Distilling Contact Planning for Fast Trajectory Optimization in Robot Air Hockey

TL;DR

This paper presents a hybrid approach combining distillation of stochastic optimal control policies with online model predictive control to enable fast, accurate, and constrained contact planning for robot air hockey, outperforming existing methods.

Contribution

It introduces a novel framework that integrates high-level contact planning via policy distillation with low-level constrained motion control for real-time robot air hockey.

Findings

Outperforms control-based methods in accuracy and speed.

Effective in leveraging bank shots and robot kinematics.

Validated in both simulation and real-world experiments.

Abstract

Robot control through contact is challenging as it requires reasoning over long horizons and discontinuous system dynamics. Highly dynamic tasks such as Air Hockey additionally require agile behavior, making the corresponding optimal control problems intractable for planning in realtime. Learning-based approaches address this issue by shifting computationally expensive reasoning through contacts to an offline learning phase. However, learning low-level motor policies subject to kinematic and dynamic constraints can be challenging if operating in proximity to such constraints is desired. This paper explores the combination of distilling a stochastic optimal control policy for high-level contact planning and online model-predictive control for low-level constrained motion planning. Our system learns to balance shooting accuracy and resulting puck speed by leveraging bank shots and the robot's kinematic structure. We show that the proposed framework outperforms purely control-based and purely learning-based techniques in both simulated and real-world games of Robot Air Hockey.