Slashdot Mirror

The Grand Challenge, I thought, was designed with a detrimentally macho mindset, with a needlessly high ratio of financial risk to scientific output. It sent many expensive cars through long stretches of very uniform-looking but occasionally high-risk (cliffs!) terrain. The cars had a road to follow. It was ok to hire an army of undergraduates to hand-design a path through the terrain (see the funny scenes involving CMU from chapter 6 of this online NOVA episode about the Grand Challenge).

Maybe they've learned a few lessons, because the DARPA LAGR competition does everything in almost the opposite manner:

• Focus on software: The teams all use the exact same hardware. We each get a robot for testing, but for the competition we just hand off our software to the government organizers, who load it into their robot and let the robot run the course.
• Focus on the difficult problems: Current robotic localization and mapping is heavily reliant on laser range scanners, which have a limited range (~30 feet) compared to vision, and therefore is unsuitable for long-range path planning. Instead of letting scientists wait for laser scanners to get incrementally better, this competition forces its entrants to adopt a vision-based solution, like people. As one organizer put it, we didn't get to the moon by incrementally improving the airplane.
• A sane funding structure: A pool of 10 or so initial teams start out with funding, with each successive competition (with increasing demands) weeding out more and more teams. I've heard that other DARPA robot competitions happening in parallel have adopted similar funding systems. Contrast this with the grand challenge teams, who didn't get any money until the end (CMU notably sank $3 million into their efforts to win the $2 million prize).

Do a google search on Sabastian Thrun, he was the team lead for Stanford, and formally at CMU (what a non-coincidence). Most of the software they used on Stanly (Stanford's bot) was either written by Sebastian in his former research or taken from experience gained on CMU's team the previous year. The ladar mapping he used, I know I saw on some former page of his that had all the gory algorithm details. It might just take a little bit of searching. He also has a c library out there somewhere that does a lot of this stuff, but I can't seem to find it now.

One paper that's of interest might be here: http://www.cs.cmu.edu/afs/cs.cmu.edu/user/thrun/pu blic_html/papers/thrun.ces-tr.html (sorry, no linky, writing in a hurry)

And that paper is mentioned in the readme of the BFL (Bayesian Filtering Library) found here:
http://people.mech.kuleuven.be/~kgadeyne/software/ bfl-trunk/

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.h tml

BTW, I wasn't on Stanford's team, but I was on another finalist team.

The project is part of a larger Program at DARPA nicknamed DARWARS that is building a series of game-based interactive training for the military.

DARWARS has been working with DoD contractors and game companies (and are interested in working with more game companies).

One of the games produced was BBN's Ambush!. Several other games are listed in the http://www.dodgamecommunity.com/ web site funded by DARWARS.

An official description of the program can be found here.

The original program was to produce a MMORPG for training in a persistent continuous war, but it has become more of a repository of games for military training.

Why is the company that is at the forefront of intelligent vehicles collaborating with the U.S. army!?

Umm... the US Army is part of the Department of Defense , and DARPA stands for Defense Advanced Research Projects Agency.

IBM does admit it. They thank DARPA
and other DoD groups for their funding
in their research papers. Most of the
current funding for speech-related research
in DoD is run through the GALE project:

        Global Autonomous Language Exploitation

        http://www.darpa.mil/ipto/programs/gale/index.htm

Salim Roukous of IBM, whom they quote in the
article, is the main player from the IBM
side and IBM is one of the main players in
this project. They were formerly a primary
player in TIDES:

Translingual Information Detection, Extraction and Summarization
http://www.darpa.mil/ipto/programs/tides/index.htm

In fact, that site has the last link I can find on
DARPA's site about TIA (Total Information Awareness),
which is a program formerly run by ex-Admiral Poindexter
(Iran-Contr fame) and shut down by an act of congress
(and erased from DARPA's site as if it never happened):

http://www.darpa.mil/ipto/programs/tides/accomplis hments.htm

These are not classified projects. You can
read about most of the techniques in the proceedings
of conferences such as ACL, ICSLP and Eurospeech.

