Forgot your password?
typodupeerror
Robotics Science

Humanoid Robots For the Next DARPA Grand Challenge? 53

Posted by samzenpus
from the thank-you-very-much-mr-roboto dept.
HizookRobotics writes "The official announcement should be out very soon, but for now here's the unofficial, preliminary details based on notes from Dr. Gill Pratt's talk at DTRA Industry Day: The new Grand Challenge is for a humanoid robot (with a bias toward bipedal designs) that can be used in rough terrain and for industrial disasters. The robot will be required to maneuver into and drive an open-frame vehicle (eg. tractor), proceed to a building and dismount, ingress through a locked door using a key, traverse a 100 meter rubble-strewn hallway, climb a ladder, locate a leaking pipe and seal it by closing off a nearby valve, and then replace a faulty pump to resume normal operations — all semi-autonomously with just 'supervisory teleoperation.' It looks like there will be six hardware teams to develop new robots, and twelve software teams using a common platform."
This discussion has been archived. No new comments can be posted.

Humanoid Robots For the Next DARPA Grand Challenge?

Comments Filter:
  • Is there such a thing as a non-bipedal humanoid design? Humanoid suggests bipedal, does it not?
  • Excellent. Can we make one weighing 60 ton's armed with dual LBX-20's, a Large Clan Laser, and 4 C Streak SRM 6's, with 12 tons of extra ammo?
    • by Anonymous Coward

      That would be awesome for all those situations in which you'd like to have a tank, only far slower, more cumbersome, and easier to disable.
      Bonus points if you can also add the BattleTech universe's atrocious targeting computers and ridiculously unsafe, overheating nuke power plants.

    • Re:Humanoid Robot... (Score:5, Interesting)

      by Sibko (1036168) on Friday April 06, 2012 @02:28AM (#39595261)
      I am afraid not.

      A 60 ton Clan chassis has enough tonnage available, (assuming it moves 1/2/0, has no armour, and utilizes an XL engine with endosteel internals) but lacks the internal critical space necessary to actually fit all of that. (Staring at the design right now in solaris skunkwerks [solarisskunkwerks.com])

      Two LBX20's and 4 cStreak6's are pretty hefty firepower though. Even for a Clan heavy. Also, 12 tons of ammo? That is enough for 45 turns of non-stop firing, most matches do not even last 16!

      Now, I *could* build a custom mech from scratch that fits some of that, but I decided a better choice would be to take ye old venerable Mad Dog (Vulture for you Freebirths) chassis and rig it with the LBX autocannons, though I was worried it would not fit them AND the ammo. Thankfully, the Mad Dog's design principle follows the 'fuck armor, put more guns on it' school of thought.

      I yanked the dual large pulse lasers and LRM20's from the prime variant, and jacked the heatsinks down to the base 12 (It is in no danger of overheating with just these autocannons). This gave me 28 tons to work with. 24 tons for the two LBX20's, and then 4 tons for the ammo. It is not the loadout I would personally take into battle with a Mad Dog, considering its armour, but for a daring young Trueborn out to claim a bloodname, I do not think it is bad.

      A 5/8/0 movement curve (reaching ~80kph at a run), enough ammo to fire the autocannons for about 10 turns (1.7 minutes) before running dry, and just enough armour to keep you mostly intact as you close into range. At 1610 battle value (2222 BV with a 3/4 standard Clan Pilot) it is actually pretty dangerous for the price. I also made sure to mix the ammo types - 2 tons cluster and 2 tons slug so that you get your holepunching and sandblasting ability all in one. Additionally, I removed the upper arm actuators to clear out some of the internal crit space in the arms. So not only does the LBX20 assembly manage to fit in just the arm itself, but since it is entirely in the arm and the upper arm actuator has been removed, the arms themselves can completely flip around, making this design even more dangerous in a close quarters fight as you maneuver through a formation and fire *backwards* (Hopefully at the rear armour of your enemy.)

      The significant downside to this design is the staying power. Light on armour (But then, all Mad Dog's are) and light on ammo. Once you run dry, there are no backup energy weapons, so you will be left adopting dezgra freebirth tactics like kicking your opponent to death. (*spit*)

      Additionally, the design lacks anything to absorb hits in the torso. Any breach of the armour that scores a critical hit is going to either blow out an ammo bin and cripple the mech, or damage the fusion reactor. (Granted, a damaged fusion reactor on this design is not so bad - thanks to the low heat output of the cannons. Even with half the reactor shielding gone, this mech is still going to run cool.)
      • by Ihmhi (1206036)

        It's posts like this that make me appreciate Slashdot.

