PhyPlan: Generalizable and Rapid Physical Task Planning with Physics Informed Skill Networks for Robot Manipulators
This addresses the challenge of data-hungry and complex physical reasoning for robot manipulators, offering a generalizable approach for rapid task learning, though it appears incremental as it builds on existing methods like PINNs and MCTS.
The paper tackles the problem of enabling robots to perform dynamic physical tasks like throwing or sliding objects to reach goals, by introducing PhyPlan, a physics-informed planning framework that combines physics-informed neural networks with Monte Carlo Tree Search. It achieves lower regret, faster skill learning, and higher data efficiency compared to state-of-the-art methods in simulated 3D environments.
Given the task of positioning a ball-like object to a goal region beyond direct reach, humans can often throw, slide, or rebound objects against the wall to attain the goal. However, enabling robots to reason similarly is non-trivial. Existing methods for physical reasoning are data-hungry and struggle with complexity and uncertainty inherent in the real world. This paper presents PhyPlan, a novel physics-informed planning framework that combines physics-informed neural networks (PINNs) with modified Monte Carlo Tree Search (MCTS) to enable embodied agents to perform dynamic physical tasks. PhyPlan leverages PINNs to simulate and predict outcomes of actions in a fast and accurate manner and uses MCTS for planning. It dynamically determines whether to consult a PINN-based simulator (coarse but fast) or engage directly with the actual environment (fine but slow) to determine optimal policy. Given an unseen task, PhyPlan can infer the sequence of actions and learn the latent parameters, resulting in a generalizable approach that can rapidly learn to perform novel physical tasks. Evaluation with robots in simulated 3D environments demonstrates the ability of our approach to solve 3D-physical reasoning tasks involving the composition of dynamic skills. Quantitatively, PhyPlan excels in several aspects: (i) it achieves lower regret when learning novel tasks compared to the state-of-the-art, (ii) it expedites skill learning and enhances the speed of physical reasoning, (iii) it demonstrates higher data efficiency compared to a physics un-informed approach.