Inside DARPA's Robot Race 135
Belfegor writes "The PBS series Nova has a great feature on their website, regarding the coverage of the DARPA-sponsored 'Robot Race' in which driverless vehicles 'competed' in a 130-mile race across the Mojave Desert. The full show is available on the website, and besides that they have plenty more information about the robotics behind the challenge, and also some pretty cool out-takes from the show."
Used to be a lot slower. (Score:3, Interesting)
But mostly these robots depend on the assumption that everything remains still.
Re:Used to be a lot slower. (Score:1)
But "Stanley" did pass whoever had been in first place while that vehicle was moving.
Re:Used to be a lot slower. (Score:2)
Surprisingly, Stanley seemed to have no trouble at all passing Highlander, but I wonder how much of that was just pure luck (and the fact that it happened during a stretch wher
Re:Used to be a lot slower. (Score:1)
Yeah, it would have been interesting to see what would happen if Stanley had caught up to Highlander at a spot where there wasn't room to pass. Would it believe that Highlander was a stationary object and stop moving, thinking there was no way to proceed, or would it just have followed behind Highlander?
I understand that it is not within the spirit of
Seen it (Score:5, Informative)
Seen it-One eyed. (Score:2, Interesting)
Re:Seen it-One eyed. (Score:3, Insightful)
This is actually not too far different from how human vision wo
Fascinating program (Score:2)
This was a fascinating program. It would have been nice if the Stanford team divulged more of their ideas, what software languages and designs they used etc. It looked like they were doing a Bayesian classification on combined laser ranging and video on the terrain ahead. Doing that for 1 image is complicated enough. Doing 10+/sec is mindblowing. The control system moderated the vehicle's need to follow a prescribed path with how safe the path was. Amazing stuff, very elegant. Pretty much done with a stock
Re:Fascinating program (Score:2)
Does anyone have any links that contain more gory details about the Stanford effort?
Re:Fascinating program (Score:5, Informative)
One paper that's of interest might be here: http://www.cs.cmu.edu/afs/cs.cmu.edu/user/thrun/p
And that paper is mentioned in the readme of the BFL (Bayesian Filtering Library) found here:
http://people.mech.kuleuven.be/~kgadeyne/software
Lastly, at one point all of us competitors were required to give our design documents to DARPA, and they put them up on their webpage here:
http://www.darpa.mil/grandchallenge05/techpapers.
BTW, I wasn't on Stanford's team, but I was on another finalist team.
Re:Fascinating program (Score:2)
[...]
> BTW, I wasn't on Stanford's team, but I was on another finalist team.
Those claims combined, it is fairly easy
Re:Fascinating program (Score:2)
I'm not stating anything false. Read his research and when it took place. The core stuff all came from CMU or Sebastian in previous efforts (like I stated). Now once at Stanford they changed how they did things entirely and wrote a ton of code to make everything play much nicer than CMU's platform. Stanly was much smarter than either of the two CMU bots.
Re:Fascinating program (Score:3, Interesting)
This sounds a little bit more like that, what I have heard. I've read, that they throw away most of the code and rewrote a large deal. E.g the classification of driveable terrain by the laser scanner was rewritten and learned. AFAIK, most of what has been published (and to what you pointed) is fairly generic stuff.
To the best of my knowledge [ira.uka.de], it has not been pu
Re:Fascinating program (Score:2)
They didn't use the near range laser scanner for the far range vision. They had a color camera they used. That was what was used to get those Red/Green video's/pictures you see on Nova's site. None of the competitors used the SICK Lasers for anything more than 25 meters, because they simply didn't work past th
Re:Seen it (Score:2, Insightful)
The main difficulty that I see, going forward, is that the laser-rangefinder systems that these robots all relied on all function by looking for obstacles and attempting to avoid them. They can spot vertical anomalie
Re:Seen it (Score:1)
Re:Seen it (Score:2)
Re:Seen it (Score:2)
Airs.. Yesterday!? (Score:2)
Re:Seen it (Score:3, Interesting)
I was really happy Stanford won the competition. The "red" team with two entries (from Carnegie Mellon?) also finished but were behind on time... the thing is though not only was Stanford's win absolute, they also did it much "smarter".
