DeepGait: Planning and Control of Quadrupedal Gaits Using Deep Reinforcement Learning

This letter addresses the problem of legged locomotion in non-flat terrain. As legged robots such as quadrupeds are to be deployed in terrains with geometries which are difficult to model and predict, the need arises to equip them with the capability to generalize well to unforeseen situations. In t...

Full description

Saved in:
Bibliographic Details
Published in:IEEE robotics and automation letters Vol. 5; no. 2; pp. 3699 - 3706
Main Authors: Tsounis, Vassilios, Alge, Mitja, Lee, Joonho, Farshidian, Farbod, Hutter, Marco
Format: Journal Article
Language:English
Published: Piscataway IEEE 01.04.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Computer simulation
Deep learning
deep learning in robotics and automation
Legged locomotion
Legged robots
Locomotion
Machine learning
Markov processes
motion and path planning
Motion planning
Optimization
Path planning
Physical simulation
Policies
Robot dynamics
Terrain
ISSN:2377-3766, 2377-3766
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This letter addresses the problem of legged locomotion in non-flat terrain. As legged robots such as quadrupeds are to be deployed in terrains with geometries which are difficult to model and predict, the need arises to equip them with the capability to generalize well to unforeseen situations. In this work, we propose a novel technique for training neural-network policies for terrain-aware locomotion, which combines state-of-the-art methods for model-based motion planning and reinforcement learning. Our approach is centered on formulating Markov decision processes using the evaluation of dynamic feasibility criteria in place of physical simulation. We thus employ policy-gradient methods to independently train policies which respectively plan and execute foothold and base motions in 3D environments using both proprioceptive and exteroceptive measurements. We apply our method within a challenging suite of simulated terrain scenarios which contain features such as narrow bridges, gaps and stepping-stones, and train policies which succeed in locomoting effectively in all cases.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2377-3766
2377-3766
DOI:10.1109/LRA.2020.2979660