      • Re: (Score:2, Funny)

        by dissy (172727)

        Two LBX20's and 4 cStreak6's are pretty hefty firepower though. Even for a Clan heavy. Also, 12 tons of ammo? That is enough for 45 turns of non-stop firing, most matches do not even last 16!

        I don't think he intended to take that mech into battle. He's probably just upset at the neighbors dog and wanted to teach it a lesson. I have the same instinctive response with cats and areal vehicles for some reason *shrug*

        • At least until the get the 3 Laws implemented and working reliably. Until then arming a them would seem a little risky.

    • by Ihmhi (1206036)

      60 tons is an awful lot.

      How about something around 10 tons? Maybe with a "Boxer" Shot Cannon. Adjust the bi-lateral angle to 3.5. And don't forget the Lambda driver!

      • by mhajicek (1582795)
        60 tons is a modern tank. Considering a mech is supposed to make tanks look like toy cars you're talking more like 600 tons minimum. If you assume the weight savings come from advanced materials, why not just use those materials on a tank? Besides, I prefer to mount three Clan AC-20's, fire linked in the torso of a Daishi with no arms.
  • by Lord_of_the_nerf (895604) on Friday April 06, 2012 @01:20AM (#39595033)
    ...the last part of the course is 'how to love' and 'question whether or not the unit has a soul'.
    • I would be fine with them dreaming of electric sheep but I would draw the line at giving them memories of fictional parents.

    • by dissy (172727)

      I'll be worried when ... the last part of the course is 'how to love' and 'question whether or not the unit has a soul'.

      Installation of a soul is trivially easy, but the licensing costs are outrageous!

  • That's ambitious (Score:5, Insightful)

    by Animats (122034) on Friday April 06, 2012 @01:31AM (#39595065) Homepage

    This is very ambitious. Yet it's almost within reach, if enough money is thrown at the problem. A lot of money.

    First, this requires solving low-speed legged locomotion over difficult terrain. Walking over rubble and climbing a ladder will be very tough. But it's not totally out of reach. A machine which can walk on either two or four limbs would be an advantage over rubble. Ladder climbing is a four-limb problem, and it's been done at least once. Grabbing the rungs with all limbs will make it easier.

    Replacing a pump is an interesting problem. I don't think anyone has yet demonstrated complex part replacement in an unstructured environment. That's probably the least well developed part of the problem.

    Building a suitable machine will be expensive. Each Boston Dynamics robot has cost upwards of $20 million. Even the Willow Robotics machine is over $100K per unit, and it's just two arms on a wheeled base. This thing has to have roughly human dimensions and be self contained. Petman isn't self-contained; it needs external power.

    Yet, looking at the problem, I can see how to approach it. Modern control theory is good enough. Machine learning and vision processing are good enough. Simulators are good enough to allow debugging in simulation. Enough people know this stuff that the job can be staffed.

    • You can climb a ladder with two limbs. I'm sure it won't be graceful, but there's no physical reason you need four limbs to climb a ladder.

      The ladder is easy, the entire thing can be solved before the robot even starts moving. The terrain isn't easy, but it's been done enough before to the point that it won't cause a huge amount of trouble. The real problems show up when trying to do things that humans use fine motor skills for. Operating screwdrivers, ratchets, a lot of actuation that has to be complia
      • The ladder is easy, the entire thing can be solved before the robot even starts moving.

