Body-Grounded Replanning for Physically Adaptive Manipulation

2026-09-24

AI Summary

This study addresses the limitation that robotic internal physical states are typically confined to low-level control, hindering high-level policy adaptation to payload and mobility variations. To overcome this, we propose a body-state-aware high-level replanning framework that innovatively incorporates joint-level physical states into high-level decision-making. By leveraging large language models integrated with execution history statistics, the approach dynamically interprets joint states and adaptively adjusts manipulation strategies without altering task objectives or low-level controllers. Experimental results demonstrate that the proposed method significantly reduces energy consumption while maintaining high task success rates, effectively enhancing policy adaptation efficiency. This work establishes a novel paradigm for robust decision-making in embodied intelligent systems.

Abstract

Manipulation requires not only reasoning about the external environment, but also about the robot's physical condition. A strategy may remain geometrically feasible while becoming physically unsuitable due to increased joint load or limited mobility, yet internal physical state is typically used only for low-level control. We propose body-grounded high-level replanning, which uses internal physical state to adapt manipulation strategies during execution. Body-state events trigger strategy replanning, and an LLM interprets the underlying joint-level state, recent execution statistics, and execution history to select a context-dependent alternative, while leaving the task objective and low-level controller unchanged. We evaluate the framework on a reaching task under controlled load and asymmetric mobility constraints in simulation and on a real robot. Our experiments show that body-grounded replanning maintains high task success while reducing physical effort and enabling more efficient strategy adaptation. Additional contact-rich manipulation experiments demonstrate the applicability of the same replanning interface beyond reaching. These results show that internal physical state can inform not only low-level control, but also high-level decisions about how a manipulation task should be performed.

Problem

• manipulation planning
• physical state awareness
• adaptive replanning
• robot body grounding
• high-level decision making

Innovation

• Body-grounded replanning
• Physically adaptive manipulation
• Large Language Model (LLM)
• Internal physical state
• Strategy adaptation