Slashdot Mirror

True, but they did build Sandstorm and H1ghlander.
And, they just won the Urban Grand Challenge recently (mentioned in TFA), so it's safe to say this is up their alley.

Many a CyberCriminal hath begged for mercy in the face of DARPA's hired mercenary, Sandstorm!

Route information is available here: http://redteamracing.org/index.cfm?method=page.dis play&page=Gc05.route&skin=popup

Was that pun intended? Given the circumstances regarding their test run, that's pretty funny.

Keep in mind that CMU has two entries this year - Sandstorm (last year's design with upgrades) and H1ghlander (still a hummer, but using different systems and software). I haven't heard anything about Sandstorm's qualification run, but their website says "H1ghlander nudged a gate and tipped one cone."

I went to CMU, so obviously I'm cheering for them however I have a tremendous amount of respect for any of the smaller teams that don't have the massive budget that most teams have.

...I'm actually following these days (next to the Space Shuttle).

So is there any place I can place my bets on which team is going to win?

I'd bet on Team Red over the OS X but only because they have two entries ;)

If you want to see some cool videos and read more about the actual design of the CMU entries (Sandstorm and H1ghlander) check out their website. The race log section is especially interesting.

Maybe its time to take bids from some of the new areospace startups instead of handing it off to old entrenched Boeing.

From TFA: Another announced team is t/Space, a consortium including such groups as Burt Rutan's Scaled Composites, Elon Musk's SpaceX, and Red Whittaker[1] (http://www.redteamracing.org/) of the Carnegie Mellon Robotics Institute. Some news reports in mid-March 2005, stemming from an interview with New Scientist have reported that t/Space intends to withdraw from the competition, citing a high paperwork burden; however, no announcement of a withdrawal has yet been made by t/Space.

However, as of May 3, it seems that only Lockheed Martin and Northrop Grumman did submit a bid.

Looks like the startups tried but said, screw it your bureaucracy sucks. We will just do it ourselves again.

Here's the link to CU. (Go Big Red!!)

Here's the link to CMU.

Nice try to sound smart, but its not like your the first to think that robots should have sensors.

Red team's sensors

It utilizes scanning radar, stereo vision, and scanning laser ranging. Position and orientation are estimated by six-axis inertial sensing and axle encoding that are fused with GPS by an Applanix pose estimator. Additionally, position is sensed by OmniStar GPS.

Generating models from digital sensor data is not simple. Integrating sensor data with machine intelligence is a profound challenge for navigating rough terrain. Sensor data is imperfect due to sensor calibration errors, visual glare, unusual surface properties, and error induced by vibration. Navigation computers must utilize this data while moving at high speed across rough terrain. It must continually make decisions with limited and imperfect information.

As for using 2D images to build a 3D map, if you really did try to implement disparity maps with webcams you should know the limitations in this. The time taken to calculate the disparity map, the semipoor resolution, noise, and lack of data in uniform areas make it only somewhat useful.

Laser range-finding is good, but keep in mind this is a laser. This means that you can really only check a range at one data point at a time. Granted, there are ways of dealing with this, such as scanning, or using prisms/mirrors to spread the laser into a plane, or a combination of the two, but it is still a limited technology.

Neither technique you mentioned is new nor the holy grail to robotic sensing. I would like to see you implement a machine vision system capable of picking out rocks in sand and small pits and other dangers while traveling at 50 mph.

My suspicion is that their sensor suite was ill-chosen. They had four line scanners, three fixed and one steerable. The fixed ones were aimed to the left, the right, and slightly upward. The steerable one was presumably aimed as far forward as it can get good data.

This sounds good, but it's not that effective a system. All of them were single-line scanners. So the vehicle has to assemble ground profiles from successive line scans. The hard-mounted units weren't stabilized, so they wouldn't produce good profiles beyond slow speeds. The long-range gimballed Reigl scanner was stabilized in three axes, but it's still a line scanner. It only gets one chance to see any point on the ground, and it sees it at maximum range and at the most oblique angle possible, the worst possible condition. Any problems, and you have to slow down and try to use the gimbal to pick up the missing data. Or you can just go plowing ahead, which is apparently what they did.