IBM does admit it. They thank DARPA
and other DoD groups for their funding
in their research papers. Most of the
current funding for speech-related research
in DoD is run through the GALE project:

        Global Autonomous Language Exploitation

        http://www.darpa.mil/ipto/programs/gale/index.htm

Salim Roukous of IBM, whom they quote in the
article, is the main player from the IBM
side and IBM is one of the main players in
this project. They were formerly a primary
player in TIDES:

Translingual Information Detection, Extraction and Summarization
http://www.darpa.mil/ipto/programs/tides/index.htm

In fact, that site has the last link I can find on
DARPA's site about TIA (Total Information Awareness),
which is a program formerly run by ex-Admiral Poindexter
(Iran-Contr fame) and shut down by an act of congress
(and erased from DARPA's site as if it never happened):

http://www.darpa.mil/ipto/programs/tides/accomplis hments.htm

These are not classified projects. You can
read about most of the techniques in the proceedings
of conferences such as ACL, ICSLP and Eurospeech.

IBM does admit it. They thank DARPA
and other DoD groups for their funding
in their research papers. Most of the
current funding for speech-related research
in DoD is run through the GALE project:

        Global Autonomous Language Exploitation

        http://www.darpa.mil/ipto/programs/gale/index.htm

Salim Roukous of IBM, whom they quote in the
article, is the main player from the IBM
side and IBM is one of the main players in
this project. They were formerly a primary
player in TIDES:

Translingual Information Detection, Extraction and Summarization
http://www.darpa.mil/ipto/programs/tides/index.htm

In fact, that site has the last link I can find on
DARPA's site about TIA (Total Information Awareness),
which is a program formerly run by ex-Admiral Poindexter
(Iran-Contr fame) and shut down by an act of congress
(and erased from DARPA's site as if it never happened):

http://www.darpa.mil/ipto/programs/tides/accomplis hments.htm

These are not classified projects. You can
read about most of the techniques in the proceedings
of conferences such as ACL, ICSLP and Eurospeech.

When I decry public funding of science, I'm blasted because people say that the free market won't pay for certain research. Now I see a more evil side of it -- and I fear that we'll see more investigations like this if I'm right. What can we do to combat humanity's deep need for self preservation in a scientist having the same human drives, especially when it is funded straight out of our pocket involuntarily?

There can be two sides to this issue.

1. If the research is funded with government money, it can be influenced by politics.
2. If the research is funded with private money, it can be influenced by its investors.

Think of it like a global warming research sponsored by a congressman who is lobbied by an oil company vs a TCO of Windows vs Linux research sponsored by Microsoft.

Both could have potential bias and complications.

Personally, I believe both private and public research can be beneficial. Take DARPA for example. I for one believe DARPA is the shining example of public research gone right. It is backed by public money, but often uses the private sector as a major part of its research. Take the recent Grand Challenge for example.

So I think there is a place for public funding at least to get the ground work. After all, the Manhattan and Apollo Project were publicly funded.

However, if you believe government funded projects are a waste of your tax money, then you can do what I do... Donate to a private non-profit research group that is tax deductible. I realized if I donate enough money to either Wikipedia or the Singularity Institute I could just write off all my taxes next year. Even though I don't get more money than I would have not donating, it means the IRS will have to give me a larger refund, hence putting my money where I want it to go and not where a congressman does.

http://www.darpa.mil/tto/programs/walrus.htm There is definitely military potential, since DARPA already has a project based on the same concepts. "The Walrus program will develop and evaluate a very large airlift vehicle concept that is designed to control lift in all stages of air or ground operations including off-loading of payload without taking onboard ballast other than air. Unlike earlier generation airships it will generate lift through a combination of aerodynamics, thrust vectoring and gas buoyancy generation and management and for much of the time, it will fly heavier than air."

I thought you can get negative refraction, when an electromagnetic wave passes through a "Metamaterial" i.e. One with Negative Permittivity and Permeability.

(for instnace, in a dispersive plasma cloud)

It's also worth noting that the Stanford vehicle was still manually drivable and street legal (according to the Technical papers). The CMU vehicle could also be driven manually, but the vehicle had been heavily modified to accept the sensors and stabilize them and it was evident from it's appearance. The technical papers are quite an interesting read, even if they don't go into much detail.

