Diff-Transfer: Model-based Robotic Manipulation Skill Transfer via Differentiable Physics Simulation
This work addresses the challenge of enabling intelligent robots to efficiently adapt mastered skills to similar but new tasks, which is incremental as it builds on existing differentiable simulation methods.
The paper tackles the problem of transferring robotic manipulation skills to novel tasks by introducing Diff-Transfer, a framework that uses differentiable physics simulation to adapt known actions along a task path, achieving successful skill transfer in four challenging simulation experiments.
The capability to transfer mastered skills to accomplish a range of similar yet novel tasks is crucial for intelligent robots. In this work, we introduce $\textit{Diff-Transfer}$, a novel framework leveraging differentiable physics simulation to efficiently transfer robotic skills. Specifically, $\textit{Diff-Transfer}$ discovers a feasible path within the task space that brings the source task to the target task. At each pair of adjacent points along this task path, which is two sub-tasks, $\textit{Diff-Transfer}$ adapts known actions from one sub-task to tackle the other sub-task successfully. The adaptation is guided by the gradient information from differentiable physics simulations. We propose a novel path-planning method to generate sub-tasks, leveraging $Q$-learning with a task-level state and reward. We implement our framework in simulation experiments and execute four challenging transfer tasks on robotic manipulation, demonstrating the efficacy of $\textit{Diff-Transfer}$ through comprehensive experiments. Supplementary and Videos are on the website https://sites.google.com/view/difftransfer