Scholarlink

<p>Path planning is a cornerstone technology for enabling the autonomous navigation and operation of Autonomous Underwater Vehicles (AUVs), directly impacting their application efficacy and scope. This paper presents a comprehensive review of the state-of-the-art in AUV path planning, categorising the methodologies into three primary classes: traditional algorithms, intelligent bionic optimisations, and machine learning-based techniques. For each category, we elucidate the fundamental principles, conduct a critical analysis of their respective strengths and limitations, and summarise prevailing improvement strategies. A key insight from our study is the contextual suitability of different methods; traditional planners excel in static, known environments, while intelligent bionic and machine learning methods offer robust solutions for dynamic and uncertain underwater realms. Furthermore, we provide a comparative discussion of these approaches, highlighting their applicability in global versus local planning contexts. Finally, the review concludes by outlining emergent trends and prospective research directions, with a particular emphasis on the integration of artificial intelligence, multi-AUV collaboration, and enhanced environmental perception for future AUV systems.</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 review of autonomous underwater vehicle path planning methods

  • 期刊论文

  • 发表日期：

    2026年2月2日

查看全文
  • Ye Dai,
  • Shichen Zhou

摘要

Path planning is a cornerstone technology for enabling the autonomous navigation and operation of Autonomous Underwater Vehicles (AUVs), directly impacting their application efficacy and scope. This paper presents a comprehensive review of the state-of-the-art in AUV path planning, categorising the methodologies into three primary classes: traditional algorithms, intelligent bionic optimisations, and machine learning-based techniques. For each category, we elucidate the fundamental principles, conduct a critical analysis of their respective strengths and limitations, and summarise prevailing improvement strategies. A key insight from our study is the contextual suitability of different methods; traditional planners excel in static, known environments, while intelligent bionic and machine learning methods offer robust solutions for dynamic and uncertain underwater realms. Furthermore, we provide a comparative discussion of these approaches, highlighting their applicability in global versus local planning contexts. Finally, the review concludes by outlining emergent trends and prospective research directions, with a particular emphasis on the integration of artificial intelligence, multi-AUV collaboration, and enhanced environmental perception for future AUV systems.