See page 11 of their DARPA Grand Challenge entry document (http://www.darpa.mil/grandchallenge05/TechPapers/ Stanford.pdf). It runs Linux, the software is written in C/C++, and it uses IPC for communication between the various components.

(This was pointed out by an AC elsewhere in this thread),

True. At least, that's what's specified in the official rules.

In the Grand Challenge, cars didn't race against one another to try to be the first across the line. They raced to try to complete the course in the shortest elapsed time .

According to the Darpa web site, Stanford won the race by finishing with an elapsed time of 6 hours and 53 minutes. They could still have won if they crossed the finish line after the CMU vehicle, as long as their elapsed time was still shorter.

CMU's Sandstorm finished in 7 hours and 4 minutes.
CMU's H1ghlander finished in 7 hours and 14 minutes.

The CMU bashing here (and subtley embedded in the wired article--everybody loves an underdog) is not really valid.

According to The Grand Challenge Tracking Site:

Stanley's official time was 6:53 and CMU's was 7:04 minutes.

I don't think that ridiculing CMU as having a "poor strategy" for doing something in an additional 11 minutes that was impossible for the entire robotics industry just a year ago is very. . . wise.

Personally, I'm overjoyed that Stanley won it. I think he's an excellent system and that Stanford deserves the praise. (Besides, those b*stards at CMU didn't let me in for my undergrad)--but making fun of their 2004 'strategy' (when they went further than any other team) and their 2005 results (when they were a scant 11 minutes behind the leader, and were 2 of only 5 teams to have a 'bot cross the finish line) seems silly to me.

And for the people wondering: Stanley is rumoured to have run linux, though last I heard the team hadn't confirmed it. In fact, most of the qualifiers for the race were running at least one linux machine.

It does: http://www.darpa.mil/grandchallenge05/TechPapers/S tanford.pdf

This reminds me of DARPA's self-healing minefield, where mines communicate with each other and 'hop' to fill any gaps in their net. DARPA's page here: http://www.darpa.mil/ato/programs/SHM/

These tasks are probably much more complex than the Ansari X prize... which rewarded 40 times this much. Offering $250K is insane. Stupid. Insanely Stupid.

Yes but the issue is all the performance doesn't matter if your researchers aren't using it to solve problems. See http://www.darpa.mil/ipto/programs/hpcs/ for more info on this. This is the big HPC push that IBM, Cray, and SUN are participating in. Also a company that I think is kinda cool http://www.orionmulti.com/ is working on a very common use of HPC tools by non-computer people. They are very focused on providing easy to use ultra low maintainance computational tools primarially for the bio-informatics community. One of the founders of the company worked an LANL on green-desitiny (or something like that) which was designed to be a low power low maintainance super computing resource at LANL. After all that, the short answer is yes performance is important but there is a lot of work and interest in making sure that this performance can actually be used by the people that are actually solving problems. Mark

Uh, ARPANet, as in Advanced Research Products Agency Network, was DOD funded for many years before it was spun off into the non-military branches of government.

Does anyone know how exactly they scored the Grand Challenge? I was watching the leaderboard the whole day as I was coding, and one of the CMU vehicles was the first to finish as far as I can recall...furthermore, the spread BETWEEN the top three vehicles (CMU, Stanford, CMU) seemed to vary from time to time. I have my doubts as to the validity of the data since there were also a couple glitches during the race where all of a sudden a bunch of vehicles' mileage and stuff were kicked back by a good amount. -------------- I also monitored the GPS based map-tracker the entire Race. Stanley was on H1landers bumper on the upper-portion of the course (west of the 15 fwy). On the bottom part of the course (before Beer Bottle Pass), Stanley made the pass. I uploaded a few screen shots: http://darpagc.textamerica.com/ Stanley was the 1st physical finisher. I was involved with robotics/vision research as a grad student & I am also involved with OffRoad Racing (see http://www.jumplive.com./ I can say this years route seemed to be way easier..mostly just dirt roads. No whoop sections. Last year, it was in the Barstow area, where it can be pretty rough (all the top OffRoad Racing teams test there). Here's a sample of some REAL OffRoad Racing (there was "pausing" for the DARPA GC '05, if a 2nd vehicle got on the bumper of the 1st vehicle): http://web.archive.org/web/20000925132425/www.seel eyracing.com/barstow/end/vgeorgel4fps3.mov It was taken from the area known as "Mile of Danger". Could the 'bots read a "fork" in the desert? Someone (who was part of a team for awhile) claims the DARPA GC was a sham. https://dtsn.darpa.mil/grandc/forum/topic.asp?topi c_id=1657&forum_id=30&Topic_Title=What+a+scam%2C+r ace+wa+sno+challenge+at+all!&forum_title=Grand+Cha llenge+Event&M=False&S= It wasn't a race, more like a rally. I did a LiveWebCast of the SCORE Primm 300 offroad race a month ago: http://primm300.textamerica.com/ These vehicles took an a far rougher/tougher course, even the stock VW Baja Bugs: http://www.dezertrangers.com/iB_html/uploads/post- 1-87711-class11s.jpg I bet one of these bugs coulda placed high (or even won) the DARPA GC. You can see this thread for more pics of the Primm 300: http://www.dezertrangers.com/cgi-bin/ib/ikonboard. cgi?s=e3be4804ea8ddd2f033271bbc686d107;act=ST;f=1; t=27465;st=0 You can see some silty areas. Stanley woulda had a problem, the CMU Hummers probably not.

