Skypilot: Fine-Tuning LLM with Physical Grounding for AAV Coverage Search

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Zhongkai, Sun, Yihao, Yan, Chao, Zhou, Han, Xiang, Xiaojia, Jiang, Jie
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917099882938368
author Chen, Zhongkai
Sun, Yihao
Yan, Chao
Zhou, Han
Xiang, Xiaojia
Jiang, Jie
author_facet Chen, Zhongkai
Sun, Yihao
Yan, Chao
Zhou, Han
Xiang, Xiaojia
Jiang, Jie
contents Autonomous aerial vehicles (AAVs) have played a pivotal role in coverage operations and search missions. Recent advances in large language models (LLMs) offer promising opportunities to augment AAV intelligence. These advances help address complex challenges like area coverage optimization, dynamic path planning, and adaptive decision-making. However, the absence of physical grounding in LLMs leads to hallucination and reproducibility problems in spatial reasoning and decision-making. To tackle these issues, we present Skypilot, an LLM-enhanced two-stage framework that grounds language models in physical reality by integrating monte carlo tree search (MCTS). In the first stage, we introduce a diversified action space that encompasses generate, regenerate, fine-tune, and evaluate operations, coupled with physics-informed reward functions to ensure trajectory feasibility. In the second stage, we fine-tune Qwen3-4B on 23,000 MCTS-generated samples, achieving substantial inference acceleration while maintaining solution quality. Extensive numerical simulations and real-world flight experiments validate the efficiency and superiority of our proposed approach. Detailed information and experimental results are accessible at https://sky-pilot.top.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18270
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Skypilot: Fine-Tuning LLM with Physical Grounding for AAV Coverage Search
Chen, Zhongkai
Sun, Yihao
Yan, Chao
Zhou, Han
Xiang, Xiaojia
Jiang, Jie
Robotics
Autonomous aerial vehicles (AAVs) have played a pivotal role in coverage operations and search missions. Recent advances in large language models (LLMs) offer promising opportunities to augment AAV intelligence. These advances help address complex challenges like area coverage optimization, dynamic path planning, and adaptive decision-making. However, the absence of physical grounding in LLMs leads to hallucination and reproducibility problems in spatial reasoning and decision-making. To tackle these issues, we present Skypilot, an LLM-enhanced two-stage framework that grounds language models in physical reality by integrating monte carlo tree search (MCTS). In the first stage, we introduce a diversified action space that encompasses generate, regenerate, fine-tune, and evaluate operations, coupled with physics-informed reward functions to ensure trajectory feasibility. In the second stage, we fine-tune Qwen3-4B on 23,000 MCTS-generated samples, achieving substantial inference acceleration while maintaining solution quality. Extensive numerical simulations and real-world flight experiments validate the efficiency and superiority of our proposed approach. Detailed information and experimental results are accessible at https://sky-pilot.top.
title Skypilot: Fine-Tuning LLM with Physical Grounding for AAV Coverage Search
topic Robotics
url https://arxiv.org/abs/2511.18270