
There's a reason why we cannot even build a simple ECU/EMS with standard off-the-shelf hardware: Polling rates are too slow.
Re: (Score:2)
Well the summary says it's a 3 mile track, so it sounds like they were using the long track.
I would have thought oval racing would be where computers could completely kick humans' butts. It's all about getting your turn in point just right, hitting the apex at exactly the right point, then applying the right amount of throttle to get away from the corner as fast as you can without losing grip. I think in a couple of iterations no human would be able to keep up.
USB latency is not necessarily 90 ms (Score:2)
According to this source [datatranslation.com] USB can get down to around 1 ms theoretically. In practical terms, you can get an expensive MIDI keyboard with input through USB, and that's certainly much faster than 90ms, otherwise it would be completely useless.
cool (Score:2)
but where's the video?
Re: (Score:3)
The TED talk has a few clips.
Re:cool (Score:5, Informative)
Try this: http://youtu.be/YxHcJTs2Sxk [youtu.be]
BRAKE (Score:5, Informative)
brake brake brake brake brake brake
the word is brake
Re: (Score:3)
Slow down, there, cowboy, before you brake Slashdot.
Re: (Score:3, Interesting)
Re:BRAKE (Score:5, Insightful)
In the context of the article, the word used to decelerate when coming into a corner should be "brake" not "break". The traditional use of the word "break" indicates that the car has become "broken" or is damaged in some way.
While it's true that the saying "make a break for it" indicates taking off in a different direction, there is no connotation of slowing down. In fact, it has the opposite connotation of picking up speed. In the case of cornering, you have to slow the vehicle down or centrifugal force will cause the vehicle to leave the road.
In this case you would BRAKE going into a corner and you then MAKE A BREAK out of the corner (i.e. quickly picking up speed to try to gain an advantage).
Re: (Score:2)
“Making a break” to me implies breaking the line that you were following, as when escaping a chain gang or somesuch.
Re: (Score:2)
In the context of the article, the word used to decelerate when coming into a corner should be "brake" not "break".
Only if you're literate.
Re:BRAKE (Score:5, Funny)
No.
If "the car accelerates to 115 miles per hour and then breaks" several fragments of it might make it around the corner as TFA suggest, but typically, the larger portions of the broken car will continue in a straight line until deflected or destroyed by a wall.
In that event, the idea that "Passengers tend to relax, putting their faith in the automatically spinning wheel" seems unlikely, unless we're talking about permanent relaxation and some kind of Tibetan prayer wheel.
Re: (Score:2)
sorry, pedantry (Score:5, Funny)
(This post brought to you by the collective might of the Oblivious Flaw In The Headline Committee, newly formed to point out the obvious flaw and thereby negating 50% of the discussion dealing with grammar and spelling.)
I can see it now (Score:5, Funny)
Turn left
Turn left
Turn left
Kill all humans
Turn left
Re: (Score:2)
From what I've seen lately, that's pretty much what NASCAR drivers are already trying to do.
Re: (Score:2)
Turn left
Turn left
Turn left
Kill all humans
Turn left
No no no no no... Thunder hill is a road course, not a Nascar circuit. It turns left AND right.
It will win soon (Score:4, Interesting)
A self-driving car doesn't have to pay much attention to the fragility of the human form when it doesn't have any on board.
Accelerate at 50g? no problem just add extra bracing.
Re: (Score:2)
Kinda misses the point of a car. Perhaps not a tractor-trailer, but cars are definitely not all that useful without passengers.
Re:It will win soon (Score:5, Insightful)
Kinda misses the point of a car. Perhaps not a tractor-trailer, but cars are definitely not all that useful without passengers.
Yes they are.
Re: (Score:2)
I know Americans are stupid about using their cars for every trip whether it makes sense to drive or not, but sending your car to pick up a box instead of letting the FedEx truck do it is absolutely retarded.
Re: (Score:3)
I know Americans are stupid about using their cars for every trip whether it makes sense to drive or not, but sending your car to pick up a box instead of letting the FedEx truck do it is absolutely retarded.
Only if you assume 1) that the FedEx truck is more efficient than your car, and 2) Waiting for the twice daily FedEx delivery cycle is a good use of your time.
But the point is that self driving cars open a whole range of possibilities that can't easily be done if you have to dedicate yourself to driving the car.
Re: (Score:2)
Re: (Score:2)
If the FedEx truck was also self driving than it would only make one trip empty.
