AdaptNet: Policy Adaptation for Physics-Based Character Control
This work addresses the challenge of slow policy adaptation in physics-based character control for robotics and animation, offering an incremental improvement over existing methods.
The paper tackles the problem of efficiently adapting existing reinforcement learning policies to new tasks, such as different locomotion styles or environmental changes, by introducing AdaptNet, which modifies the latent space and policy layers to achieve new behaviors with significantly reduced training times compared to learning from scratch.
Motivated by humans' ability to adapt skills in the learning of new ones, this paper presents AdaptNet, an approach for modifying the latent space of existing policies to allow new behaviors to be quickly learned from like tasks in comparison to learning from scratch. Building on top of a given reinforcement learning controller, AdaptNet uses a two-tier hierarchy that augments the original state embedding to support modest changes in a behavior and further modifies the policy network layers to make more substantive changes. The technique is shown to be effective for adapting existing physics-based controllers to a wide range of new styles for locomotion, new task targets, changes in character morphology and extensive changes in environment. Furthermore, it exhibits significant increase in learning efficiency, as indicated by greatly reduced training times when compared to training from scratch or using other approaches that modify existing policies. Code is available at https://motion-lab.github.io/AdaptNet.