GeRM: A Generalist Robotic Model with Mixture-of-experts for Quadruped Robot

arXiv:2403.13358v1 Announce Type: cross
Abstract: Multi-task robot learning holds significant importance in tackling diverse and complex scenarios. However, current approaches are hindered by performance issues and difficulties in collecting training datasets. In this paper, we propose GeRM (Generalist Robotic Model). We utilize offline reinforcement learning to optimize data utilization strategies to learn from both demonstrations and sub-optimal data, thus surpassing the limitations of human demonstrations. Thereafter, we employ a transformer-based VLA network to process multi-modal inputs and output actions. By …

abstract arxiv cs.cv cs.lg cs.ro current data datasets diverse experts however importance offline paper performance reinforcement reinforcement learning robot robotic strategies training training datasets type