        You're obviously a CS guy. :)

        Any real system has unmodelled mechanics, sensor noise and disturbances, all of which lead to uncertainty. You cannot "solve" the problem open-loop; you need feedback control and updates to your trajectory as you go, because over time your position will drift. And that's just for ladders. As you increase the problem to unstructured problems like walking on rubble or object manipulation, the difficulty increases by leaps and bounds. I agree that changing the pump will like

    • by chrb (1083577)
      You forgot the driving part. How is that going to be done? Just getting to the locked door is, in many ways, harder than the autonomous car challenge - the control and sensing has to be done completely from and by the humanoid robot, with limited power supply and (I would guess) not enough power/cpu/physical space to integrate as the advanced numerous sensors that the cars had. And instead of having the software wired in to the hardware control, the humanoid robot is going to have to drive using its limbs.
    • There are a lot of different platforms doing every aspect of this. The trick is to get them all on one system. There are dextrous arms with fingers, torsos with facial recognition and expression, and the Petman than seems to have the arms and legs (although I would like to see it without the support cables).
    • "First, this requires solving low-speed legged locomotion over difficult terrain. " This is why people have servos in their ankles for fine balance control.
    • by na1led (1030470)
      I don't think we have the technology to build a Terminator. The Power Source required is not available yet, and then you need very sophisticated A.I. which is still at least a decade away. They should concentrate on building Mechs instead.
    • by Missing.Matter (1845576) on Friday April 06, 2012 @10:02AM (#39597449)

      Machine learning and vision processing are good enough.

      I disagree on both counts. Machine learning is very good at specific problems. Current best techniques to train classifiers need millions of samples of ground truth training data, which is incredibly expensive to produce. Most classifiers like this make several assumptions about the training data, namely that it is representative of a stable underlying unknown distribution, which of course is hardly true. The best uses for machine learning techniques are for very specific applications where this assumption can be enforced, like optical character recognition. Machine learning has amazing applications, but it's hardly a field where I would say the results are "good enough," especially for most outdoor mobile robotics applications, like this DARPA project.

      Same goes for vision processing. The reason most robots today use LIDARs as their primary "vision" sensor is that image processing is very slow and not very good. With a 3d LIDAR you can do mapping, object detection and recognition, people detection, grasp planning, obstacle avoidance, etc. all very easily in polynomial time. Most of these tasks are computationally intractable using images, if they're possible at all (for example, night time operations).

      • Re:That's ambitious (Score:4, Informative)

        by Animats (122034) on Friday April 06, 2012 @11:25AM (#39598439) Homepage

        Same goes for vision processing. The reason most robots today use LIDARs as their primary "vision" sensor is that image processing is very slow and not very good. With a 3d LIDAR you can do mapping, object detection and recognition, people detection, grasp planning, obstacle avoidance, etc. all very easily in polynomial time. Most of these tasks are computationally intractable using images, if they're possible at all (for example, night time operations).

        10 years ago, that was true. Not any longer. Take a look at DARPA's current robot manipulation project. [youtube.com] Watch a vision-guided robot with two three-fingered hands put a key in a lock, turn the key, and open a door. That's being done with stereo vision and force feedback. The gold-colored device with two lenses is a Point Grey stereo vision device.

        Much of early vision processing is done in custom hardware or GPUs now. It's not "computationally intractable". It's fairly expensive computationally, because the amount of work per pixel is high. But it's not exponential.

        When I ran a DARPA Grand Challenge team a decade ago, we relied on LIDAR too much, and vision not enough. The winning teams relied almost entirely on vision when they were going fast - the LIDARs didn't have the range or angular accuracy to read terrain out to the stopping distance. What they got from vision was basically "distant part of road looks like near part of road - OK to go fast if near part of road is good", obtained from a classifier system. So they could out-drive their LIDARs. We couldn't exceed 17mph.

        • No it is true even today. Experiment's for DARPA's ARM project were conducted under bright lighting in ideal, static conditions. I've used the bumblebee cameras and other stereo vision systems, and they have limitations that inhibit the operation of many mobile robots including frame rate and maximum discernible range, (which is limited by the baseline of the camera), and most of all differences in lighting conditions.

          When I ran a DARPA Grand Challenge team a decade ago, we relied on LIDAR too much, and vision not enough. The winning teams relied almost entirely on vision when they were going fast - the LIDARs didn't have the range or angular accuracy to read terrain out to the stopping distance.

          Look at the more recent DARPA Urban challenges. Of the 6 teams that completed the challeng

          • by Animats (122034)

            Of the 6 teams that completed the challenge, all of them used the Velodyne 3D LIDAR as their primary vision sensor, in addition to several peripherial LIDARs. None of the teams who relied on vision primarily finished.

            True. The Urban Challenge was a slow-speed run, within LIDAR range. The latest Velodyne units see out to 70 meters. Note that you can't profile the road (to see potholes, etc.) out that far; the angle is too oblique. But you can see cars, pedestrians, telephone poles, etc.

