为解决无人机在三维环境下的路径规划问题，通过麻雀搜索算法研究了路径规划方法。传统的麻雀搜索算法求解该问题时存在易陷入局部最优、收敛精度低等问题，针对该问题提出改进方法。首先，对种群中的发现者加入动态权重因子，使其能够提高局部搜索能力，同时提高收敛速度，同时引入高斯变异；追随者采用量子粒子群生成新解的方式；并且加入额外的柯西变异进行扰动，柯西变异的扰动幅度较小，可以增强局部搜索能力。通过仿真实验，算法改进后分别与麻雀算法以及其他改进的麻雀算法进行对比，结果表明该算法收敛速度更快，求解精度更高，证明了该算法的有效性和可行性，可见在无人机三维路径规划中具有很大的发展前景。

To solve the path planning problem for unmanned aerial vehicles (UAVs) in three-dimensional environments, path planning methods through the sparrow search algorithm was investigated. Traditional sparrow search algorithms have issues such as easily falling into local optima and low convergence accuracy when solving this problem. To address these issues, an improved method was proposed. First, dynamic weight factors were added to the discoverers in the population to enhance their local search ability and increase convergence speed, while introducing Gaussian mutation. Followers used a quantum particle swarm approach to generate new solutions, and an additional Cauchy mutation was introduced for perturbation, with a smaller perturbation amplitude to enhance local search ability. Through simulation experiments, the improved algorithm was compared with the sparrow algorithm and other improved sparrow algorithms, showing that the improved algorithm has faster convergence speed and higher solution accuracy, proving the effectiveness and feasibility of the algorithm. This indicates that the algorithm has great potential for UAV three-dimensional path planning.