Safe Planner: Empowering Safety Awareness in Large Pre-Trained Models for Robot Task Planning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Siyuan, Ma, Zhe, Liu, Feifan, Lu, Jiani, Xiao, Qinqin, Sun, Kewu, Cui, Lingfei, Yang, Xirui, Liu, Peng, Wang, Xun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929587275956224
author Li, Siyuan
Ma, Zhe
Liu, Feifan
Lu, Jiani
Xiao, Qinqin
Sun, Kewu
Cui, Lingfei
Yang, Xirui
Liu, Peng
Wang, Xun
author_facet Li, Siyuan
Ma, Zhe
Liu, Feifan
Lu, Jiani
Xiao, Qinqin
Sun, Kewu
Cui, Lingfei
Yang, Xirui
Liu, Peng
Wang, Xun
contents Robot task planning is an important problem for autonomous robots in long-horizon challenging tasks. As large pre-trained models have demonstrated superior planning ability, recent research investigates utilizing large models to achieve autonomous planning for robots in diverse tasks. However, since the large models are pre-trained with Internet data and lack the knowledge of real task scenes, large models as planners may make unsafe decisions that hurt the robots and the surrounding environments. To solve this challenge, we propose a novel Safe Planner framework, which empowers safety awareness in large pre-trained models to accomplish safe and executable planning. In this framework, we develop a safety prediction module to guide the high-level large model planner, and this safety module trained in a simulator can be effectively transferred to real-world tasks. The proposed Safe Planner framework is evaluated on both simulated environments and real robots. The experiment results demonstrate that Safe Planner not only achieves state-of-the-art task success rates, but also substantially improves safety during task execution. The experiment videos are shown in https://sites.google.com/view/safeplanner .
format Preprint
id arxiv_https___arxiv_org_abs_2411_06920
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Safe Planner: Empowering Safety Awareness in Large Pre-Trained Models for Robot Task Planning
Li, Siyuan
Ma, Zhe
Liu, Feifan
Lu, Jiani
Xiao, Qinqin
Sun, Kewu
Cui, Lingfei
Yang, Xirui
Liu, Peng
Wang, Xun
Robotics
Robot task planning is an important problem for autonomous robots in long-horizon challenging tasks. As large pre-trained models have demonstrated superior planning ability, recent research investigates utilizing large models to achieve autonomous planning for robots in diverse tasks. However, since the large models are pre-trained with Internet data and lack the knowledge of real task scenes, large models as planners may make unsafe decisions that hurt the robots and the surrounding environments. To solve this challenge, we propose a novel Safe Planner framework, which empowers safety awareness in large pre-trained models to accomplish safe and executable planning. In this framework, we develop a safety prediction module to guide the high-level large model planner, and this safety module trained in a simulator can be effectively transferred to real-world tasks. The proposed Safe Planner framework is evaluated on both simulated environments and real robots. The experiment results demonstrate that Safe Planner not only achieves state-of-the-art task success rates, but also substantially improves safety during task execution. The experiment videos are shown in https://sites.google.com/view/safeplanner .
title Safe Planner: Empowering Safety Awareness in Large Pre-Trained Models for Robot Task Planning
topic Robotics
url https://arxiv.org/abs/2411.06920