Stanford took an approach of focusing on software, to make their vehicle more smart. They gave it the course, but left it up to the vehicle to decide how fast to go and the specifics of how soon to turn, etc.
Meanwhile Carnegie Mellon took th
and the choice of vehicles... (Score:2)
And guess what vehicles they started with? Stanford built around a Volkswagen diesel (Passat? Golf?)... and Carnegie Mellon worked with.... Hummers.
Very appropriate.
It was definitely brains (Volkswagen with sophisticated software) versus brawn (Hummer with less sophisticated mapping).
Stanford 0wn3d Carnegie (Score:2, Interesting)
Also, the SuperDAD Toyota pickup looked like it had a tenth of the tech of Stanley but it was doing almost as well. If only the laser sensor hadn't detached itself from the roof.
Re:Stanford 0wn3d Carnegie (Score:1)
Who cares though. Look at the improvement from two years ago to this past race. Now we have (forgive me if I forget the details) 5-6 vehicles that finished the race? That says great things about
Re:Stanford 0wn3d Carnegie (Score:2, Funny)
Re:Stanford 0wn3d Carnegie (Score:3, Insightful)
Great show but... (Score:5, Insightful)
it is interesting just how involved the contestants are. This contest is their life. They mentioned several times in the show how many months of long workdays they spent to build and program these cars. And, then, who owns the work? Do they at least get patent recognition on some of the innovations? Some of the software they talked about was truly seriously cool stuff.
Sidenote: One hour of Nova or Frontline is like watching 5 days worth of "learning" and "discovery" shows elsewhere. It's amazing how good some of these shows are.
Re:Great show but... (Score:2, Interesting)
Maybe all these guys are geniuses and get grants to work on the stuff. Maybe university supported or something like that. Or! They make their money in half a year, and build robot cars the rest of the time.
Re:Great show but... (Score:1)
Re:Great show but... (Score:1)
"Because you are sponsored by VW, all respective research and technologies are ours, muahahah"
or vice versa, or... something like that. There would have to be a pretty clear relationship between the sponsor and the grunt, whether it be royalties or whatever
Re:Great show but... (Score:2)
I would wager that Stanford would be on the high ground if it came down to a legal battle.
Re:Great show but... (Score:3, Informative)
Mostly paid employees and purchased parts (Score:3, Informative)
The big breakthrough was Stanford's texture vision system. I was very impressed
Re:Mostly paid employees and purchased parts (Score:2, Informative)
Re:Mostly paid employees and purchased parts (Score:2)
Re:Great show but... (Score:2, Interesting)
Hmm (Score:3, Insightful)
Re:Hmm (Score:1)
Re:Hmm (Score:1)
DARPA competitions encourage innovation in technology. Technology which may well end up on the battlefield some day. Not necessarily a bad thing if it prevents the loss of life, but after viewing the aforementioned film, I've got to thinking about how improved technology may be encouraging to those who would start wars. Why We Fight goes a ways toward exploring the military-industrial complex, congress' complicity (i.e. parts of a bomber are made in all 50 states, any representative p
MOD PARENT UP- this is DAMN relevant (Score:3, Insightful)
I'm perhaps one of four people (an exaggeration, I hope) on my campus that isn't gung-ho about helping the DOD build driverless vehicles, and it's lonely at times.
Whatever m
Nice to see CMU... (Score:2)
I'm such a Nova junkie, and this was an excellent episode.
Re:Nice to see CMU... (Score:2)
I'm a geek, so I watched this twice last night. (Score:5, Interesting)
I will say, I was impressed, and surprised that I did not see an article on it at
I will say, that aside from "Stanley" winning the race on completion and time, I also believe that Stanley was the best technology. The H1lander and friend were micromanaged, and there were two vehicles that had different strategies (the tortoise and the hair) and it took almost the whole 2 hours of a team of people to map out the course and program the robots. They then added the fudge factor for human error with the fast and slow strategies.
