Scholarlink

This paper introduces the Heuristic Bidirectional Rapidly-exploring Random Tree (HB-RRT) algorithm, a 4D approach integrating three spatial dimensions and time, specifically designed for obstacle avoidance in dynamic settings. Unlike traditional RRT, RRT*, and PF-RRT, HB-RRT leverages heuristic sampling, greedy connections, and lightweight RRT* reconnection to enhance path quality and reduce computational effort, employing temporal sampling and collision prediction for safe navigation amidst moving obstacles. Simulations demonstrate HB-RRT’s superiority, reducing path length by 18% and computation time by over 90% compared to RRT, with collision-free paths validated in a UAV base station inspection scenario under dynamic conditions. Its adaptive strategy ensures enhanced responsiveness, making it suitable for dynamic environments.

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

Heuristic Bidirectional-RRT: A 4D Path Planning Solution for UAVs in Dynamic Environments

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Daxin Gong,
  • Haijin Zeng,
  • Zhiyuan Shan

摘要

This paper introduces the Heuristic Bidirectional Rapidly-exploring Random Tree (HB-RRT) algorithm, a 4D approach integrating three spatial dimensions and time, specifically designed for obstacle avoidance in dynamic settings. Unlike traditional RRT, RRT*, and PF-RRT, HB-RRT leverages heuristic sampling, greedy connections, and lightweight RRT* reconnection to enhance path quality and reduce computational effort, employing temporal sampling and collision prediction for safe navigation amidst moving obstacles. Simulations demonstrate HB-RRT’s superiority, reducing path length by 18% and computation time by over 90% compared to RRT, with collision-free paths validated in a UAV base station inspection scenario under dynamic conditions. Its adaptive strategy ensures enhanced responsiveness, making it suitable for dynamic environments.