一种巡检无人机航迹自适应规划算法

A Adaptive Planning Algorithm for the Trajectory of Patrol Unmanned Aerial Vehicle

关键词:  巡检无人机  航迹规划  高斯云  α稳定分布  白鲸优化器

Key words:  patrol UAV  trajectory planning  Gaussian cloud  α stable distribution  Beluga Whale Optimizer (BWO)

摘要:

针对传统方法求解复杂环境中巡检无人机航迹规划容易出现精度差、收敛于局部最优航迹的不足，提出一种自适应高斯云与α稳定分布变异白鲸优化器(G-α-BWO)的巡检无人机航迹规划算法。该算法引入分段线性混沌映射(Piecewise Linear Chaotic Map,PWLCM)-准对立学习初始化策略提升初始航迹解的多样性和质量，利用自适应高斯云策略提升算法对全局最优航迹的搜索能力，利用自适应α稳定分布变异策略使算法跳离局部最优航迹，丰富搜索空间并提高收敛速度。首先构建巡检无人机航迹规划模型，设计考虑航程大小、高度、转角及障碍物威胁的权重代价函数，将三维航迹规划问题转换为约束条件多目标优化问题，然后通过G-α-BWO迭代搜索巡检无人机最优航迹。结果表明，改进算法搜索航迹可安全避障，且代价更小，有助于提高巡检任务效率和安全性。

Abstract:

The conventional methods have poor accuracy and tend to converge to a local optimal trajectory in the trajectory planning for Unmanned Aerial Vehicle (UAV) in complex environments.In view of these problems,an adaptive Gaussian cloud and α stable distribution variation Beluga Whale Optimizer (G-α-BWO) algorithm for tracking UAV is proposed.Piecewise Linear Chaotic Map (PWLCM)-based quasi-oppositional learning initialization is introduced to improve the diversity and quality of the initial trajectory solutions.An adaptive Gaussian cloud strategy is used to improve the search ability of the algorithm for the global optimal trajectory.An adaptive α stable distribution variation is designed to make the algorithm skip a local optimal trajectory,enrich the search space and improve the convergence speed.Firstly,the trajectory planning model is constructed for patrol UAV,and the weight cost function considering the flight range,height,angle and obstacle threat is designed.The three-dimensional trajectory planning problem is transformed into a multi-objective optimization problem with constraints.Then,the optimal trajectory of patrol UAV was iteratively searched by G-α-BWO.The results showed that the improved algorithm avoided obstacles and had less cost,being helpful to improve the efficiency and security of patrol tasks.