HiCRISP: An LLM-based Hierarchical Closed-Loop Robotic Intelligent Self-Correction Planner

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ming, Chenlin, Lin, Jiacheng, Fong, Pangkit, Wang, Han, Duan, Xiaoming, He, Jianping
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910400868515840
author Ming, Chenlin
Lin, Jiacheng
Fong, Pangkit
Wang, Han
Duan, Xiaoming
He, Jianping
author_facet Ming, Chenlin
Lin, Jiacheng
Fong, Pangkit
Wang, Han
Duan, Xiaoming
He, Jianping
contents The integration of Large Language Models (LLMs) into robotics has revolutionized human-robot interactions and autonomous task planning. However, these systems are often unable to self-correct during the task execution, which hinders their adaptability in dynamic real-world environments. To address this issue, we present a Hierarchical Closed-loop Robotic Intelligent Self-correction Planner (HiCRISP), an innovative framework that enables robots to correct errors within individual steps during the task execution. HiCRISP actively monitors and adapts the task execution process, addressing both high-level planning and low-level action errors. Extensive benchmark experiments, encompassing virtual and real-world scenarios, showcase HiCRISP's exceptional performance, positioning it as a promising solution for robotic task planning with LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2309_12089
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle HiCRISP: An LLM-based Hierarchical Closed-Loop Robotic Intelligent Self-Correction Planner
Ming, Chenlin
Lin, Jiacheng
Fong, Pangkit
Wang, Han
Duan, Xiaoming
He, Jianping
Robotics
The integration of Large Language Models (LLMs) into robotics has revolutionized human-robot interactions and autonomous task planning. However, these systems are often unable to self-correct during the task execution, which hinders their adaptability in dynamic real-world environments. To address this issue, we present a Hierarchical Closed-loop Robotic Intelligent Self-correction Planner (HiCRISP), an innovative framework that enables robots to correct errors within individual steps during the task execution. HiCRISP actively monitors and adapts the task execution process, addressing both high-level planning and low-level action errors. Extensive benchmark experiments, encompassing virtual and real-world scenarios, showcase HiCRISP's exceptional performance, positioning it as a promising solution for robotic task planning with LLMs.
title HiCRISP: An LLM-based Hierarchical Closed-Loop Robotic Intelligent Self-Correction Planner
topic Robotics
url https://arxiv.org/abs/2309.12089