Stanley was programmed in minutes of receiving the map, and it calculated its speed dynamically on its own. Stanley had "adaptive vision" which overlaid laser, video, and other sensory data to create a dynamic field of view of what was safe to drive through.
Now, what shocked me, was that so many teams finished this year. Nobody got past 7 or 9 miles last year, and many vehicles passed the entire 132 mile trip this year. Watching the vehicles drive was impressive. Most of the time, they appeared to be manned.
The course was not easy, by any stretch of the imagination. With the success of Stanley, I believe that this will increase the adaptive and learning capabilities in current software controlled systems. Currently, software is brute forced into trying to accommodate all possible logical conditions, which is impossible, and often just wrong.
Re:I'm a geek, so I watched this twice last night. (Score:5, Funny)
Let me guess; in the end it was a close shave and the tortoise only won by a whisker? ;)
Re:I'm a geek, so I watched this twice last night. (Score:3, Insightful)
Re:I'm a geek, so I watched this twice last night. (Score:2)
Exactly. I believe that the Stanley approach was more "real life" for what we do now, and what will be done in the future. When I go on a trip, or even go to somewhere locally where I don't know the exact location of where I go, I at least get the address and correlate it to something I do know. With the ease and availability of Go
Re:I'm a geek, so I watched this twice last night. (Score:2, Informative)
Why did they make it easier? My personal theory is the act of congress that calls for 2/3 of the armed forces to be autonomous vehicles by 2008 (or something of the sort; I'm probably wrong about the date).
By making people win (not to denigrate their achievement... debuggin
You almost never see the words (Score:2, Insightful)
Another interesting point is that it seems to me that this is the development arena for the military's new autonomously roving gun platform.
Re:You almost never see the words (Score:5, Interesting)
Re:You almost never see the words (Score:1, Offtopic)
She should have not volunteered for such a dangerous job.
I have a pet peeve with sob stories about how people enter known dangerous jobs, especially the military, where their existence is to be disposable to help the poor people of Iraq and ensure the economic welfare of the people back home.
Now, the initial invasion of Iraq is at best controversial. The continued occupation with no plan
Re:You almost never see the words (Score:2)
Overall though, a good show. Go watch it.
Re:You almost never see the words (Score:2)
But I do understand the desire to attempt to make resupply trucks autonomous. I'm not entirely sure it's really possible... but I do understand the desire.
Modern militaries consume enourmous amounts of supplies, and those supplies are big, bulky, and heavy - and more often than not, highly explosive.
The main gun round for an M1A1 tank is around 200mm in diameter, weighs ~23kg,
Re:You almost never see the words (Score:1)
At some point this may be true, but more immediately this is the development arena for the military's new autonomous supply delivery system. If you look at the people that are most often getting attacked it's the supply caravans. If we didn't have to have people in those vehicles then the loss when a caravan is attacked is much less. Just because we can make a robot t
torrent (Score:5, Informative)
My Robot (Score:2, Funny)
Too bad I've been so busy slacking this year.
Haven't I seen this before...? (Score:1, Funny)
Note to David Hasselhoff: Now's the time to re-invest into your (American) acting career!!
Re:Haven't I seen this before...? (Score:1)
Hell yeah! Now instead of having friends or parents drive us everywhere, our cars can! And secondly... if I talk to my car, it just keeps bitching about low oil pressure. I told it to stop whining and go complain to the auto-shop. But the automated system at the shop accidently ate my car. Poor thing...
Needs Serious funding (Score:1)
The robotics is taking a long time to mature.
*very very* long time
I believe the problem is that only small set of individual professors and small group of students cannot achieve huge breakthroughs in engineering.
Creating an Atom Bomb was just engineering (since sustained reaction was experimentally proved in 1933 itself).
But how many people could do it?
It took a huge set of scientists (*read - not engineers) and a huge set of engineers wo
Re:Needs Serious funding (Score:1)
While the fact that so "few" (I am not sure of the validity of that statement)are working on improving robotics, you have to realize the massive task that it is to translate reality into a machine. And then... for that machine to independently make a decision based on the generalizations of their enviroment.
