SCIENCE@home

Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations

A. Rajeswaran,Vikash Kumar,Abhishek Gupta,John Schulman,E. Todorov,S. Levine

Published 2017 in Robotics: Science and Systems

ABSTRACT

Dexterous multi-fingered hands are extremely versatile and provide a generic way to perform multiple tasks in human-centric environments. However, effectively controlling them remains challenging due to their high dimensionality and large number of potential contacts. Deep reinforcement learning (DRL) provides a model-agnostic approach to control complex dynamical systems, but has not been shown to scale to high-dimensional dexterous manipulation. Furthermore, deployment of DRL on physical systems remains challenging due to sample inefficiency. Thus, the success of DRL in robotics has thus far been limited to simpler manipulators and tasks. In this work, we show that model-free DRL with natural policy gradients can effectively scale up to complex manipulation tasks with a high-dimensional 24-DoF hand, and solve them from scratch in simulated experiments. Furthermore, with the use of a small number of human demonstrations, the sample complexity can be significantly reduced, and enable learning within the equivalent of a few hours of robot experience. We demonstrate successful policies for multiple complex tasks: object relocation, in-hand manipulation, tool use, and door opening.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-61 of 61 references · Page 1 of 1

CITED BY

Showing 1-100 of 1301 citing papers · Page 1 of 14

123414
Next