Another helpful website for anyone interested in what the DOD intends to do with grand challenge technology:

http://www.darpa.mil/grandchallenge/overview.html

Note the complete lack of any mention of 'non lethal' and the use of such interesting catch phrases as "autonomous ground vehicles that will help save American lives on the battlefield "

For far better info than the anemic (and completely flash based) gc.org site:

http://www.darpa.mil/grandchallenge/discussion.htm l -- DARPA's GC message boards
http://www.tgdaily.com/2005/10/08/darpagrandchalle nge2005/ -- Was updated throughout the actual event. Best coverage I've seen yet.
http://www.popsci.com/popsci/darpachallenge/ -- Popular Science's rather disorganized site

I'm still looking for "highlight" video myself... or pretty much any non-bland video (seeing them cross the finish line is nifty and all, but that was not a challenging part of the race). I particularly want video of Alice trying to take out some reporters!

There was a 10 mile error in the data from the start. Darpa corrected it which made it jump back as you noted. It is believed that the current data is more accurate read the thread here: https://dtsn.darpa.mil/grandc/forum/topic.asp?topi c_id=1652&forum_id=30&Topic_Title=Post+Time!&forum _title=Grand+Challenge+Event&M=False&S=

hint: http://www.darpa.mil/

Video coverage here (there's a whole bunch. The overview, stanley and ghostrider ones are awesome!):

http://www.cartv.com.nyud.net:8090/content/researc h/channels/index.cfm/channel/cartv_video/action/sh owvideo/vid/e_0145/vcat/Event/
NQE final paper:

http://www.darpa.mil/grandchallenge/NQEfinal1.pdf

And more announcements can be found on:

http://www.grandchallenge.org/

Also, a good summary of things that have been happening can be found in the discussion forum:

https://dtsn.darpa.mil/grandc/forum/topic.asp?topi c_id=1636&forum_id=30&Topic_Title=NQE&forum_title= Grand+Challenge+Event&M=False&S=

