LLM-Driven Self-Refinement for Embodied Drone Task Planning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Deyu, Zhang, Xicheng, Li, Jiahao, Long, Tingting, Dai, Xunhua, Fu, Yongjian, Zhang, Jinrui, Ren, Ju, Zhang, Yaoxue
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911114150805504
author Zhang, Deyu
Zhang, Xicheng
Li, Jiahao
Long, Tingting
Dai, Xunhua
Fu, Yongjian
Zhang, Jinrui
Ren, Ju
Zhang, Yaoxue
author_facet Zhang, Deyu
Zhang, Xicheng
Li, Jiahao
Long, Tingting
Dai, Xunhua
Fu, Yongjian
Zhang, Jinrui
Ren, Ju
Zhang, Yaoxue
contents We introduce SRDrone, a novel system designed for self-refinement task planning in industrial-grade embodied drones. SRDrone incorporates two key technical contributions: First, it employs a continuous state evaluation methodology to robustly and accurately determine task outcomes and provide explanatory feedback. This approach supersedes conventional reliance on single-frame final-state assessment for continuous, dynamic drone operations. Second, SRDrone implements a hierarchical Behavior Tree (BT) modification model. This model integrates multi-level BT plan analysis with a constrained strategy space to enable structured reflective learning from experience. Experimental results demonstrate that SRDrone achieves a 44.87% improvement in Success Rate (SR) over baseline methods. Furthermore, real-world deployment utilizing an experience base optimized through iterative self-refinement attains a 96.25% SR. By embedding adaptive task refinement capabilities within an industrial-grade BT planning framework, SRDrone effectively integrates the general reasoning intelligence of Large Language Models (LLMs) with the stringent physical execution constraints inherent to embodied drones. Code is available at https://github.com/ZXiiiC/SRDrone.
format Preprint
id arxiv_https___arxiv_org_abs_2508_15501
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLM-Driven Self-Refinement for Embodied Drone Task Planning
Zhang, Deyu
Zhang, Xicheng
Li, Jiahao
Long, Tingting
Dai, Xunhua
Fu, Yongjian
Zhang, Jinrui
Ren, Ju
Zhang, Yaoxue
Robotics
Artificial Intelligence
We introduce SRDrone, a novel system designed for self-refinement task planning in industrial-grade embodied drones. SRDrone incorporates two key technical contributions: First, it employs a continuous state evaluation methodology to robustly and accurately determine task outcomes and provide explanatory feedback. This approach supersedes conventional reliance on single-frame final-state assessment for continuous, dynamic drone operations. Second, SRDrone implements a hierarchical Behavior Tree (BT) modification model. This model integrates multi-level BT plan analysis with a constrained strategy space to enable structured reflective learning from experience. Experimental results demonstrate that SRDrone achieves a 44.87% improvement in Success Rate (SR) over baseline methods. Furthermore, real-world deployment utilizing an experience base optimized through iterative self-refinement attains a 96.25% SR. By embedding adaptive task refinement capabilities within an industrial-grade BT planning framework, SRDrone effectively integrates the general reasoning intelligence of Large Language Models (LLMs) with the stringent physical execution constraints inherent to embodied drones. Code is available at https://github.com/ZXiiiC/SRDrone.
title LLM-Driven Self-Refinement for Embodied Drone Task Planning
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2508.15501