Related papers: Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motions

Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motions

URL: http://arxiv.org/abs/2209.07899v1
Date: Fri, 16 Sep 2022 12:49:04 GMT
Title: Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motions
Authors: Chenhao Li, Sebastian Blaes, Pavel Kolev, Marin Vlastelica, Jonas Frey, Georg Martius
Abstract summary: We propose a cooperative adversarial method for obtaining versatile policies with controllable skill sets from unlabeled datasets. We show that by utilizing unsupervised skill discovery in the generative imitation learning framework, novel and useful skills emerge with successful task fulfillment. Finally, the obtained versatile policies are tested on an agile quadruped robot called Solo 8 and present faithful replications of diverse skills encoded in the demonstrations.
Score: 19.626042478612572
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Learning diverse skills is one of the main challenges in robotics. To this end, imitation learning approaches have achieved impressive results. These methods require explicitly labeled datasets or assume consistent skill execution to enable learning and active control of individual behaviors, which limits their applicability. In this work, we propose a cooperative adversarial method for obtaining single versatile policies with controllable skill sets from unlabeled datasets containing diverse state transition patterns by maximizing their discriminability. Moreover, we show that by utilizing unsupervised skill discovery in the generative adversarial imitation learning framework, novel and useful skills emerge with successful task fulfillment. Finally, the obtained versatile policies are tested on an agile quadruped robot called Solo 8 and present faithful replications of diverse skills encoded in the demonstrations.