Towards Fatigue Modeling for Locomotion Tasks

Rui Xu, Noshaba Cheema, Erik Herrmann, Perttu Hämäläinen, Philipp Slusallek

In: Motion, Interaction and Games - Posters. ACM SIGGRAPH Conference on Motion in Games (MIG-2021) ACM 2021.


Modern deep reinforcement learning (RL) methods allow simulated characters to learn complex skills such as locomotion from scratch. To generate realistic and smooth movements, domain-specific knowledge, such as motion capture data, finite state machines or morphology-specific attributes are needed to guide the motion generation algorithms. Here we investigate biomechanical fatigue to improve symmetry and periodicity of the generated locomotion movements compared to methods found in previous literature.


Weitere Links

Towards_Fatigue_Modeling_for_Locomotion_Tasks___MIG_Poster.pdf (pdf, 2 MB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence