ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills
Loading...
Date
2020
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association and John Wiley & Sons Ltd.
Abstract
Humans are highly adept at walking in environments with foot placement constraints, including stepping-stone scenarios where footstep locations are fully constrained. Finding good solutions to stepping-stone locomotion is a longstanding and fundamental challenge for animation and robotics. We present fully learned solutions to this difficult problem using reinforcement learning. We demonstrate the importance of a curriculum for efficient learning and evaluate four possible curriculum choices compared to a non-curriculum baseline. Results are presented for a simulated humanoid, a realistic bipedal robot simulation and a monster character, in each case producing robust, plausible motions for challenging stepping stone sequences and terrains.
Description
@article{10.1111:cgf.14115,
journal = {Computer Graphics Forum},
title = {{ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills}},
author = {Xie, Zhaoming and Ling, Hung Yu and Kim, Nam Hee and Panne, Michiel van de},
year = {2020},
publisher = {The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {10.1111/cgf.14115}
}