Scholarlink

<p>A collaborative multi-UAV path planning algorithm based on a multi-strategy particle swarm optimization algorithm is proposed for the task of multi-UAV cooperative operations in complex 3D environments. The algorithm takes into account flight efficiency, threat exposure, trajectory smoothness, and inter-UAV collision avoidance to ensure safe and coordinated mission execution. Firstly, a clearance-guided corridor initialization strategy is developed to generate a diverse and constraint-satisfying initial population, which enhances early-stage search stability. Next, a dual-layer search mechanism is introduced, in which the upper layer adopts Enhanced Genetic Multi-Level Learning (EGML) to promote global exploration and the lower layer employs Enhanced Laplacian Acceleration (ELA) to accelerate local convergence. Finally, an elite segment repair operator is applied to detect and resolve inter-UAV path conflicts, ensuring safe separation without compromising solution continuity. Experimental results demonstrate that MSPSO outperforms several existing constrained optimization algorithms in terms of convergence rate, solution accuracy, and overall robustness. MSPSO enables multi-UAV to execute collaborative missions more efficiently, safely, and smoothly without violating kinematic or inter-vehicle safety constraints, while delivering a high-quality flight plan.</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

MSPSO: a multi-strategy particle swarm optimization for safe and efficient collaborative multi-UAV path planning in complex 3D environments

  • 期刊论文

  • 发表日期：

    2026年6月20日

查看全文
  • Chuanyun Wang,
  • Xipei Chen,
  • Huilong Zheng,
  • Qian Gao,
  • Linlin Wang

摘要

A collaborative multi-UAV path planning algorithm based on a multi-strategy particle swarm optimization algorithm is proposed for the task of multi-UAV cooperative operations in complex 3D environments. The algorithm takes into account flight efficiency, threat exposure, trajectory smoothness, and inter-UAV collision avoidance to ensure safe and coordinated mission execution. Firstly, a clearance-guided corridor initialization strategy is developed to generate a diverse and constraint-satisfying initial population, which enhances early-stage search stability. Next, a dual-layer search mechanism is introduced, in which the upper layer adopts Enhanced Genetic Multi-Level Learning (EGML) to promote global exploration and the lower layer employs Enhanced Laplacian Acceleration (ELA) to accelerate local convergence. Finally, an elite segment repair operator is applied to detect and resolve inter-UAV path conflicts, ensuring safe separation without compromising solution continuity. Experimental results demonstrate that MSPSO outperforms several existing constrained optimization algorithms in terms of convergence rate, solution accuracy, and overall robustness. MSPSO enables multi-UAV to execute collaborative missions more efficiently, safely, and smoothly without violating kinematic or inter-vehicle safety constraints, while delivering a high-quality flight plan.