Physics-Informed Eikonal Caging for Whole-Arm Manipulation Planning

2026-06-20

AI Summary

This work addresses the challenge of insufficient planning robustness in whole-arm manipulation due to difficulties in modeling contact dynamics. The authors reformulate caging as a shortest-escape-time problem, wherein an object seeks the minimal time to escape the robot’s grasp configuration. They construct an escape-time field satisfying the eikonal equation and represent it as a continuous, differentiable function using physics-informed neural networks. This approach is the first to cast geometric caging into a differentiable optimization primitive, enabling seamless integration into gradient-based whole-arm manipulation planners. As a result, the method significantly enhances robustness against contact model inaccuracies and external disturbances. Simulations and real-world experiments demonstrate superior performance over existing baselines under both perturbations and model mismatch, validating the escape-time field as an effective primitive for robust manipulation.

Abstract

Planning contact-rich whole-arm manipulation is challenging because interactions that involve extended robot geometry give rise to complex contact dynamics that are difficult to model accurately. This creates a need for planning principles that do not rely heavily on precise contact models. Caging offers one such geometric notion of robustness to modeling inaccuracy by restricting object escape through geometrically enclosing the object. However, existing caging formulations are difficult to incorporate into continuous optimization-based manipulation planning. We reformulate caging as a minimum-time escape problem in which the object seeks to leave an enclosing robot geometry in the shortest time. This yields a continuous escape-time field that measures the robot's enclosure quality and we show it satisfies an eikonal equation. We therefore can approximate this field using a physics-informed neural network, producing a smooth differentiable representation that can be embedded directly into manipulation planning. The resulting objective supports whole-arm manipulation planning to favor robot configurations resisting object escape. This improves the manipulation robustness to contact model mismatch, thus enabling planning with simplified contact models, including quasi-dynamic approximations and simplified object geometry. Across simulation and real-world experiments, we show improved robustness to disturbances and contact-model mismatch relative to baselines. These results suggest that geometric enclosure can serve as a practical robustness primitive for whole-arm manipulation. A supplementary video, which includes an intuitive overview of our method and experiment video results, is available on our project webpage.

Problem

• whole-arm manipulation
• caging
• contact-rich planning
• modeling inaccuracy
• manipulation robustness

Innovation

• Physics-Informed Neural Networks
• Eikonal Equation
• Caging
• Whole-Arm Manipulation
• Contact-Rich Planning