LLM-based Robot Task Planning with Exceptional Handling for General Purpose Service Robots

TL;DR

The paper addresses the challenge of translating natural-language commands into executable action sequences for general-purpose service robots while mitigating LLM hallucinations. It introduces a constrained LLM prompt scheme that confines outputs to eight predefined primitive actions and represents plans as executable Pythonic lists that feed into a SMACH state machine, coupled with an exceptional-handling module to correct format or action violations. Empirical results show high plan-decomposition accuracy (83%) and substantial full executability (69%), with real-world demonstrations on a TRACER MINI robot and cross-LLM comparisons highlighting ERNIE-Bot 4.0’s superior performance in decomposition. The work advances practical, robust robot task planning and provides a framework for reducing hallucinations, with potential impact on service robots' reliability in daily scenarios.

Abstract

The development of a general purpose service robot for daily life necessitates the robot's ability to deploy a myriad of fundamental behaviors judiciously. Recent advancements in training Large Language Models (LLMs) can be used to generate action sequences directly, given an instruction in natural language with no additional domain information. However, while the outputs of LLMs are semantically correct, the generated task plans may not accurately map to acceptable actions and might encompass various linguistic ambiguities. LLM hallucinations pose another challenge for robot task planning, which results in content that is inconsistent with real-world facts or user inputs. In this paper, we propose a task planning method based on a constrained LLM prompt scheme, which can generate an executable action sequence from a command. An exceptional handling module is further proposed to deal with LLM hallucinations problem. This module can ensure the LLM-generated results are admissible in the current environment. We evaluate our method on the commands generated by the RoboCup@Home Command Generator, observing that the robot demonstrates exceptional performance in both comprehending instructions and executing tasks.

Figures (5)

• Figure 1: The work process of our prompt scheme.
• Figure 2: An overview of our method.
• Figure 3: Flowchart of "look for person".
• Figure 4: The typical commands and test results.
• Figure 5: The process of executing commands in spoken language.