Scholarlink

<p>This paper presents a novel multi-robot coverage path planning (MCPP) algorithm focusing on large-scale indoor closed obstacle-constrained area. We enhance traditional spanning tree coverage (STC) algorithm with backtracking for energy saving by implementing back-end optimization based on chaotic mapping to reduce the number of turns. The proposed method simplifies the non-deterministic polynomial hard (NP-hard) problem into an NP problem by stochastic optimization method, ensuring that the computational complexity remains manageable in a large-scale environment. To substantiate the feasibility of the proposed method, we have developed a technical framework for algorithm deployment and conducted real-world experiments on a custom platform.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A multi-UGV coverage path planning method with reduced the number of turns for structured obstacle-enclosed area

  • 期刊论文

  • 发表日期：

    2026年1月5日

查看全文
  • Chao Wang,
  • Wei Dong,
  • Huajian Liu,
  • Hui Dong,
  • Yanlong Yang,
  • Renjie Li,
  • Yongzhuo Gao

摘要

This paper presents a novel multi-robot coverage path planning (MCPP) algorithm focusing on large-scale indoor closed obstacle-constrained area. We enhance traditional spanning tree coverage (STC) algorithm with backtracking for energy saving by implementing back-end optimization based on chaotic mapping to reduce the number of turns. The proposed method simplifies the non-deterministic polynomial hard (NP-hard) problem into an NP problem by stochastic optimization method, ensuring that the computational complexity remains manageable in a large-scale environment. To substantiate the feasibility of the proposed method, we have developed a technical framework for algorithm deployment and conducted real-world experiments on a custom platform.