Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior

Gabriel B. Margolis, Pulkit Agrawal
Proceedings of The 6th Conference on Robot Learning, PMLR 205:22-31, 2023.

Abstract

Learned locomotion policies can rapidly adapt to diverse environments similar to those experienced during training but lack a mechanism for fast tuning when they fail in an out-of-distribution test environment. This necessitates a slow and iterative cycle of reward and environment redesign to achieve good performance on a new task. As an alternative, we propose learning a single policy that encodes a structured family of locomotion strategies that solve training tasks in different ways, resulting in Multiplicity of Behavior (MoB). Different strategies generalize differently and can be chosen in real-time for new tasks or environments, bypassing the need for time-consuming retraining. We release a fast, robust open-source MoB locomotion controller, Walk These Ways, that can execute diverse gaits with variable footswing, posture, and speed, unlocking diverse downstream tasks: crouching, hopping, high-speed running, stair traversal, bracing against shoves, rhythmic dance, and more. Video and code release: https://gmargo11.github.io/walk-these-ways

Cite this Paper

BibTeX
@InProceedings{pmlr-v205-margolis23a, title = {Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior}, author = {Margolis, Gabriel B. and Agrawal, Pulkit}, booktitle = {Proceedings of The 6th Conference on Robot Learning}, pages = {22--31}, year = {2023}, editor = {Liu, Karen and Kulic, Dana and Ichnowski, Jeff}, volume = {205}, series = {Proceedings of Machine Learning Research}, month = {14--18 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v205/margolis23a/margolis23a.pdf}, url = {https://proceedings.mlr.press/v205/margolis23a.html}, abstract = {Learned locomotion policies can rapidly adapt to diverse environments similar to those experienced during training but lack a mechanism for fast tuning when they fail in an out-of-distribution test environment. This necessitates a slow and iterative cycle of reward and environment redesign to achieve good performance on a new task. As an alternative, we propose learning a single policy that encodes a structured family of locomotion strategies that solve training tasks in different ways, resulting in Multiplicity of Behavior (MoB). Different strategies generalize differently and can be chosen in real-time for new tasks or environments, bypassing the need for time-consuming retraining. We release a fast, robust open-source MoB locomotion controller, Walk These Ways, that can execute diverse gaits with variable footswing, posture, and speed, unlocking diverse downstream tasks: crouching, hopping, high-speed running, stair traversal, bracing against shoves, rhythmic dance, and more. Video and code release: https://gmargo11.github.io/walk-these-ways} }
Endnote
%0 Conference Paper %T Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior %A Gabriel B. Margolis %A Pulkit Agrawal %B Proceedings of The 6th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Karen Liu %E Dana Kulic %E Jeff Ichnowski %F pmlr-v205-margolis23a %I PMLR %P 22--31 %U https://proceedings.mlr.press/v205/margolis23a.html %V 205 %X Learned locomotion policies can rapidly adapt to diverse environments similar to those experienced during training but lack a mechanism for fast tuning when they fail in an out-of-distribution test environment. This necessitates a slow and iterative cycle of reward and environment redesign to achieve good performance on a new task. As an alternative, we propose learning a single policy that encodes a structured family of locomotion strategies that solve training tasks in different ways, resulting in Multiplicity of Behavior (MoB). Different strategies generalize differently and can be chosen in real-time for new tasks or environments, bypassing the need for time-consuming retraining. We release a fast, robust open-source MoB locomotion controller, Walk These Ways, that can execute diverse gaits with variable footswing, posture, and speed, unlocking diverse downstream tasks: crouching, hopping, high-speed running, stair traversal, bracing against shoves, rhythmic dance, and more. Video and code release: https://gmargo11.github.io/walk-these-ways
APA
Margolis, G.B. & Agrawal, P.. (2023). Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior. Proceedings of The 6th Conference on Robot Learning, in Proceedings of Machine Learning Research 205:22-31 Available from https://proceedings.mlr.press/v205/margolis23a.html.

Related Material