=====
A post by Espina reads:
Hi! ...stopped by the NQE last week and this whole Tuesday and I must say that all the work accomplished on all the AGVs was very impressive. ...for those who couldn't be there the following bots all had runs in the morning session: "Mojavaton, DAD, CIMAR, Insight Racing, Golem Group, ENSCO, Princeton, MonsterMoto, Team Jefferson, UCF,, AION, Cajunbot, Banzai, Gray Team, Mitre Meteorites, Virginia Tech Grand Challenge Team, Austin Robot, Desert Buckeyes." All had full runs except five. Majavaton and Insight Racing which both collided with a vehicle/obstacle within 100 yards of the finish line. Aion decided to skip the course and circle back directly to the finish line but a K- rail barrier refused to co-operate. The UCF bot went walkabout on the back 40 towards the NASCAR track and Austin Robotics got sulky in the first loop when the crowd left for lunch during its run. MonsterMoto was given a restart because a chase truck encroached on the route near the start. ...according to some team members from Ensco, the afternoon session was a chance for the teams "on the cusp" to improve their standings. Austin Robotics, CajunBot, VT, Team Banzai, Mojavaton, the Mitre Group, and the Gray Team all had additional runs. ...Mojavaton, VT, Mitre (had two) and the Gray Team all had full runs. The Gray Team had two runs but was unable to to get GPS back after the tunnel on the first run so they made a few adjustments and had a stellar 2nd run. It seemed like a time/constelation problem. CajunBot made it to the last Obstacle/vehicle to the chagrin of the crowd. Team Banzai froze contemplating a witch's hat on a downhill transition at the end of the first loop and Austin Robotics lost GPS (and its way) after the tunnel... ...after that the best of the rest ran (Autonosys, Blue Team, Overbot, Indiana Robotic NAV, BJB Engineering, Team Juggernaut, Autonomous Vehicle Systems, Team Tormenta, Indy Robot Racing, Terra Engineering, PVHA Road Warriors, CyberRider, AI Motorvators, Team Underdawg. )with most of the teams wiping out the first barrier, and/or re-arranging the hay bales at the tunnel entrance, colliding with the tunnel entrance and losing GPS after the tunnel. However, IT, from AI motorvaters had a full run on the shortened RDDF and TerraHawk made it thru most of the hard parts. Overbot ran very thoroughly and cautiously but froze on the downhill transition. ...if any of this information is incorrect please feel free to fix...I could be suffering the effects of sunstroke... ...anyhow, good luck all and I admire dedication of all of the teams on completing an AGV. ...see y'all in Primm, Espina

Video coverage here (there's a whole bunch. The overview, stanley and ghostrider ones are awesome!):

http://www.cartv.com.nyud.net:8090/content/researc h/channels/index.cfm/channel/cartv_video/action/sh owvideo/vid/e_0145/vcat/Event/
NQE final paper:

http://www.darpa.mil/grandchallenge/NQEfinal1.pdf

And more announcements can be found on:

http://www.grandchallenge.org/

Also, a good summary of things that have been happening can be found in the discussion forum:

https://dtsn.darpa.mil/grandc/forum/topic.asp?topi c_id=1636&forum_id=30&Topic_Title=NQE&forum_title= Grand+Challenge+Event&M=False&S=

=====
A post by Espina reads:
Hi! ...stopped by the NQE last week and this whole Tuesday and I must say that all the work accomplished on all the AGVs was very impressive. ...for those who couldn't be there the following bots all had runs in the morning session: "Mojavaton, DAD, CIMAR, Insight Racing, Golem Group, ENSCO, Princeton, MonsterMoto, Team Jefferson, UCF,, AION, Cajunbot, Banzai, Gray Team, Mitre Meteorites, Virginia Tech Grand Challenge Team, Austin Robot, Desert Buckeyes." All had full runs except five. Majavaton and Insight Racing which both collided with a vehicle/obstacle within 100 yards of the finish line. Aion decided to skip the course and circle back directly to the finish line but a K- rail barrier refused to co-operate. The UCF bot went walkabout on the back 40 towards the NASCAR track and Austin Robotics got sulky in the first loop when the crowd left for lunch during its run. MonsterMoto was given a restart because a chase truck encroached on the route near the start. ...according to some team members from Ensco, the afternoon session was a chance for the teams "on the cusp" to improve their standings. Austin Robotics, CajunBot, VT, Team Banzai, Mojavaton, the Mitre Group, and the Gray Team all had additional runs. ...Mojavaton, VT, Mitre (had two) and the Gray Team all had full runs. The Gray Team had two runs but was unable to to get GPS back after the tunnel on the first run so they made a few adjustments and had a stellar 2nd run. It seemed like a time/constelation problem. CajunBot made it to the last Obstacle/vehicle to the chagrin of the crowd. Team Banzai froze contemplating a witch's hat on a downhill transition at the end of the first loop and Austin Robotics lost GPS (and its way) after the tunnel... ...after that the best of the rest ran (Autonosys, Blue Team, Overbot, Indiana Robotic NAV, BJB Engineering, Team Juggernaut, Autonomous Vehicle Systems, Team Tormenta, Indy Robot Racing, Terra Engineering, PVHA Road Warriors, CyberRider, AI Motorvators, Team Underdawg. )with most of the teams wiping out the first barrier, and/or re-arranging the hay bales at the tunnel entrance, colliding with the tunnel entrance and losing GPS after the tunnel. However, IT, from AI motorvaters had a full run on the shortened RDDF and TerraHawk made it thru most of the hard parts. Overbot ran very thoroughly and cautiously but froze on the downhill transition. ...if any of this information is incorrect please feel free to fix...I could be suffering the effects of sunstroke... ...anyhow, good luck all and I admire dedication of all of the teams on completing an AGV. ...see y'all in Primm, Espina

