Robot motion planning sees collision risk beyond simple hits
CaSCo: Cascade-Aware Soft-Collision Motion Planning
Robotics
Summary
Most robots treat collisions as a simple yes-or-no problem, but some things you bump into are more fragile or risky than others. The authors created CaSCo, a system that uses a language model to understand how risky different objects are and a physics simulator to predict what happens if the robot moves and pushes objects around. This helps the robot plan paths that avoid knocking over or damaging important or fragile items, not just avoiding any contact. The system also tracks how objects might shift after collisions, making planning smarter in cluttered spaces.
What this means in practice
- •For robotics engineers: Plan robot movements that minimize damaging or risky object interactions in cluttered areas by predicting cascaded object collisions.
- •For automation system developers: Develop safer automation systems that understand object fragility using semantic risk to reduce damage during complex tasks.
Authors
Shivaram Kumar, Gaoyuan Liu, Yoonchang Sung
Abstract
Conventional motion planning treats collision as a binary constraint, although contact with different objects can have drastically different consequences. A robot may safely brush against a cardboard box while even minor contact with a glass, laptop, or unstable object may be undesirable. Moreover, a direct robot--object collision can move the contacted object and trigger secondary object--object collisions, making the risk of a motion depend on the physical evolution of the scene rather than only on the robot's geometric path. We present CaSCo, a cascade-aware soft-collision motion planning framework in which a vision-language or language model assigns semantic risk to objects and a physics simulator predicts the consequences of candidate robot motions. CaSCo searches for a path that minimizes the total semantic risk of the unique objects displaced either directly by the robot or indirectly through cascaded collisions. Because collisions change the environment, we augment roadmap states with the predicted object arrangement and the set of objects whose risk has already been incurred. We develop an optimal graph-search algorithm with an admissible and consistent cascade-relaxed heuristic and caching and pruning mechanisms for efficient search. Experiments in cluttered manipulation environments evaluate semantic risk, cascade reasoning, planning efficiency, and real-robot operation.