I think this establishes that line scanners aren't going to solve this problem. CMU had the fanciest single-line scanner ever built, and they crashed into a very clear obstacle. CMU was more successful in the 1980s with a 3D laser rangefinder on the Navlab project. That unit was too slow for this event, but was the right idea.

We'd been through this analysis a year ago on Team Overbot, and knew we needed something better. We know what's needed, but couldn't get it built in time. Our custom laser rangefinder vendor went out of business, and the alternative vendor couldn't deliver in time. Next time.

CMU's race log is silent about this. Their last entry ends "We can win this. Spare nothing. Victory or demise."

It beats that "dead" feeling you get when you lose" - Buffy

According to the team's website, the hero of our story, known as "Red", a self described dynamic individual, wrote merely that:

"There is no more practice, just impecable execution." redteamracing.org

So, it would seem that after all, the red team failed because it lacked "impecable execution". Still, getting a humvee to go 7 miles is impressive. I once got a lawnmower to go 14 miles unattended through some corn fields. But I suppose for college boys, 7 miles on a multi-million dollar car is a good start.

Best of luck to all competitors, and may they some day attain the "impeccable execution" shown by the seven-milers from CMU.

AMD sponsored an entry.

So did Intel

I think this was the whole idea, DARPA wants to make sure they have some kind of "race" on their hands, not just one team putting along till the end.

The Carnegie Mellon entry looks to be functioning properly at low speeds, but has had some problems staying up at higher speeds.

I think at least having more teams will force the CM team to actually make this a race instead of a walk.

I withheld midterm grades of our course, pending the outcome of QID, since this is a third of the grade. Since QID went well, grades will be good, and it will be a pleasure to assign these.

Withold the grades, and students will do anything. (Note to Theory profs out there: withhold grades until someone proves P != NP )

Attendance was about the same today except it didn't appear that there was as many media representatives present. Again temperatures were in the 90's. I acquired a media pass today and was allowed access to almost every area of the speedway including the pits and the start line. This will allow me to film each entry up close and interview members of the teams. DARPA is also publishing the daily events here and here.

Vehicle inspections on the rest of the field were performed today. DARPA is reacting as fast as they can to modify the rules and give every opportunity to each of the teams in hopes they will be able to qualify. DARPA is now allowing the teams as many appearances on the Q&D course as requested by the teams. The Q&D that was scheduled today became an opportunity for teams to iron out their problems on the track.

The Blue team with the CyberRider (the motorcycle entry Web Site) was the first on the Q&D course. It traveled about 20 feet when it fell over and exposed it's greasy side.

Team ENSCO Web Site traveled to the first major turn and failed to navigate the first sharp turn. This is also the same place Team TerraMax web site failed today and the Sci Autonics web site team.

The first turn appears to be difficult for the vehicles that make it there.

The ASI/Florida State Web Site had two more runs today. They disabled all of the perception systems and successfully ran about 1/3 of the course. The first run, the vehicle was driving like a "drunk sailor" according to a member of the team. Florida State took the recoded path data to tune their vehicle controller. ASI reported the vehicle tracked much better on the second run. I get the impression that a few more Q&D course test runs will be performed before the vehicle is ready to qualify.

Team Caltech Web Site made two more runs on the Q&D course. Well, both runs were consistent, but not as good as yesterday. Both times, cleared the start line and banked hard left as if it were going to the last way point. I hope to find out today what happened.

There were several other teams that attempted the Q&D course today with similar or worse results.

The highlight of the day was CMU's Red Team web site. As anticipated they made it to the finish line of the course and electrified the spectators and increased pressure to the rest of the field.

That's the highlights of the second day of Q&D testing.

Basicly, it took a turn too fast.

a picture.

this page is the running log put out by the group, and includes a description of the accident.

