Slashdot Mirror
Still More on the DARPA Grand Challenge
The SF Chronicle has an in-depth story on the DARPA Grand Challenge, with emphasis on the several teams from the San Francisco area. The three teams covered are using a pickup truck, a six-wheeled all-terrain vehicle, and a self-balancing motorcycle...
Which of these three doesn't belong?
a pickup truck,
a six-wheeled all-terrain vehicle,
and a self-balancing motorcycle
Esoteric reference.
no VW van?
*insert joke about how blue screen of death will be literal here*
Do you think I could enter with a beat up buick and a brick on the accelerator?
01100010 01101001 01110100 01100101 00100000 01101101 01100101
so if that pickup truck had come from texas instead of SF would it have it's own computer controlled fully autonomous gun turret too?
Which one will be least elaborite. I mean, yes you need some complexity, but the less things that can go wrong the better. I like the sound of a self balancing motorcycle myself, but I bet the simplest will have a better chance at winning.
http://www.beyourowneviloverlord.tk
http://www.frozenchickenthrowing.tk
http://www.killercamel.tk
I wonder how much do NASA and DARPA collaborate. Much of the technology used to create the mars rovers seems like they would be useful for this challenge.
From the article: "The biggest hurdle has been making vehicles see obstacles and react to them"
The mars rovers use a pair of cameras to build a 3d model when it decides its path. Put this system into a 4x4, give it a small cluster for computation, and it should work well enough to make it across the desert, I would think.
Leave the vechicle unlocked in a bad part of Barstow with the keys, a pile of Vegas casino chips and case of booze in the car.
If no one watches the car, I predict at least 50% chance that it with disappear from Barstow and reappear in Vegas. (Or in a ditch on the way there.)
Cannonball run. For robots. Cool.
EEPD Profive 700 Mhz Pentium III PC -104 onboard computer running Real Time Linux
http://robots.mit.edu/projects/darpa/ (with videos)
CC.
TaijiQuan (Huang, 5 loosenings)
From the article:
DARPA won't disclose the exact route of the Grand Challenge until two hours before the race March 13; it has promised a rigorous route that will include rocks, gullies and streams.
Some of the world's best dirtbike riders wouldn't be able to easily cross stuff your average Land Rover or Land Cruiser would laugh at. I think the team with [what looks like] the 6-wheeled ATV stands the best chance, at least from a vehicle-choice perspective. Those things are amazing in terms of what they can cross- some of them even float and can ford -rivers- using the tires as paddlewheels.
Description of the pickup truck entry:Two tons of steel rolled forward and made a jerky left out of a parking lot in Morgan Hill. It gained speed and settled into a lane. It followed a curve to an intersection. It stopped. Then it turned right and continued down the road.
Probably stands a better chance(and has better fuel economy than the 6-wheeler- though a MUCH higher center of gravity), but taking a trip through suburbia hardly qualifies as suitable testing grounds for what DARPA has described...and depending on the truck, it might not stand up to the abuse. A jeep(or, a Land Cruiser, or a Land Rover) would have been a much better idea than a pickup truck, which really isn't designed for off-roading.
Even the guys who do insane things with their jeeps and whatnot come fully equipped. Air suspensions. Winches. Huge tanks of air or compressors to re-seat the giant tires(did I mention giant tires? :-)
I can also think of a lot better things to spend money on than that giant LCD display they put in the truck's passenger side; that thing has got to be what, 21"? The money would have been much better spent on the truck itself. It's all fun and games until that rock takes out your transfer case and your truck's transmission rips itself to pieces.
Please help metamoderate.
The article mentions things like your rental car showing up from the airport all by itself, which I admit would be quite cool. It also makes me wonder about the first collision between two autonomous vehicles on a public highway. Would the programmers get the tickets? Lots more interesting questions to be answered when these things start selling...
Kindness is the language which the deaf can hear and the blind can see. - Mark Twain
I just read an article in Scientific American about this. I'm not into robotics myself, but I'm curious: how much of the difficulty is due to the time it takes to process the input data (from cameras, lasers, etc.)? how much is from the necessary ruggedness of the components? how much difficulty comes from lack of funding for and access to top-of-the-line components? I'm also curious to see what DARPA plans to do with the winning vehicle, if there is a winner. Will they pay for, and then take, any vehicle that is innovative (for example, the motorcycle that can stand on its own)?. Kudos to DARPA for their clever method of conducting research--instead of tying funds up in someone's brainchild, they are allowing a lot of different ideas to proliferate.
