面向巡检任务的无人机三维航迹自适应规划算法

Unmanned Aerial Vehicle 3D Path Adaptive Plan Algorithm for Patrol Tasks

关键词:  巡检无人机  航迹规划  爬行动物搜索算法  混沌映射  威布尔飞行算子

Key words:  patrol unmanned aerial vehicle  path planning  reptile search algorithm  chaos map  weibull flight operator

摘要:

传统方法求解复杂环境下巡检无人机航迹规划问题容易出现收敛于局部最优航迹、全局搜索精度差的不足。为此,提出一种自适应非线性进化与威布尔飞行变异爬行动物搜索的巡检无人机航迹规划算法NWFRSA。引入改进无限折叠迭代混沌映射ICMIC初始化策略提升初始航迹解的多样性和质量,利用S型特征函数非线性进化因子实现全局最优航迹的探采均衡,并设计威布尔飞行算子变异策略使算法能跳离局部最优航迹,丰富搜索空间并提高收敛精度。构建了巡检无人机的航迹规划模型,设计了考虑航迹总长度、飞行高度、飞行转角及威胁模型的权重目标函数,将三维空间内的航迹规划转换为约束条件下的多目标优化问题,并利用改进爬行动物搜索算法NWFRSA迭代搜索无人机的最优航迹。构建了多场景下不同复杂程度的障碍物与威胁区域分布的地形图进行对比实验。结果表明,改进算法的搜索航迹不仅实现了安全避障,而且航迹代价更小,有助于提高巡检任务效率和安全性。

Abstract:

The traditional method to solve the track planning problem of inspection UAV in complex environment is prone to poor accuracy and convergence to the local optimal track. Therefore, a tracking planning algorithm NWFRSA based on adaptive nonlinear evolution and Weibull flight mutant reptiles search is proposed. An improved ICMIC initialization strategy for infinite folding iterative chaotic mapping is introduced to improve the diversity and quality of the initial flight path solutions. And the exploration and mining equilibrium of the global optimal flight path is realized by using the nonlinear evolution factor of S-type feature function. The Weibull flight operator mutation strategy is designed to make the algorithm skip form a local optimal flight path, enrich the search space and improve the convergence accuracy. The track planning model of inspection UAV is constructed, and the weight objective function of the model considering track length, flight altitude, flight angle and threat is designed. The track planning in three-dimensional space is converted into a multi-objective optimization problem under constraints, and the improved reptile search algorithm NWFRSA is used to iteratively search the optimal track of UAV. The topographic maps of the distribution of obstacles and threats with different complexity in multiple scenarios are constructed for comparative experiments. The results show that the improved algorithm not only can achieve safe obstacle avoidance, but also can reduce the cost of track, which is helpful to improve the efficiency and safety of inspection tasks.