It is true that we keep hearing about combat robots, but the simple truth is that it is incredible difficult to reproduce what boots on the ground can do. Let's face it, even when stumbling drunk, human beings are better at negotiating difficult terrain (such as a combat zone) than even the most sophistocated robots. Just consider the amount of brainpower and resources are being poured into the DARPA Grand Challenge. How many years, millions of dollars, and engineers has Honda spent just to get the Asimo to walk around on its own, on a perfectly flat floor or up a preprogrammed flight of stairs? The difficulties of over-land mobility in robotics are just too great to think that $32 million in development funds (as mentioned in the article) will be able to develop a robot that is even half as capable, or intelligent, as a grunt that spent a few months in basic. This is why the greatest advances in military robotics have come in aircraft - the mobility issues are much simpler. It is a lot easier to program an autopilot for a drone than it is to program an autodrive for an off-road Humvee (which, incidentally, they haven't yet been able to do, hence the Grand Challenge). I'll call it vaporware.

It is possible to design complex digital systems without central clocks. You obviously know little of digital design beyond that which is taught in an level 100 University course.

If you don't believe me then check out this page at DARPA on the subject:

http://www.darpa.mil/mto/clockless/

Hmm...maybe those DARPA folks didn't realize that all of their flip flops have a clock input on them!

oil is inherently a finite resource. the US government is always reluctant to fund alternate sources of automobile fuel until an absolute oil crisis is on the horizon. sure, the hybrid thing is a stop-gap solution, but at least it pushes a little against the suppliers and can temporarily delay price increases while we work towards a more permanent solution (maybe fuel cell, but it's just too damn expensive now to be viable for private consumers).

for those of you who tell me this gas problem is only temporary and that we need to innovate to improve oil extraction productivity, bear in mind that this is not the oil shock of the 70s: we now have a strong China and India on the scene, who do not help OPEC's wheeling and dealing of supply/demand. those countries' demand will not go away, and for the short term, neither will the political unrest in the Middle East which threatens our supply of crude. c'mon thinking we can stabilize the Middle East is like Englishmen talking about actively Westernizing China: people have been trying unsuccessfully for so many years; what makes you think you can do it?

having worked for an R&D consulting firm, i often spoke with the big gas extracting and refining firms. they are so hesitant to move into a new technology because they keep getting paid for gas. i can see where they're coming from. still, the smartest companies practice technology roadmapping to allocate resources towards the development of a disruptive technology (read Xerox copy machines, VOIP) which will leave their competitors in the dust. so yes, it does make business sense to diversify and innovate, but large public companies are reluctant to allocate too many resources towards these activities because they are busy meeting quarterly earnings requirements and satisfying their stakeholders.

with the big companies unable to wholly fund alternate fuel R&D efforts, the onus lies on the US government to help fund these programs. yet, non-defense/DARPA funding towards science programs has been cut. i wonder where that money is going?

OHHH:

tie fighters it's a bit of a catch-22; i only wish it was as funny as the book.

woot

HELLAD DARPA PAGE

Or more likely Linux (if the DARPA Grand Challenge goes anywhere-I hope it does). It looks like a lot of the bots run on Linux of some kind (at least one OS X too).

CMU tried that last year. They'd obtained custom aerial and LIDAR imagery of the route, and had a semitrailer full of people at workstations manually programming the route in the two hours before the race.

It didn't work. They crashed three times in eight miles. Notably, they plowed through a sheet metal fence DARPA had placed on the route, and which apparently wasn't in their aerial photographs.

This year, the USAF colonel running the Grand Challenge made it very clear that preprogramming wouldn't work. Many of the teams groaned when he put up the slide with the pictures of tank traps.