War does not determine who is right, war determines who is left.
As a lover of the desert and card carrying member of Earth First, I thought I'd point out another contest that happens to be occurring on the same day. Its called "Bag the Unmanned Vehicle". Contestants compete to disable unmanned vehicles trashing desert flora and fauna for fun and prizes.
=-+
Just because DARPA is collaborating with NASA, don't get your hopes up if you're thinking about some 'geekcool' super-Star-Trek-beam-me-up-scotty rocket their buddy. DARPA is strictly defense, and anything they can get to the benefit of a defense project is worth gold.
If DARPA is doing something with NASA, it will likely use this for the killing fields nothing more nothing less.MoFscker
The difficulty with autonomous land vehicles is using sensor data to figure out what the environment is like, and using that information to plan what to do next. Both are AI problems, not hardware problems (though, certainly, clever sensor hardware and lots of computer power helps).
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
Of course the simple solution would be to give a monkey a quad bike. But don't give him a full-blown road vehicle, or he will turn it into a V8 intercepter and conquer the post-apocolyptic wasteland...
Even those who arrange and design shrubberies are under considerable economic stress at this period in history.
Yes.
Each vehicle is followed by a manned one. Specifically, one of the team members and a contest official.
The team member has a "big read button" - which is a mandatory safety device - that is the vehicle is in danger of or actually goes off course can be used to shut it down.
Then you can get disqualified for it, upon the disgression of the cheif judge.
Check out the latest copy of the rules
=Smidge=
Vehicles must cross 200 miles of rugged terrain between Barstow and Las Vegas
Sweet Jesus! That's bat country. I suppose the poor bastards will figure that out soon enough.
?
Automobiles should be restricted to 4 mph, and preceded by a man carrying a red flag (an presumably singing the Internationale for good measure).
Machines such as the Spinning Jenny will destroy our way of life.
I salute you, Ned Ludd, for your foresight and insight into the human condition.
oh brave new world, that has such people in it!
For that matter, I read recently about a study done by a couple of psychologists in which they described to schoolchildren about accidents in which seatbelts saved lives, and then about ones in which they caused injuries. The children, after hearing about the first, said, ``oh, then you should wear your seatbelt always.'' After hearing about the latter, they said the opposite. When asked repeatedly by the researcher, ``so, when should you wear it and when shouldn't you?'', one subject replied, ``well, I guess you should wear the seatbelt half the time.''
People aren't rational; one theory is that we interpret probabilities by ``representativeness'', a heuristic in which the situation being judged is compared to a similar situation thought to be probable or frequently heard of. So the more people hear about robotic-automobile-caused deaths (which would certainly be more publicised than the same old same old), they'd assume such vehicles are less-safe than traditional cars.
Many people judge the risk of very rare, unlikely deaths (from rare diseases, freak accidents, and the like) to be far higher than they are, while they judge the risk of death from things like car accidents and other more normal causes to be significantly lower than it really is. This is because they hear about far more of the freak accident deaths--precisely because they are freak accidents--than the ordinary, normal deaths.
Of course, just as with airbags, after the breaking-in period, I think people would probably get used to it. And the economic demand, if great enough, would be enough incentive to let them on the road.
200 miles in 10 hours equals an average of 20mph.
A blog about stuff.
Sadly, we (Team Overbot) aren't going to be ready in time. We lost five members in January. Two got better jobs, and two were Stanford students who needed to get their grades back up. This left us with too few people to finish in time. We have all the hardware, and most of the software, Most of it is working, but it hasn't been integrated and tested. We'll finish the vehicle, and we'll have some public demos at some point, even if we're not at the Grand Challenge.
It's up to DARPA whether anyone wins this year. They're going to provide 5000 GPS waypoints, and if you can drive the route described by connecting the dots, somebody will probably win. If the vehicle has to find its own gully crossing, it's unlikely that anyone will win, unless somebody figured out, by hand, in advance, where the crossing is. It's all up to DARPA. As one of the DARPA people put it, "This is turning into a breadcrumb following exercise". If somebody wins by connecting the dots, this whole thing was a waste of time.
Several teams are using aerial photographs and manual planning. The general route leaked weeks ago, and it's since been oveflown by Airborne 1 in San Diego. High-resolution photos and depth maps from LIDAR scans have been obtained. Still, you won't see a fence in those depth maps. The emphasis on preplanning surprised us. The whole point of the Grand Challenge was originally that preplanning was made impossible by the large area to be covered and the release of the waypoints only two hours before the race. That all changed when the route leaked.
