PhysCaP: Grounding Code-as-Policy Agent with Physics-Informed Exploration

2026-08-21

AI Summary

为解决机器人操作中物理属性感知问题，PhysCaP通过引入基于物理信息的探索层和无需训练的物理属性提取模块，实现高效主动感知。

Abstract

We present PhysCaP, a Physics-Informed Code-as-Policy agent for active perception in robotic manipulation. While vision-language-action policies excel at imitating demonstrations, they rely on passive observation and fail to infer latent physical properties critical for manipulation. PhysCaP augments code-as-policy frameworks with a physics-informed exploration layer that enables explicit information-seeking through interaction. It introduces training-free physical property extraction modules that estimate object mass and stiffness from robot proprioception without additional sensors. To balance exploration costs and the efficiency of information obtained, PhysCaP employs a dual-agent design: a Planner that decides when to explore and when to stop, and a Prioritizer that filters implausible interactions and ranks the remainder using a heuristic priority score, enabling efficient, targeted exploration. We evaluate PhysCaP on real-world tabletop manipulation tasks (searching for hidden objects, detecting empty cans, and finding ripe avocados) and a simulated task in LIBERO. The results show that existing passive and naive interactive baselines either fail when physical properties are hidden or over-explore, whereas PhysCaP achieves comparable performance with fewer interactions and reduced execution time. Ablation studies further validate the effectiveness of the proposed physical property extraction modules. Project page: https://physcap.github.io

Problem

• Physics-Informed
• Code-as-Policy
• Active Perception
• Physical Properties
• Robotic Manipulation

Innovation

• Physics-Informed Exploration
• Code-as-Policy
• Physical Property Extraction
• Dual-Agent Design