We just put in the waypoint file, exactly as DARPA gives it to us, and go. We don't do any manual preprocessing.

Just so all you geeks out there know, the final lineup for the DARPA GC has not been decided yet.

I wouldn't get my hopes that high. Somebody has to falter for you to get on the track.

The alternates will take the place of any of the 40 semifinalist teams announced last
month that are unable to attend the September 2005 National Qualification Event (NQE) at the
California Speedway, Fontana, California. During Grand Challenge 2004, three semifinalist
teams experienced vehicle problems in the final weeks before the event and were unable to
attend. In identifying alternates for the NQE, DARPA is ensuring there will be a full field of
competitors at the NQE.

http://www.darpa.mil/grandchallenge/Press_Release_ Final_July_5.pdf

the us is doing this type of research as well, and are much further advanced than the japanese.

http://www.darpa.mil/tto/programs/hyfly.html

Something like five URLS in the story and not a single link directly to the semifinalists?

The route will be no more than 175 miles over desert terrain featuring natural and man-made obstacles. The exact route will not be revealed until two hours before the event begins.

Ok, all these experts say we'll run out of fossil fuels in about 50 years at our current consumption.

Ok, all these experts say we'll run out of fossil fuels in about 50 years at our current consumption.

Ok, all these experts say we'll run out of fossil fuels in about 50 years at our current consumption.

Ok, all these experts say we'll run out of fossil fuels in about 50 years at our current consumption.

Ok, all these experts say we'll run out of fossil fuels in about 50 years at our current consumption.

The program started last December. 8 teams have been selected to participate. Here is the short description from the DARPA announcement:

The envisioned outcome of the LAGR program is a revolutionary methodology for developing autonomous navigation software systems. Current systems typically rely on hand-crafted, hand-tuned algorithms for obstacle detection and avoidance. In LAGR, algorithms will be created that learn how to navigate based on their own experience and by mimicking human teleoperation. It is expected that systems developed in LAGR will provide a performance breakthrough in navigation through complex terrain.

More details here: http://www.darpa.mil/ipto/programs/lagr/.

Does anyone know a bookie accepting bets on the outcome? With 117 teams still in the race and plenty of media coverage there would be a huge market.

Go Stanley!

I know NASA has had some difficulties, but science fiction has clearly influenced our expectations for the feasible.

For an example of what I mean, check out the DARPA Grand Challenge. That's here on EARTH and they don't get far.

Robotics is an extremely interesting field--one which I've had the pleasure of working in at UTexas, Austin but it is still quite in its infancy.

You say "Oh just give it extra legs like a spider". You have no idea of the shear logistics that go into something like that. Giving it a system like that would be another expensive project by itself involving years of work on control mechanisms, sensor mechanisms, PID controllers for feedback to every joint, swivle, whatnot. If one TINY thing is mis-anticipated; if the sand has a little bit more friction or a little less; if the wind blows it into the joints of your spider; if the system has to rely on hydraulics (lord have mercy on you) the entire thing can grind to a hault.

It's not iPods or laptops we should be thinking about when it comes to these batteries, its exoskeletons and robots. Who can forget Arnie's 10-year nuclear battery in T2? Once these become commonplace, and I am sure that they will, nuclear boogeymen notwithstanding, one of the major hurdles in large scale robotics production is bypassed.

Heck, while we're at it, why not slip a rack of them into your car? I could do with not having to stop at a gas station for 10 years! Aeroplanes, helicopters even might be limited by the weight of these things, but assuming no shortage of materials, you could put one of these batteries into every home, and run your house from it! The possibilities are endless...

Let's see here, should the blind allow robots to lead them around.?...Hmmm....let me think about that one. MMMmmm... No.

Simple things were serious issues like - if you're going slowly and your wheel comes up against a rock and at the current amount of throttle, it can't get over, what do you do - how do you know to just give it a little more gas and drive right over vs. you're up against something a little bigger that you should back up and drive around.

"The 'D' in DARPA stands for Defence"

No it doesn't, at least not in the USA. :-)

Defense Advanced Research Projects Agency

The DARPA Grand Challenge is a lot more interesting, and useful to boot...