And about the A-bomb... they had Einstein, they cheated.
Re:Needs Serious funding (Score:2)
In some arenas, the technology already exists. Roomba vacuums are fully functional independent robots. You can get (for a price, and with limited capability) robot lawn mowers. Some subway systems use automated trains; they're fully functional and independe
Saw this show last night (Score:1)
Agressive Robot Drivers (Score:5, Funny)
Do you scream and give it the finger?
Throw rocks at it?
Run it off the road?
Launch a homing missile at it?
Any way around it, driverless vehicles will have no rights in our future society!
Who will speak up for the robots?
Re:Agressive Robot Drivers (Score:2)
What do you do in the future when one of these is mass-produced and forgets its turn signal and cuts you off?
...
driverless vehicles will have no rights in our future society!
If your fears are realized, hopefully they'll have no lefts either.
Re:Agressive Robot Drivers (Score:2)
Finally, all my knowledge aquired over the years becomes extremely useful!
I know this one... What you do is get way out in front of it, and get out of your car. Walk over and grab the yellow-line in the middle of the road. Rip it so that you have an end, and carry your end to the nearest wall, then set it down. The robotic vehicle will follow the curved line at full speed, straight into the brick
video is worth a thousand words (Score:1)
Seminar about Stanley (Score:1, Informative)
Tell PBS Thanks! (Score:4, Informative)
Let PBS know [pbs.org] what you thought about the format, show, or anything else.
-Ian
Sensors (Score:2)
One thing that I noticed from the article is that one of the teams has problems with dust accumulating on the sensors. How would one get rid of this dust, so that you don't recieve incorrect readings?
Re:Sensors (Score:2)
Although software was a big focus, having good mechanical engineers was the key. Keeping stuff cool and free from vibration type failures was a BIG deal.
Re:Sensors (Score:2)
Re:Sensors (Score:3, Informative)
Interview with Director and Team Leader (Score:2, Interesting)
but would it work? (Score:3, Interesting)
It seems to me like 21st century warfare is a whole different animal - how hard would it be for a motivated, talented individual to figure out some simple attacks for the navigation systems on these vehicles, and get loads of sweet US munitions delivered to their doorstep? How effective would one of these vehicles be in an urban setting? How easy would it be to create a series of obstacles that would paralyze one of these vehicles?
It's amazing technology, for sure, and the Stanford and CMU teams deserve kudos. I'm just concerned that with the current rush to technological solutions and shift away from "boots on the ground", this technology will be in battle zones far too quickly.
Re:but would it work? (Score:2)
Anyways, I wonder what percentage of military trucki
Re:but would it work? (Score:2)
How hard? I would say next to impossible. Tricking the GPS system on board is impossible, while staying alive. You'd have to emulate enough GPS sats to give the vehicle improper coordinates (not only would these Satellite emulators have to work, but they would have to be synchronized properly, don't forg
Re:but would it work? (Score:1)
Re:but would it work? (Score:2, Interesting)
This is just conjecture based on a half-recollection but I don't thik it would be too difficult to attack a rel
I can see it now (Score:1)
Virus outbreak causes cars to crash, responsible for thousands of deaths.
Re:I can see it now (Score:1)
Sebastien Thrun's book (Score:3, Informative)
Der Steppenwolf (Score:1)
Re:Der Steppenwolf (Score:2)
Re:Der Steppenwolf (Score:1)
What is the issue with the m/c becoming self-aware?
For humans, it is essential to fight with others (human or otherwise) to have enough space for survival.
So, this is implanted in our genes, that we fight.. or hunt or whatever.
But in robots, unless we specifically code to attack humans, they do *not* have any reason to do that. It is not in their genes*, anyhow.
Even replicating - even that is not in their genes..
So there is no isse of them becoming self-aware.
