Abstract: Path planning of UAV（unmanned aerial vehicle） is an important part of UAV mission planning system. It is necessary to obtain the optimal path in a search space with threat area. In order to solve the problems of slow convergence speed and easy to fall into local optimization, a mutation gray wolf optimizer algorithm based on A* initialization is proposed. The algorithm first discretizes the model, and then uses A* algorithm to initialize the wolf, so that the subsequent algorithm has a better starting point. Then, the simplified gray wolf optimizer algorithm is used to build and update the population on the continuous model. In the iterative process, the population is optimized by a new modified mutation operator proposed in the paper. The UAV track smoothed by cubic B-spline meets the performance requirements of UAV. The experimental results show that the algorithm is superior to PSO（particle swarm optimization）, GWO（grey wolf optimizer） and SOS（symbiotic organisms search） algorithms in cost convergence speed, final path and algorithm stability, and has high application value in solving UAV path planning problems.

Key words: unmanned aerial vehicle（UAV）, path planning, A* algorithm, gray wolf optimizer（GWO）

摘要： 无人机（unmanned aerial vehicle，UAV）路径规划问题是无人机任务规划系统的重要组成部分，需要在一个存在威胁区的搜索空间中获得最优路径。为解决灰狼优化算法存在收敛速度慢、容易陷入局部最优等问题，提出了一种基于A*初始化的变异灰狼优化算法。该算法首先将模型离散化，进而使用A*算法进行头狼的初始化，使后续算法有一个较优的起点，随后通过简化后的灰狼优化算法在连续模型上构建和更新种群，在迭代过程中，通过新提出的一种新型修正变异算子优化种群。利用三次B样条平滑后的无人机航迹，符合无人机的性能要求。经实验验证，算法在代价收敛速度、求取的最终路径以及算法稳定性方面均优于粒子群算法（particle swarm optimization，PSO）、灰狼优化算法（gray wolf optimizer，GWO）、共生生物搜索算法（symbiotic organisms search，SOS）算法，在解决无人机路径规划问题上具有较高的应用价值。

关键词: 无人机（UAV）, 路径规划, A*算法, 灰狼优化（GWO）