Scholarlink

<p>Multi-unmanned aerial vehicle (multi-UAV) path planning in complex three-dimensional threat environments requires both robust constrained optimization and efficient repeated computation. To improve convergence accuracy, constraint handling, and computational efficiency, this study proposes the Meta-heuristic Enhanced Tribe Competition and Cooperation of Members algorithm (META_CTCM) based on the Competition of Tribes and Cooperation of Members with Kent chaotic and <i>t</i>-distribution mutation (CTCMKT) framework. Four mechanisms, namely a time-varying update operator, dual-phase threshold write-back, gradient nudging, and adaptive explosion, are introduced to coordinate global exploration, feasible-solution preservation, local refinement, and diversity restoration. A multi-constraint objective function is established by integrating path length, altitude deviation, turning smoothness, and comprehensive safety cost, and an eight-dimensional evaluation system is used to assess path quality, convergence behavior, stability, success rate, and computational overhead. Experimental results show that META_CTCM achieves superior overall performance compared with the direct baseline CTCMKT and other representative meta-heuristic algorithms. Compared with CTCMKT, the path-length cost, altitude-deviation cost, and turning-smoothness cost are reduced by 7.4%, 62.0%, and 94.4%, respectively, corresponding to shorter trajectories, smaller deviation from the preferred altitude band, and smoother turning behavior. The computational overhead is reduced by 94.4% relative to CTCMKT, indicating that the proposed mechanisms improve optimization quality without excessive runtime burden. Furthermore, a coarse-grained parallel scalability experiment shows that the batch runtime of 32 independent runs decreases from 670.1527 to 170.4796&#xa0;s using eight workers, achieving a speedup of 3.9310. These results demonstrate that META_CTCM provides an effective and computationally scalable solution for complex 3D multi-UAV path planning.</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

META_CTCM: a multi-constraint meta-heuristic algorithm for 3D multi-UAV path planning

  • 期刊论文

  • 发表日期：

    2026年7月2日

查看全文
  • Guoqing Yang,
  • Yongkang Gao,
  • Wanxiang Yao,
  • Kun Zhao,
  • Qinxiao Zhang,
  • Yingjie Li

摘要

Multi-unmanned aerial vehicle (multi-UAV) path planning in complex three-dimensional threat environments requires both robust constrained optimization and efficient repeated computation. To improve convergence accuracy, constraint handling, and computational efficiency, this study proposes the Meta-heuristic Enhanced Tribe Competition and Cooperation of Members algorithm (META_CTCM) based on the Competition of Tribes and Cooperation of Members with Kent chaotic and t-distribution mutation (CTCMKT) framework. Four mechanisms, namely a time-varying update operator, dual-phase threshold write-back, gradient nudging, and adaptive explosion, are introduced to coordinate global exploration, feasible-solution preservation, local refinement, and diversity restoration. A multi-constraint objective function is established by integrating path length, altitude deviation, turning smoothness, and comprehensive safety cost, and an eight-dimensional evaluation system is used to assess path quality, convergence behavior, stability, success rate, and computational overhead. Experimental results show that META_CTCM achieves superior overall performance compared with the direct baseline CTCMKT and other representative meta-heuristic algorithms. Compared with CTCMKT, the path-length cost, altitude-deviation cost, and turning-smoothness cost are reduced by 7.4%, 62.0%, and 94.4%, respectively, corresponding to shorter trajectories, smaller deviation from the preferred altitude band, and smoother turning behavior. The computational overhead is reduced by 94.4% relative to CTCMKT, indicating that the proposed mechanisms improve optimization quality without excessive runtime burden. Furthermore, a coarse-grained parallel scalability experiment shows that the batch runtime of 32 independent runs decreases from 670.1527 to 170.4796 s using eight workers, achieving a speedup of 3.9310. These results demonstrate that META_CTCM provides an effective and computationally scalable solution for complex 3D multi-UAV path planning.