We shouldnt try to see in
Airs on PBS Tuesday, March 28 (Score:2)
Seriously though, I'd been hoping someone would be putting together something like this (though I'd been expecting it form Discovery or TLC - yay for public television). Fortunately, it is available online [pbs.org] for those of us who missed it.
Re:Airs on PBS Tuesday, March 28 (Score:1)
of course (Score:2)
"The wars of the future will not be fought on the battlefield or at sea. They will be fought in space, or possibly on top of a very tall mountain. In either case, most of the actual fighting will be done by small robots. And as you go forth today remember always your duty is clear: To build and maintain those robots. Thank you."
-- Military school Commandant's graduation address, "The Secret War of
Lisa Simpson"
Very good NOVA documentary (Score:2)
It wasn't as much the fact that Stanley won the race as how Stanley won the race and the differing approaches of the builders that made it interesting.
Unfortunately, it was not in HD. It was widescreen low definition. T
Robot Wars (Score:3, Funny)
Some Random Thoughts about This (Score:2)
There was nothing quite like seeing, for the first time in my life,
On the weekend of the race... (Score:1)
Details in the program (Score:3, Insightful)
-- The teams get the GPS waypoints a few hours before the race. The waypoints are purposefully vague, so the robots have the choice of driving off a cliff (or into one) while still being within GPS parameters. This is supposed to prevent the race from reducing to "Who can follow GPS the best?" The Red Team had a group of what looked like 20 or 30 people who immediately sat down with the waypoints mapped out on satellite imagery, going through and adding waypoints of their own and adding speed commands for their robots. This seems to me to be a big violation of the spirit of the competition.
-- The Red Team had two entries, which they programmed differently: one more aggressive, the other more conservative (on speed). The faster robot, Highlander, was pulling away from Stanley for the first part of the race, until some unknown issue starting causing problems. Nova didn't say what was wrong, but it looked literally like Highlander was slipping out of gear and rolling back down hills. It _might_ have been doing it on purpose, i.e. a software glitch, but it didn't look that way.
-- One of the Red Team's entries completed the last portion (the hardest portion) of the course with its main sensor non-functional -- it was stuck pointed 90 degrees to the side. This argues even more strongly that the Red Team's vehicles weren't doing much route-finding and were pretty much just following GPS waypoints.
The conclusion I draw from this is that we are still a long way from the DOD's goal of autonomous transport vehicles. In a combat situation, transports need to be able to avoid obstacles put in their way _by the enemy_. The only time during this challenge that the vehicles did anything like this was during the initial trials before the race, and that was very limited. The actual race course was hard -- off-road, dirt, narrow, slippery -- but it didn't have tank traps painted the same color as the dirt they rest on. It didn't have razor-wire barricades, forcing the cars to figure out a route through the bushes around them.
I'm confident that if I had been on the course fifteen minutes before the cars showed up, I could have stalled or disabled all of them. Pile a bunch of bushes across the road and all of them would have stopped. During the trials and race, none of them demonstrated the ability to work around such a very limited obstacle.
All of this is not to minimize what was accomplished. But we're a long way from sitting back sipping champagne while robots do the dirty work of war.
Re:Details in the program (Score:2)
Your point is certainly valid, but also consider that a robot with backup systems that rely on totally different strategies will have a better chance of success when (not "if") something
Re:Details in the program (Score:2)
Re:Details in the program (Score:1)
Agreed that in a military situation you're likely to have extensive pre-planning of the route, but you're not goin
David vs Goliath (Score:1)
My Friend at CMU (Score:1)
Also, both CMU robotos ran some form of linux with a all of the hardware donated by Intel. I believe they had a 2TB RAI
Re:My Friend at CMU (Score:1)
Lessons from the winners/losers (Score:2)
Things that Nova left out (Score:2)
* It mentioned the Gray team being a dark horse, but in reality, they took only about half an hour longer than Stanley. If anything, it was probably even more of a newcomer than Stanley. CMU has been in the robot driving business for a long time (they had neural-net based self-driving vehicles since the early 90's), so for this unknown team to finish so close
I was on Red Team year 1 (Score:2)
Re:I saw this last night too (Score:2)