Learning Dexterous In-Hand Manipulation

08/01/2018
by   OpenAI, et al.
0

We use reinforcement learning (RL) to learn dexterous in-hand manipulation policies which can perform vision-based object reorientation on a physical Shadow Dexterous Hand. The training is performed in a simulated environment in which we randomize many of the physical properties of the system like friction coefficients and an object's appearance. Our policies transfer to the physical robot despite being trained entirely in simulation. Our method does not rely on any human demonstrations, but many behaviors found in human manipulation emerge naturally, including finger gaiting, multi-finger coordination, and the controlled use of gravity. Our results were obtained using the same distributed RL system that was used to train OpenAI Five. We also include a video of our results: https://youtu.be/jwSbzNHGflM

READ FULL TEXT

page 1

page 2

page 3

page 5

page 9

page 22

research
10/25/2022

DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality

Recent work has demonstrated the ability of deep reinforcement learning ...
research
11/20/2018

Reinforcement Learning of Active Vision forManipulating Objects under Occlusions

We consider artificial agents that learn to jointly control their grippe...
research
08/04/2023

Nonprehensile Planar Manipulation through Reinforcement Learning with Multimodal Categorical Exploration

Developing robot controllers capable of achieving dexterous nonprehensil...
research
03/15/2022

Vision-Based Manipulators Need to Also See from Their Hands

We study how the choice of visual perspective affects learning and gener...
research
08/04/2023

Getting the Ball Rolling: Learning a Dexterous Policy for a Biomimetic Tendon-Driven Hand with Rolling Contact Joints

Biomimetic, dexterous robotic hands have the potential to replicate much...
research
07/02/2022

Learning Switching Criteria for Sim2Real Transfer of Robotic Fabric Manipulation Policies

Simulation-to-reality transfer has emerged as a popular and highly succe...
research
05/20/2023

DexPBT: Scaling up Dexterous Manipulation for Hand-Arm Systems with Population Based Training

In this work, we propose algorithms and methods that enable learning dex...

Please sign up or login with your details

Forgot password? Click here to reset