CAPE: Corrective Actions from Precondition Errors using Large Language Models
This addresses the issue of robust failure recovery in LLM-driven robot planning, enabling more reliable task execution, though it is incremental over existing methods.
The paper tackles the problem of LLM-based planning failing to recover from action failures by proposing CAPE, which uses few-shot reasoning from action preconditions to propose corrective actions, improving plan correctness from 28.89% to 49.63% in VirtualHome and by 76.49% on a Spot robot compared to SayCan.
Extracting commonsense knowledge from a large language model (LLM) offers a path to designing intelligent robots. Existing approaches that leverage LLMs for planning are unable to recover when an action fails and often resort to retrying failed actions, without resolving the error's underlying cause. We propose a novel approach (CAPE) that attempts to propose corrective actions to resolve precondition errors during planning. CAPE improves the quality of generated plans by leveraging few-shot reasoning from action preconditions. Our approach enables embodied agents to execute more tasks than baseline methods while ensuring semantic correctness and minimizing re-prompting. In VirtualHome, CAPE generates executable plans while improving a human-annotated plan correctness metric from 28.89% to 49.63% over SayCan. Our improvements transfer to a Boston Dynamics Spot robot initialized with a set of skills (specified in language) and associated preconditions, where CAPE improves the correctness metric of the executed task plans by 76.49% compared to SayCan. Our approach enables the robot to follow natural language commands and robustly recover from failures, which baseline approaches largely cannot resolve or address inefficiently.