Why make any empty trips? Do pickups as well as drop offs and the truck can avoid being empty altogether. Of course, that only applies to generic package shippers like FedEx and UPS. More specific delivery vehicles may not be able to do that.
Re: (Score:2)
Why make any empty trips? Do pickups as well as drop offs and the truck can avoid being empty altogether.
This doesn't work because the pickups and deliveries are mostly on different routes. Residential routes are mostly deliveries. Business routes are mostly pickups. Also, businesses want their deliveries in the morning as early as possible (repair parts, JIT inventory, etc), and the pickups late in the afternoon to ship out the days orders.
Downside (Score:2)
Re: (Score:2)
Taxi's would get a whole lot cheaper.
Re: (Score:2)
Taxi's what would get a whole lot cheaper? Their spare tires? Their gasoline? What??
Re: (Score:2)
Taxi's hire cost would get a whole lot cheaper because there would be no driver to pay for.
Re: (Score:2)
You miss an obvious problem... Modern race cars don't accelerate, corner, or brake anywhere near the limits of human endurance. Why? Those pesky physics get in the way. It's hard to make a car accelerate at 50G. Top fuel drag racers don't achieve anything near that kind of acceleration, and accelerating is pretty much all they can do.
If you want to talk about aircraft or rockets, that's a different matter.
(FWIW, you could probably make a self driving car much lighter than a human powered car by eliminating
Re: (Score:2)
Re:It will win soon (Score:4, Funny)
Retinas are breaking?
Re: (Score:2)
F1 is having drivers 'red out' and there have been some detached retinas under braking.
From BRAKING? Sorry, without a citation I simply don't believe that. From smashing into the wall at 200 mph, sure, but not braking. If that were true you'ld have shitloads of fighter pilots getting vitrectomies, let alone prize fighters.
Sorry, I'm calling bullshit.
Re: (Score:2)
F1 cars are closer to inverted aircraft than cars anyhow, but fair point. :)
Re: (Score:2)
The fragility of the human form is in no way a limiting factor. The factors are:
* The vehicles are designed to carry people
* They therefore have certain performance characteristics
* The computer was not as good as the person at pushing to the limits of those characteristics
Re: (Score:2)
A self-driving car doesn't have to pay much attention to the fragility of the human form when it doesn't have any on board.
Accelerate at 50g? no problem just add extra bracing.
Or add extra braking / breaking (see discussion above).
Betteridge's Law (Score:5, Funny)
No.
Re: (Score:2)
Re: (Score:2)
In a Race Between a Self-Driving Car and a Pro Race-Car Driver, what is that wooshing sound?
Re: (Score:2)
we lose
Horrible article (Score:2)
I was really disappointed, the title sounded really promising...
Re: (Score:2)
Re: (Score:3)
In the context of racing, you are correct : seconds are not measly
In the context of my daily commute, seconds are completely irrelevant. Even a few minutes are pretty measly.
Also consider the competition. The car is roughly on par with a professional race car driver... or rather a seasoned one, though the fail to mention what seasoning they used. Either way, that's already leaps and bounds ahead of 95% of drivers
Re: (Score:2)
Ultimately where you car is going is determined by four little contact points between the car and the ground, that has a certain grip (rubber on road friction). That grip can be used for accelerating, braking or keeping a turn. The math here is a few integrals and nothing humans do in their head, but for computers that's pretty straightforward. It doesn't make the car road safe though. The main reason your typical driver isn't so good at this is that you don't actually want to find out where that limit goes
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Also, did it have a detailed 3D GPS map of the course to use as a reference to pre-compute the optimum strategy?
If not, it would have had to figure out the best braking, acceleration, and steering strategy in real time, which would be much more impressive.
Why'd the human driver really beat the robot? (Score:2)
Re: (Score:2)
There's room for improvement (Score:2)
Re: (Score:2)
Don't forget the moose attacks!
http://www.theprovince.com/Moose+attacks+police+with+Prince+George+Mountie+inside/7447294/story.html [theprovince.com]
Re: (Score:2)
The beginning of the end... (Score:2)
One more step towards ideocracy and less freedom.
Get used to the idea that in a decade or two's time it will probably be illegal to drive a car manually. New cars might not even have controls, just a microphone to speak the destination into. Mothers Against Drivers, the government and all car manufacturers will successfully collaborate in brainwashing the general population into believing that humans are mentally/physically incapable of actually driving a car at all, certainly never safely.
Re: (Score:3)
Just like it's illegal to fly a plane manually? oh wait. More alarmist bullshit form the 'League of Alarmist through FUD."
