Robel: Robotics Benchmarks for Learning with Low-Cost Robots


88 points | by theafh 136 days ago


  • netinstructions 136 days ago

    They may be lower cost than the status quo, but $3199 to $3699 is out of my reach as a (non professional) hobbyist. I hope maybe one day though!

    Not quite in the realm like the $50 to $90 'Google Voice / Google Vision AI' kits

    • pmorais 136 days ago

      You’re not the intended audience. This is meant for small research labs that are just starting up and want to enter the RL/Robotics research space and can’t afford a $400,000 PR2 or $150,000 shadow hand.

    • gene-h 136 days ago

      The main cost is the actuators. The servos they use are the only ones out there with integrated feed back from the sensors which is required so they can calculate hardware safety. They also have, albeit pretty bad, torque control which is necessary for many approaches to controlling legged robots and manipulators.

      • contingencies 135 days ago

        According to a Chinese SCARA motor producer, Tamagawa is a notable Japanese manufacturer of servos used in SCARA robots.

        • lnsru 136 days ago

          Wouldn’t stepper motor with encoder and current sensor be cheaper?

          • gene-h 136 days ago

            I'm skeptical that a stepper motor would be able to support it's own weight and more without a gearbox. This is especially important for making serial manipulators. Also keep in mind that what they're aiming for here is repeatability so that researchers can easily compare their algorithms. That means using parts that don't vary too much and minimizing the amount of assembly researchers have to do. There are much cheaper knock off versions of these actuators, although they may not be as repeatable. At the very least they aren't as well documented.

            • Iv 136 days ago

              Steppers are highly repeatable.

              • tastroder 136 days ago

                This is more about repeatable experiments than particular movements.

                Regarding the price, the actuators seem to be ~240USD. A good stepper motor with the appropriate feedback mechanism to make it suitable for servo-like control plus a modern stepper motor controller that is suited for the robotics context will likely not be (much) cheaper and you have to hack together the servo functionality, tune settings, etc. - which seems detrimental if the goal is repeatability accross teams. I'm not in the target audience for these robots either but from the perspective of robustness and repeatable research they don't look too shabby.

            • codesushi42 136 days ago

              That wouldn't work for a robotic hand.

        • contingencies 136 days ago

          Maybe I am missing something but fundamentally you can "learn" in the ML sense with any sensor and effector.

          Examples of cheap sensors: MEMS microphone, camera, voltage sensor, mass sensor, distance sensor, multi-axis position sensors (gyroscope/magnetometers).

          Examples of effectors: Any kind of motor, solenoid, LEDs, etc.

          If you want to constrain the question to 3D motion, here is a suggestion - hack the controller of any existing RC car platform. Add overhead position sensing within a fixed arena added via external camera. Maybe add a MEMS microphone (USD$3) or position sensor (~USD$10) to verify airtime/orientation. ML problem #1: Add a ramp. Try to get it to jump highest (longest time airborne). ML problem #2: Same with a power efficiency metric. ML problem #3: Same with a time efficiency metric applied to navigation from a random start point and orientation. ML problem #4: Motor noise vs. jump height optimization.

          • cbames89 136 days ago

            Byron Boot's lab at Georgia Tech does a lot of interesting work with larger-scale R/C vehicles. That being said, by adding more degrees of freedom you're making the configuration space of the robot higher. In many cases, higher dimension configuration spaces are more difficult for traditional sampling-based approaches. Learning approaches may be able to bias their solutions to avoid large sections of this configuration space.

          • dekhn 136 days ago

            I'm interested in hearing from anybody who has sucessfully built a useful research robot for < $1K. Even just a SCARA arm.

            my experience has been everything < $1K uses crappy servos.

            • mojolozzo 136 days ago

              The servos are certainly where you can spend most money when building a robot for research. I tried building small research robots with small servos ($25 each), see results at the end of this video:

              The total cost for one robot was well below 1k. The servos are $25 each (Turnigy 306G if remember correctly), all connecting parts are 3d printed, the electronics and batteries are about $50, and you could add a raspberry pi for $50. So in total <$500, depending on number of DOFs.

              Of course, these servos don't have the same torque as Dynamixles, for example. But they are lighter!

              • Judgmentality 136 days ago

                Do you have any videos of the actual robots or just the simulation?

                • dekhn 136 days ago

                  thanks, that's what I was looking for. Do you find the 306Gs work reliably and the cables don't get messed up in routine research work?

                • Iv 136 days ago

                  I am trying to build a suspended robot for about $1000 as an open source project:


                  Unfortunately at that price it comes without a manipulator, but I have bought (very) cheap arms and will experiment with them soon.

                • natvert 136 days ago

                  our lab uses Dynamixles in the widow x ( and the servos alone are 70% of the total cost.

                  is anyone aware of an open-source alternative to these servos? this would be a great contribution to the community, if not available already

                • toxik 136 days ago

                  Curious that they wouldn’t include PPO in their related work, but DDPG. Did I miss something important that makes PPO inapplicable here?

                  • DrNuke 136 days ago

                    You can translate this short article of mine to get real, low-cost “toys” for a small mechatronics lab suited to almost any dl & rl starting level endeavor you want. (tldr: less than $1.5k would suffice in first instance)