Scholarlink

<p>In order for unmanned aerial vehicles (UAVs) to achieve energy-saving, safe and smooth flight in a wide range of application scenarios, pre-planning of flight paths is essential. This study introduces a new UAV trajectory planning method. To address the local optimization problems, poor trajectory quality, and insufficient obstacle avoidance in complex environments in existing algorithms, a Global-enhanced Crown Porcupine Optimization (GCPO) algorithm is proposed. On one hand, the algorithm combines a dynamic reverse learning strategy to improve the quality of the initial solution, a dynamic weighting mechanism of whale spiral search and an adaptive step-size perturbation optimization mechanism to escape the local optimum, an optimal value guidance strategy to balance exploration and development, and a cross-cutting method to prevent search stagnation. On the other hand, this study develops a new cost function model that combines constraints such as trajectory length, height, obstacles, and smoothness with a trajectory discretization method to improve planning accuracy through precise control of discrete points. Finally, an advanced obstacle avoidance strategy is proposed to address both global and local planning challenges. Simulation results show that GCPO outperforms other state-of-the-art meta-heuristic optimization algorithms, including Particle Swarm Optimization (PSO), Grey Wolf Algorithm (GWO), Rime Ice Algorithm (RIME), Northern Goshawk Algorithm (NGO), and Classic Crested Porcupine Algorithm (CPO), achieving more efficient and higher quality trajectory planning with smoother, lower cost flight paths, and enhanced obstacle avoidance capabilities in complex 3D environments.</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

Three-Dimensional Trajectory Planning for Unmanned Aerial Vehicles Using an Enhanced Crowned Porcupine Optimization Algorithm

  • 期刊论文

  • 发表日期：

    2025年3月6日

查看全文
  • Xingyu Liu,
  • Li Ding,
  • Ahmed Tijani Musa,
  • Hongtao Wu

摘要

In order for unmanned aerial vehicles (UAVs) to achieve energy-saving, safe and smooth flight in a wide range of application scenarios, pre-planning of flight paths is essential. This study introduces a new UAV trajectory planning method. To address the local optimization problems, poor trajectory quality, and insufficient obstacle avoidance in complex environments in existing algorithms, a Global-enhanced Crown Porcupine Optimization (GCPO) algorithm is proposed. On one hand, the algorithm combines a dynamic reverse learning strategy to improve the quality of the initial solution, a dynamic weighting mechanism of whale spiral search and an adaptive step-size perturbation optimization mechanism to escape the local optimum, an optimal value guidance strategy to balance exploration and development, and a cross-cutting method to prevent search stagnation. On the other hand, this study develops a new cost function model that combines constraints such as trajectory length, height, obstacles, and smoothness with a trajectory discretization method to improve planning accuracy through precise control of discrete points. Finally, an advanced obstacle avoidance strategy is proposed to address both global and local planning challenges. Simulation results show that GCPO outperforms other state-of-the-art meta-heuristic optimization algorithms, including Particle Swarm Optimization (PSO), Grey Wolf Algorithm (GWO), Rime Ice Algorithm (RIME), Northern Goshawk Algorithm (NGO), and Classic Crested Porcupine Algorithm (CPO), achieving more efficient and higher quality trajectory planning with smoother, lower cost flight paths, and enhanced obstacle avoidance capabilities in complex 3D environments.