Nobody seems to be deploying anything new in the sensor area. Everybody with a laser rangefinder that we know of is using an off-the-shelf line scanner. Nobody has a true 3D scanner, although several teams have line scanners on tilt heads. It's quite possible to build a true 3D LIDAR depth measurement system. But it's hard to make money doing it, as the five companies that exited the field learned the hard way.
We hear talk of new vision algorithms, but no details yet. Stereo vision doesn't work well on dirt or sand; there aren't enough edges for the stereo algorithms to register the images properly. Optical flow doesn't work well for the same reason. If somebody can do good stereo from motion in this demanding environment, that will be an achievement.
Still, the Grand Challenge has done quite a bit to get autonomous vehicle work moving again. Just getting CMU off the dime (DARPA's real intent, we hear) was worth the whole effort.
If DARPA does this every year for the next decade, with a tougher course every time somebody wins, we will have battlefield robotics that works within ten years.
I was the lead engineer of the Grand Challenge team 'R Junk Works'. Our paper was submitted and approved by DARPA for the Grand Challenge. They then put us in the 'Partially Acceptable' category just after approval. This was no big deal for us as we already had our prototype vehicle built and were testing it in October of 2003 before we submitted our Technical Paper. Their 'Site Visit' seemed like just another hoop to jump through before being in the race. There are only three people on our team, and we all have worked in one form or another for the 'Lockheed Martin Skunk Works', so our little group we called the spin offs - or the 'R Junk Works'. We are also located in Palmdale California. In retrospect, perhaps it was our team name that gave DARPA hesitation. Heaven forbid that only three guys in a garage in Palmdale with a name like ours win their Grand Challenge! Anyway, the four DARPA site visitors/inspectors arrived at my garage on the 5th of December to inspect our progress. They road around in our test vehicle that had: Integrated DGPS, LN200INU and V4L2 Vision systems running under LINUX Fedora Core 1.0 all installed and functional in my personal 1998 Toyota Tacoma pickup truck. They let it slip that this was a 'first for them' to be riding around in one of the contestants vehicles. They road around in the drivers seat around the desert next to my house here in Palmdale along a pre programmed course that took them down dirt desert roads and washes here in the Mojave Desert - only a few mile away from where they are going to have the actual race. Members of this inspection team jump in front of our vehicle as it was traversing the pre programmed course and watched how it avoided them by driving around them and continuing down the course. They watched with amazement as our vehicle raced along in excess of 35 mph across the rutty Mojave Desert roads. Almost everything worked perfectly for the demo except our main vision system camera had been damaged the day before and we were using our backup camera that was having intermittent problems, but did not take away from the totality of the demo. It was probably one of the best demos I have ever given in my entire professional life. If I could summarize their attitude of the demonstration, it would be that they were amazed, enthused and excited over our participation in the Grand Challenge. They also let it somewhat slip that we were the farthest along team that they had seen as yet! I tend to think that the inspection team was 'On our side' as possible contestants. After the demo, we assumed that it was inevitable that we would be selected for one of the remaining six contestant slots left. This was far from the case. They called us on December 17th (my birthday) and told us that we were not selected to participate. One of the inspection team members said: 'After a much heated discussion amongst the DARPA Program Director and our inspection team, I have been told to tell you that you were not selected to participate in the Grand Challenge.' The transparent reason they gave us was that our team did not, as yet, have an actual 'race vehicle'. A very trivial problem for us when it comes right down to it. This was by their very own undoing, as once our vehicle sponsors got wind that we were not in the 'Totally Acceptable' list; they backed out and were waiting for our team to be on that list before donating our actual race vehicle. We even supplied statements of sponsorship from that sponsor = they obviously did not read them = OR = perhaps there was another incentive. After talking to a guy called 'Dan' who is the editor of a national magazine and good friend of mine, he also went to the 'Kick Off' for the Grand Challenge that DARPA had in LA last year. He was able to 'Liquor Up' one of the DARPA legal reps, and SHE intimated that the reason they were holding the Grand Challenge was to put the fear of god into their current contractor and show that they could go else ware for technical projects. She also said that it had already achieved this goal and that even if the
If I'd paid more attention to what Berkeley was doing with ACK delays, TCP would work better in that area today. Both algorithms went in around the same time, and they don't play well together.