And if it goes where you want, why does it matter?
Re: (Score:2)
Because, I for one do not want a future that resembles shitty Rush songs, red little boat indeed.
Re: (Score:2)
what about utility trucks that need to be place th (Score:2)
what about utility trucks that need to be place that just having a microphone / touch screen will be a very poor way to get them in to place to work on lines / even be used as a temp prop on a power pole to hold it in place.
Re: (Score:2)
The problem with utility truck drivers, they won't like not being able to stop at the strip club and bars along the way... They really hate GPS tracking...
Re: (Score:2)
I doubt these cars will be required but I will definitely get one. I hate driving and consider it an option of last resort when buses, bikes and walking are not available.
Re: (Score:2)
Re: (Score:2)
Its people like you that happily give up their rights that make the erosion of rights possible.
Re: (Score:2)
Should have had my coffee. I mean freedoms not rights.
I want to see (Score:3)
a dozen driverless cars designed to race go at this. Would emergent behavior appear? Can we make them so decisions are recorded and then applied to the next situation?
Re: (Score:2)
See: RARS [wikipedia.org]
a race track is a poor test for a day to day use o (Score:2)
a race track is a poor test for a day to day use of a car.
Re: (Score:3)
Re: (Score:2)
Rainy day one thing.
Rainy day with a bunch of kids, trees jumping out (honest ossifer), ostriches running around, moose attacks vehicle... Would depend on how reliable the system has proven to be.
Reasonably Priced Car (Score:3)
Personally, I would LOVE if Top Gear (UK) brought in this team to test how quickly they can make the Reasonably Priced Car go around the track.
Re: (Score:2)
Yes, please!
subject (Score:2)
"But only by a few measly seconds."
In auto racing, a few seconds is generally considered an enormous lead.
I've experienced that "freak out" in a way (Score:2)
Obviously I wasn't riding shotgun with a robot. Instead, I was on an airboat in the Florida everglades. The first time the captain pointed us straight at the mangroves and gunned the engine, I freaked. Then I realized that with an airboat, you change the attitude, drift, and then accelerate. This guy was a master. He was able to navigate through the groves with just a couple feet on either side (at slower speeds of course). After the first few high speed turns, I sat back and enjoyed it just like the
Breaks / Brakes (Score:2)
Re: (Score:2)
I could care less about that.
Re: (Score:3, Funny)
Not that I will ever trust an autonomous vehicle with my life.
Because you're irrational. I bet you'd trust a human, though!
Re: (Score:3)
I bet you'd trust a human, though!
Either way, you're trusting humans. The question is, which choice offers fewer points of failure? ;)
Re:ONLY a few seconds? (Score:4, Insightful)
Re: (Score:2)
And, unfortunately, software has little or no responsibility if it gets it wrong while a human does.
Don't think you'll see a CEO jailed any time soon because the company had inadequate testing procedures and they ran a kid over. Money will change hands, regulations will tighten, but people inside a software-producing company going to jail when their product fails? Very rare.
The other thing about humans is they have a superior (if slower) decision making capability. If you've seen the movie I-Robot (hardl
Re: (Score:2)
In the iRobot example, the robot saved the person they perceived to be the most likely to survive. A human would have tried to save the child and the more likely outcome would be two deaths instead of one.
Re: (Score:2)
Don't think you'll see a CEO jailed any time soon because the company had inadequate testing procedures and they ran a kid over.
I don't think so either. But then again, I also don't think he should.
How worried are you that a bridge you drive over won't fall if the winds pick up. How worried are you that the skyscraper you're on in LA won't collapse to the ground if an earthquake happens? These are all real dangers. And an engineer signed off on the design, indicating it meets all of the applicable codes. If something happens as a result of negligence from that engineer, his ass is on the line.
I am an engineer, and I know for a
Re: (Score:2)
The human brain goes through a little over 140,000 hours of development and testing before being set loose on the road.
Re: (Score:2)
Re: (Score:2)
Re:ONLY a few seconds? (Score:4, Insightful)
For every one of them that drives off a cliff, it'll save 100 people who would have not noticed a stopped vehicle ahead while checking over their shoulder while merging or running a red light because were momentarily distracted, or looked down to check the stereo for a second and drifted into oncoming traffic.
Re: (Score:3)
Humans are idiots. Distracted humans in a reflex situation doubly so.
Re: (Score:3)
Re:Been there, done that..... (Score:4, Funny)