A GAN-Like Approach for Physics-Based Imitation Learning and Interactive Control

dc.contributor.authorXu, Peien_US
dc.contributor.authorKaramouzas, Ioannisen_US
dc.contributor.editorNarain, Rahul and Neff, Michael and Zordan, Victoren_US
dc.date.accessioned2022-02-07T13:32:37Z
dc.date.available2022-02-07T13:32:37Z
dc.date.issued2021
dc.description.abstractWe present a simple and intuitive approach for interactive control of physically simulated characters. Our work builds upon generative adversarial networks (GAN) and reinforcement learning, and introduces an imitation learning framework where an ensemble of classifiers and an imitation policy are trained in tandem given pre-processed reference clips. The classifiers are trained to discriminate the reference motion from the motion generated by the imitation policy, while the policy is rewarded for fooling the discriminators. Using our GAN-like approach, multiple motor control policies can be trained separately to imitate different behaviors. In runtime, our system can respond to external control signal provided by the user and interactively switch between different policies. Compared to existing method, our proposed approach has the following attractive properties: 1) achieves state-of-the-art imitation performance without manually designing and fine tuning a reward function; 2) directly controls the character without having to track any target reference pose explicitly or implicitly through a phase state; and 3) supports interactive policy switching without requiring any motion generation or motion matching mechanism. We highlight the applicability of our approach in a range of imitation and interactive control tasks, while also demonstrating its ability to withstand external perturbations as well as to recover balance. Overall, our approach has low runtime cost and can be easily integrated into interactive applications and games.en_US
dc.description.number3
dc.description.sectionheaderspapers
dc.description.seriesinformationProceedings of the ACM on Computer Graphics and Interactive Techniques
dc.description.volume4
dc.identifier.doi10.1145/3480148
dc.identifier.issn2577-6193
dc.identifier.urihttps://doi.org/10.1145/3480148
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1145/3480148
dc.publisherACMen_US
dc.subjectComputing methodologies
dc.subjectAnimation
dc.subjectPhysical simulation
dc.subjectReinforcement learning
dc.subjectcharacter animation
dc.subjectphysics
dc.subjectbased control
dc.subjectreinforcement learning
dc.subjectGAN
dc.titleA GAN-Like Approach for Physics-Based Imitation Learning and Interactive Controlen_US
Files