Basicly, it took a turn too fast.

a picture.

this page is the running log put out by the group, and includes a description of the accident.

Well, CMU was probably working pretty hard since their robot flipped and destroyed a bunch of equipment. But there is a post about successful qualification on monday .

Too bad CMU rolled their vehicle on March 4th. At this point I'm not sure if they're still the favorites. I sure hope so, I have a couple friends on the team.

They lost a significant amount of their equipment when it rolled, including a quarter million dollar gimbal

For more info on their roll, check out their race log

I'm disappointed that there is not more information available about this event as it happens. I've been following it vicariously for months, and now I'd like to hear about what is happening at the speedway in Fontana. How many teams showed up? How many tried out today? Which passed, which failed? I haven't been able to find out any of that information.

The so-called Science Blog article was from February 10! That's not exactly timely, is it?

Nagle's later posting here does present some information about Caltech. The Caltech team web page provides the same basic info, with a little different spin. But I guess we're lucky they posted today; the previous entry on the team's news page was dated November 16, 2003.

CMU has been updating almost every day, but their last entry was Saturday, saying "The curtain goes up Monday morning". Again, what happened?

You'd think in this age of bloggers, when every windbag on the net sees fit to tell us what he had for lunch that day, someone would be watching this event and posting some updates in the evening. If this isn't happening, I beg anyone who is attending to step up and start writing! Maybe I'm spoiled by the usual instant access to information, but I'm passionately interested in this event and starving for news.

Some parts are off road, thus you need to navigate around rocks, hills/gullies, potholes etc etc etc. Here are some previous generation stuff that the Robotics Institute at CMU has done over the las 20 years, many of which involve autonomous highway driving, but driving off road is much harder as you cant follow any lane markers and you need to determine the best path so you don't flip over.

Some parts are off road, thus you need to navigate around rocks, hills/gullies, potholes etc etc etc. Here are some previous generation stuff that the Robotics Institute at CMU has done over the las 20 years, many of which involve autonomous highway driving, but driving off road is much harder as you cant follow any lane markers and you need to determine the best path so you don't flip over.

Red Team: "Muahaha! Bow before the greatest CS school in the universe and our giant Hummer that is smarter than you!"

Blue Team: "Feh! We can beat you with two wheels tied behind our backs!" *obscene gesture*

Red Team: "Come a little closer and say that."

Clearly what everyone is missing is to have the car drive itself!

see here and here

At least I know the redteam has already done this. If you're building an autonomous car, making it remote controlled is a very early step along the way.

The Red Team has taken serious safety procautions that this Monster Garge thing seems not to address. There are E-Stop buttons all along the perimeter of the car, and the E-Stop mechanism isn't just an electrical cutoff, but also a huge brake on the drivetrain which engages if power is cut. I would advise anyone trying to build one of these things to use a similar system, since a runaway car could be expensive both in terms of dollars and lives.

That red Jeep has nothing to do with the Grand Challenge. That's Navlab 11, the Robotics Institute's latest test vehicle. the Robotics Institute, headed by Charles Thorpe, took a look at the Grand Challenge and decided to pass. He told me "If we entered, we'd have to win", and since he's mostly Government-funded, he'd need another source of funding, which he didn't have. Whittaker, who heads a related but separate operation, the Field Robotics Center, decided to do it on his own.

Whittaker issues a constant stream of trival press releases, like Team Equipped with Laptops and Office Equipment. We have considerable respect for the Robotics Institute at CMU, but this is becoming embarassing.

We take Team Caltech seriously, but not Whittaker's operation.

We will give a presentation on September 24, in EE380 at Stanford, on how we're doing it, and will show our vehicle, which isn't vaporware.

John Nagle
Team Overbot

I checked out Red Teams site and all I can say is "wow".

That and Mopar never gave me these options when I bought my Jeep! :(

Anybody got a mirror or a BitTorrent link for the Red Team video (http://www.redteamracing.org/include/media/movies /red_team.mpeg)?

CMU is one step ahead of you. There is already a seperate project devoted to the challenge in question.