            There are small short-range LIDAR devices, like the Swiss Ranger [mesa-imaging.ch] and photonic mixer devices [pmdtec.com], which have finally reached the Consumer Electronics Show level. For short ranges (a few meters), triangulation devices like the Kinect work. Lo

    • by mcgrew (92797) *

      Each Boston Dynamics robot has cost upwards of $20 million.

      That's what ENIAC, the world's first pocket calculator, cost in the late 1940s. And it wouldn't even fit in a pocket (took a large building to house) and didn't even have an LCD readout!

      Today's pocket calculators are more powerful than ENIAC and cost a dollar or two. So you can expect cheap robots in fifty years.

    • Yet, looking at the problem, I can see how to approach it. Modern control theory is good enough. Machine learning and vision processing are good enough.

      That's fine for the software... what about the hardware? Do we have actuators (hydraulic, pneumatic, whatever) that are good enough? And most important, do we have a power supply that's good enough?

      • by Animats (122034)

        That's fine for the software... what about the hardware? Do we have actuators (hydraulic, pneumatic, whatever) that are good enough? And most important, do we have a power supply that's good enough?

        Yes. They're not common items, but the right stuff does exist. There are military robots powered by small Diesel engines. [hdtglobal.com] The LS3 (the militarized successor to BigDog) is powered by a small Diesel engine, which drives both a generator and a hydraulic pump. (The U.S. military is all-Diesel; one standard fuel runs everything.)

        Miniature hydraulic cylinders and proportional valves provide actuators with very good power to weight ratios. Modern brushless servomotors aren't bad, either. Pneumatics are also an

  • The ultimate test of robotic ability is if it can go in the kitchen and build me a sammich! With limited or no tele-presense.
  • by Snospar (638389) on Friday April 06, 2012 @06:43AM (#39596005)

    Say one, perhaps gold in colour with a slightly annoying, camp voice, that's humanoid and another that's really the brains of the operation in a tri-wheeled body that communicates with a series of clicks & beeps (it's also got a useful rotary arm for shutting off that valve).

    It would make for a more entertaining video at the end of the competition.

  • Cyborgs!

    • by mcgrew (92797) *

      You're behind the curve. We cyborgs are here, now.

      You will be assimilated. You will not resist, you will pay good money for your cybernetic implants when your eyes or hips or knees no longer work.

      Most of us cyborgs are geezers, with cybernetic implants replacing natural but worn out body parts.

  • by scharkalvin (72228) on Friday April 06, 2012 @07:52AM (#39596351) Homepage

    The advantage of a humanoid robot, especially a bipedal one is that it can make use of tools and vehicles designed to be used by humans. Such a robot can perform tasks too dangerous for a human that could be done by a human. More specialized robots in non human form can be designed especially for high risk jobs, such as clean up of a nuclear disaster. The Japanese have been specializing in humanoid robots for some time as they have special needs. With an aging and shrinking population they see the need to develop service robots to take care of the ill and elderly. A proposed 'first step' was Project Atom to develop a humanoid robot with the Intelligence level of a 9 year old child. (The name for the project comes from Tezuka Osmau's Tetsuwan Atom manga, better known outside of Japan as Astro Boy).

    • by mark-t (151149)

      The advantage of a humanoid robot, especially a bipedal one is that it can make use of tools and vehicles designed to be used by humans.

      The idea of a non humanoid robot is so that it can make use of tools and vehicles that are *NOT* designed for human use, where if they were used usable by humans they would be less efficient, more costly to manufacture, etc.

      The only real so-called advantage of a humanoid robot is that we simply don't know of any other form that is as adaptable to as wide a range of tasks.

  • - Should they be nuclear powered? Or perhaps sufficient to trail a power and comms cable for 1km from nearest comms station.
    - Should study the exact jobs done by cleanup workers in Fukushima.
    - It would be great if the workers could stay miles away while an army of robots was slogging through radioactive ground water etc instead.
    - IANA Robotics Engineer but perhaps very good teleoperation and ability to get through tight spaces, reconfigure, work without wireless capability, and be radiation hardened, is muc

Never ask two questions in a business letter. The reply will discuss the one you are least interested, and say nothing about the other.

Working...