Scholarlink

<p>To improve safety of autonomous underwater vehicle operates in the ocean current environment, a framework for multiple populations for multiple objectives (MPMO) based on ant colony optimization (ACO), called MPMO-ACO, is proposed. MPMO-ACO uses three populations to optimize path length, turning amount and energy consumption. Firstly, different local pheromone update modes are designed in different iteration stages of MPMO-ACO to increase the global search ability. Secondly, by biasedly focusing on the optimization objectives of the corresponding population, the biased sorting (BS) method is used for each population to enhance the effectiveness of non-dominated sorting (NDS). Thirdly, store all solutions in the Archive, perform NDS, and select solutions with larger crowding distances. Meanwhile, to share information among populations, the pheromone matrix is merged for global pheromone update. Experiments were conducted on three-dimensional ocean environment maps of different scales. The results indicate that MPMO-ACO demonstrates superior diversity and convergence efficiency.</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

Autonomous underwater vehicle 3D path planning based on multiple populations for multiple objectives ant colony optimization

  • 期刊论文

  • 发表日期：

    2025年9月19日

查看全文
  • Zhaojun Zhang,
  • Shun Lu,
  • Jiawei Lu,
  • Simeng Tan,
  • Kuansheng Zou

摘要

To improve safety of autonomous underwater vehicle operates in the ocean current environment, a framework for multiple populations for multiple objectives (MPMO) based on ant colony optimization (ACO), called MPMO-ACO, is proposed. MPMO-ACO uses three populations to optimize path length, turning amount and energy consumption. Firstly, different local pheromone update modes are designed in different iteration stages of MPMO-ACO to increase the global search ability. Secondly, by biasedly focusing on the optimization objectives of the corresponding population, the biased sorting (BS) method is used for each population to enhance the effectiveness of non-dominated sorting (NDS). Thirdly, store all solutions in the Archive, perform NDS, and select solutions with larger crowding distances. Meanwhile, to share information among populations, the pheromone matrix is merged for global pheromone update. Experiments were conducted on three-dimensional ocean environment maps of different scales. The results indicate that MPMO-ACO demonstrates superior diversity and convergence efficiency.