Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Robel: Robotics Benchmarks for Learning with Low-Cost Robots

ai.googleblog.com

Post reply on HN
Print view View original

11–20 of 26 posts

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#11

Post by markisus »

pmorais wrote: post #8
netinstructions wrote: post #2

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! https://sites.google.com/view/roboicsbenchmarks/getting-star... https://www.trossenrobotics.com/d-kitty.aspx https://www.trossenrobotics.com/d-claw.aspx Not quite in the realm like the $50 to $90 'Google Voice / Google Vision AI' kits https://aiyprojects.withgoogle.com/

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.

How do you know these prices? I generally find these companies do not publish them on their websites

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#12

Post by watersb »

rficcaglia wrote: post #10
pmorais wrote: post #8

Earlier quoted context omitted.

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.

Totally worth the money :) https://youtu.be/c3Cq0sy4TBs

"The Proud Robot", Henry Kuttner, Astounding 1943

https://www.prosperosisle.org/spip.php?article863

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#13

Post by mojolozzo »

dekhn wrote: post #9

I'm interested in hearing from anybody who has sucessfully built a useful research robot for my experience has been everything < $1K uses crappy servos.

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: https://m.youtube.com/watch?v=q8jgu-EtCFc

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

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#14

Post by dekhn »

mojolozzo wrote: post #13
dekhn wrote: post #9

I'm interested in hearing from anybody who has sucessfully built a useful research robot for my experience has been everything < $1K uses crappy servos.

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: https://m.youtube.com/watch?v=q8jgu-EtCFc 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 ar…

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?

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#15

Post by toxik »

markisus wrote: post #11
pmorais wrote: post #8

Earlier quoted context omitted.

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.

How do you know these prices? I generally find these companies do not publish them on their websites

Hardly secret around institutions that have them.

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#17

Post by Judgmentality »

mojolozzo wrote: post #13
dekhn wrote: post #9

I'm interested in hearing from anybody who has sucessfully built a useful research robot for my experience has been everything < $1K uses crappy servos.

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: https://m.youtube.com/watch?v=q8jgu-EtCFc 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 ar…

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

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#18

Post by natvert »

our lab uses Dynamixles in the widow x (https://www.trossenrobotics.com/widowxrobotarm) 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

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#19

Post by contingencies »

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.

Re: Robel: Robotics Benchmarks for Learning with Low-Cost Robots

#20

Post by cbames89 »

contingencies wrote: post #19

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…

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.
Post reply on HN