A reinforcement learning approach to synthesizing climbing movements

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

Researchers

Research units

Abstract

This paper addresses the problem of synthesizing simulated humanoid climbing movements given the target holds, e.g., by the player of a climbing game. We contribute the first deep reinforcement learning solution that can handle interactive physically simulated humanoid climbing with more than one limb switching holds at the same time. A key component of our approach is Self-Supervised Episode State Initialization (SS- ESI), which ensures diverse exploration and speeds up learning, compared to a baseline approach where the climber is reset to an initial pose after failure. Our results also show that training with a multi-step action parameterization can produce both smoother movements and enable learning from slightly fewer explored actions at the cost of increased simulation time per action.

Details

Original languageEnglish
Title of host publicationIEEE Conference on Games 2019, CoG 2019
Publication statusPublished - 1 Aug 2019
MoE publication typeA4 Article in a conference publication
EventIEEE Conference on Games - London, United Kingdom
Duration: 20 Aug 201923 Aug 2019

Publication series

NameIEEE Conference on Computatonal Intelligence and Games
PublisherIEEE
Volume2019-August
ISSN (Print)2325-4270
ISSN (Electronic)2325-4289

Conference

ConferenceIEEE Conference on Games
Abbreviated titleCoG
CountryUnited Kingdom
CityLondon
Period20/08/201923/08/2019

    Research areas

  • Action parameterization, Climbing movements, Reinforcement learning, State initialization

ID: 37821635