🤖 AI Summary
Dynamic motion generation faces challenges in adapting policies to changing targets and perturbations in physical parameters.
Method: This paper proposes a multi-task reinforcement learning framework featuring an encoder–decoder architecture and a policy selection mechanism to learn generalized foundational locomotion policies. Inspired by human sensorimotor adaptation, it enables zero-shot transfer of a single foundational policy across distinct motor skills (e.g., heading → volley kicking).
Results: Evaluated on a monopedal robot simulation platform, the framework outperforms state-of-the-art methods on multi-gait benchmark tasks. Crucially, it successfully reuses a heading policy to generate high-quality volleys under unseen target positions and ball restitution coefficients—without task-specific fine-tuning—demonstrating significantly enhanced adaptability and generalization in motor skill synthesis.
📝 Abstract
In this study, we propose a multitask reinforcement learning algorithm for foundational policy acquisition to generate novel motor skills. extcolor{hcolor}{Learning the rich representation of the multitask policy is a challenge in dynamic movement generation tasks because the policy needs to cope with changes in goals or environments with different reward functions or physical parameters. Inspired by human sensorimotor adaptation mechanisms, we developed the learning pipeline to construct the encoder-decoder networks and network selection to facilitate foundational policy acquisition under multiple situations. First, we compared the proposed method with previous multitask reinforcement learning methods in the standard multi-locomotion tasks. The results showed that the proposed approach outperformed the baseline methods. Then, we applied the proposed method to the ball heading task using a monopod robot model to evaluate skill generation performance. The results showed that the proposed method was able to adapt to novel target positions or inexperienced ball restitution coefficients but to acquire a foundational policy network, originally learned for heading motion, which can generate an entirely new overhead kicking skill.