Adaptive Slope Locomotion with Deep Reinforcement Learning

William Jones, Tamir Blum, Kazuya Yoshida

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

In this paper we present a model free Deep Reinforcement Learning based approach to the motion planning problem of a quadruped moving from a flat to an inclined plane. In our implementation, we do not provide any prior information of the location of the inclined plane, nor pass any vision data during the training process. With this approach, we train a 12 degree of freedom quadruped robot to traverse up and down a variety of simulated sloped environments, in the process demonstrating that deep reinforcement learning is able to generate highly dynamic and adaptable solutions.

Original languageEnglish
Title of host publicationProceedings of the 2020 IEEE/SICE International Symposium on System Integration, SII 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages546-550
Number of pages5
ISBN (Electronic)9781728166674
DOIs
Publication statusPublished - 2020 Jan
Event2020 IEEE/SICE International Symposium on System Integration, SII 2020 - Honolulu, United States
Duration: 2020 Jan 122020 Jan 15

Publication series

NameProceedings of the 2020 IEEE/SICE International Symposium on System Integration, SII 2020

Conference

Conference2020 IEEE/SICE International Symposium on System Integration, SII 2020
Country/TerritoryUnited States
CityHonolulu
Period20/1/1220/1/15

Keywords

  • DPPO
  • Quadruped
  • Reinforcement Learning
  • Slope Walking
  • V-REP

Fingerprint

Dive into the research topics of 'Adaptive Slope Locomotion with Deep Reinforcement Learning'. Together they form a unique fingerprint.

Cite this