Slashdot Mirror
Japanese SCHAFT Takes the Gold at DARPA Robot Challenge
savuporo writes "The two days of DARPA's humanoid robotics challenge are now over. 16 teams entered in three categories — custom built humanoid, DARPA supplied Atlas platform, and a non-humanoid form — and competed in eight different tasks. The all-Japanese SCHAFT team scored 27 out of 32 maximum points, followed by IHMC Robotics and Tartan Rescue, with 20 and 18 points. The tasks included challenges like driving a vehicle, climbing ladders and walls, using handheld tools to cut through walls, etc. All robots had a mix of autonomy and teleoperated controls to accomplish the tasks. Full details on scores can be found here. The eight teams that scored highest will get continued funding from DARPA to compete in the final challenge in 2014. Two NASA teams also entered, and the JPL-built non-humanoid RoboSimian placed 5th, whereas the JSC built and touted 'Valkyrie' came out of competition with zero points. Team SCHAFT and Boston Dynamics (building the Atlas platform) were recently acquired by Google."
Reader mikejuk says the scores "[make] the performance sound better than it actually was": " Each task could take 30 minutes and most of the robots took their time and moved as slow as ice. It seems that the teams were precomputing every move and taking a lot of time rather than getting on with the task as quickly as possible. As a result there is farther to go in creating useful rescue bots than the scores might suggest."
Looked the coolest but amounted to nothing. It's a wonder that NASA rockets don't break more often
Looked the coolest but amounted to nothing. It is a wonder that NASA rockets don't break more often.
I'm talkin' 'bout SCHAFT!
what is this, FARK?
This is amazing, and it always the same pattern - first sci-fi movie or a book (Jules Verne), then from 10 to 50 years and actual invention follows. Looks like we will have our terminator overlords quite soon.
I would guess the team name refers to
Patlabor
"Help! I'm drowning!"
"TARGET SIGHETED! CALCULATING RESCUE STRATEGY..."
"Hel....(glub)"
"TARGET HAS DESCENDED BELOW SURFACE BY 1 METER. RECALCULATING..."
"..."
"TARGET HAS DESCENDED BELOW SURFACE BY 3 METERS. RECALCULATING..."
"..."
"TARGET HAS DESCENDED BELOW SURFACE BY 6 METERS. RECALCULATING..."
"..."
"TARGET HAS EXPIRED. RESCUE NO LONGER POSSIBLE."
Elsewhere...
"Help! Shark!"
"TARGET SIGHETED! CALCULATING RESCUE STRATEGY..."
"HELP! Oh! ARGGHHH!!!! IT BIT MY LEG OFF!!!"
"TARGET MASS HAS DECREASED BY 10 KILOGRAMS. RECALCULATING..."
"ARRGGHHH!!!"
I bet the other 5 points will be addressed in the BR release.
The other guys got SCHAFTed!
this is a robotic response to reports of unrestrained hooliganism. look away from your POT (Personal Open Terminal) but do not leave your station, further instructions to follow. the innocent stem cells remain as hostage with us.
I always have routed for robotics. And I have always been disappointed by them.
Let's talk basics, pre-robotics. Video production 101 at the community college. Whoever put the show on could not sync audio and video, and the final 'closing ceremony' was nothing but a 'technical difficulty' screen. Face it, when we can't use technology to broadcast an event, 60 years after TVs invention, we can not do anything that resembles rescue oriented robotics.
The robots themselves, were lacking. Not the hardware platforms, they seemed ok, but the software sucked. Dabbling around robotics myself, I understand why, and acknowledge the teams efforts. But the fact remains; even for a first attempt, I saw nothing 'promising'. We (as a planet) spend too much emphasis on blinking light arduinos, and far too little time encouraging software skills. Again, from personal experience, I can see how the teams had to use compilers with no remote debugging, probably archaic monolithic code ('C' and Assembler), hard to use cross compiling, and basically the kind of stuff you would only force on a development team if you wanted them to fail.
And while I said I thought the hardware looked OK, I will make an observation; 90% of the time the robots stood there doing nothing, that stupid single LIDAR was spinning its ass off. Was that just to keep its grease warm, or was it indeed a huge bottle neck to have only 1 apparently limited LIDAR? To me, the chassis builder (BDI) should have provided at least 3 LIDARS at varying elevations, and the software to analyse point cloud data and provide solid models of the environment should have been on their own independent processors. I get the feeling entirely too much time was required for the developers to do that themselves. Its something everybody needed to do, it should be provided with the hardware.
Then again, I could be completely wrong. I looked at the simulator code back when it was released but have long since forgotten it. Government funded, intended to be open, but I haven't seen where anyone has published anything. Maybe just too early.
slashdot troll = you make a compelling argument I do not like the implications of.
The whole concept of humanoid robots is kind of dumb to me. I mean I can get industrial robots, or autonomous vehicles like Google cars, but trying to make a humanoid robot is more like an obsession with dolls than anything. There's a reason the Japanese are best there, and I think it's because nobody else really cares about making moving dolls.
The pragmatic people who are trying to make robots that do useful work don't think about making them look like humans in the first place.
Wanted a SCHAFT robot. I got her one, then she dumped me.
Silence is a state of mime.
The competition accept the SCHAFT.
"Be nice, veer left, and never stop thinking" Iain Banks - Walking On Glass
"DARPA gets the SCHAFT"
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Here's the missing video for those of us who want to see it in action:
http://www.youtube.com/watch?v=diaZFIUBMBQ
Liberty.
Most of the discussion so far has been rather lame. But there's detailed video of everything. Not that silly feed for the general public, but the detail videos. For each event, there's video from four cameras - three watching the robot, and one watching the operators, presented as a four-quadrant image. That's all on line, unedited, 9 hours a day from each camera, no audio. In there, you can see what's really happening.
Most of the time, the robots are being teleoperated. DARPA introduced substantial lag (100ms during "good comms" periods, 1 second during "bad comms" periods, alternating once a minute) and bandwidth limits (1,000,000 bps each direction during "good comms" periods, 100,000 bps during "bad comms" periods.) This was to prevent excessive teleoperation and encourage autonomy. What they mostly got was really slow teleoperation. Event time limits were very generous and scoring wasn't time based. Many of the events were dumbed down from the original specs. DARPA wanted some successes.
Watch the debris removal task by Team Schaft starting at 02:33:00. That's one of the better runs. It's really slow, but the operator isn't constantly teleoperating. Once in a while the operator pushes a button. Some other teams used game controllers.
SHAFT won and he did all the tasks walking BACKWARD!! How cool is that!
Look at the guy on the left put down the game controller (PS3 style) on the left side of the table when the robot cleared.
He was using the gamepad and from time to time use the keyboard to grab/move/release the "hand".
> using handheld tools to cut through walls
I, for one, welcome or new katana-wielding nekomimi overladies!