Understanding Physical Effects for Effective Tool-use

TL;DR

This paper introduces a robot learning framework that uses FEM simulation and symbolic regression to develop effective tool-use strategies with minimal effort, adaptable to new objects and physical effects.

Contribution

It combines FEM-based simulation, symbolic regression, and optimal control to learn and plan effective tool-use strategies that generalize beyond training data.

Findings

Framework produces more effective tool-use strategies in simulation.

Strategies differ significantly from observed behaviors, showing adaptability.

Demonstrates effectiveness across two exemplar tasks.

Abstract

We present a robot learning and planning framework that produces an effective tool-use strategy with the least joint efforts, capable of handling objects different from training. Leveraging a Finite Element Method (FEM)-based simulator that reproduces fine-grained, continuous visual and physical effects given observed tool-use events, the essential physical properties contributing to the effects are identified through the proposed Iterative Deepening Symbolic Regression (IDSR) algorithm. We further devise an optimal control-based motion planning scheme to integrate robot- and tool-specific kinematics and dynamics to produce an effective trajectory that enacts the learned properties. In simulation, we demonstrate that the proposed framework can produce more effective tool-use strategies, drastically different from the observed ones in two exemplar tasks.