Learning How to Walk: Warm-starting Optimal Control Solver with Memory of Motion

Teguh Santoso Lembono, Carlos Mastalli, Pierre Fernbach, Nicolas Mansard, Sylvain Calinon

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a framework to build a memory of motion for warm-starting an optimal control solver for the locomotion task of a humanoid robot. We use HPP Loco3D, a versatile locomotion planner, to generate offline a set of dynamically consistent whole-body trajectory to be stored as the memory of motion. The learning problem is formulated as a regression problem to predict a single-step motion given the desired contact locations, which is used as a building block for producing multi-step motions. The predicted motion is then used as a warm-start for the fast optimal control solver Crocoddyl. We have shown that the approach manages to reduce the required number of iterations to reach the convergence from ~9.5 to only ~3.0 iterations for the singlestep motion and from ~6.2 to ~4.5 iterations for the multi-step motion, while maintaining the solution’s quality.
Original languageEnglish
Title of host publication2020 IEEE International Conference on Robotics and Automation (ICRA)
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages1357-1363
Number of pages7
ISBN (Electronic)978-1-7281-7395-5
ISBN (Print)978-1-7281-7396-2
DOIs
Publication statusPublished - 15 Sep 2020
Event2020 International Conference on Robotics and Automation - Virtual conference, France
Duration: 31 May 202031 Aug 2020
https://www.icra2020.org/

Publication series

Name
PublisherIEEE
ISSN (Print)1050-4729
ISSN (Electronic)2577-087X

Conference

Conference2020 International Conference on Robotics and Automation
Abbreviated titleICRA 2020
CountryFrance
CityVirtual conference
Period31/05/2031/08/20
Internet address

Fingerprint

Dive into the research topics of 'Learning How to Walk: Warm-starting Optimal Control Solver with Memory of Motion'. Together they form a unique fingerprint.

Cite this