A reinforcement learning approach to synthesizing climbing movements

Loading...
Thumbnail Image

Access rights

openAccess
acceptedVersion

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Major/Subject

Mcode

Degree programme

Language

en

Pages

7

Series

IEEE Conference on Games 2019, CoG 2019, IEEE Conference on Computatonal Intelligence and Games ; Volume 2019-August

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.

Description

Other note

Citation

Naderi, K, Babadi, A, Roohi, S & Hamalainen, P 2019, A reinforcement learning approach to synthesizing climbing movements. in IEEE Conference on Games 2019, CoG 2019., 8848127, IEEE Conference on Computatonal Intelligence and Games, vol. 2019-August, IEEE, IEEE Conference on Games, London, United Kingdom, 20/08/2019. https://doi.org/10.1109/CIG.2019.8848127