基于改进DPSO非退出故障下多无人机任务规划

针对非退出故障下多无人机协同任务规划问题，提出了一种基于混合策略改进的离散粒子群算法。该方法首先采用Sobol序列进行种群初始化，提高解空间的覆盖率；然后，提出非线性时变策略，加快算法的收敛速度；并引入柯西算子,增强离散粒子群算法的搜索空间；同时，还提出自适应交叉学习策略，丰富种群多样性，进而提升算法的全局寻优能力。综合改进的离散粒子群算法不仅加快了收敛速度，并且解的最优性也得到了提高。此外，运用三次样条插值算法进行无人机航迹规划，最后，将改进算法在三维空间中进行无人机故障前后的对比仿真实验，结果表明所设计的算法具有显著的寻优有效性，为部分无人机发生轻微故障后，多机协同执行任务规划的问题提供了理论依据。

Improved DPSO based mission planning for UAVS with minor fault

A discrete particle swarm optimization (DPSO) algorithm based on hybrid strategy is proposed to solve the problem of multi UAV cooperative task planning with minor failures. Firstly, Sobol sequence is used to initialize the population to improve the coverage of solution space; Then, a nonlinear time-varying strategy is proposed to accelerate the convergence of the algorithm; The Cauchy operator is introduced to enhance the search space of the discrete particle swarm optimization algorithm; At the same time, an adaptive cross learning strategy is proposed to enrich the diversity of the population, thereby improving the global optimization ability of the algorithm. The improved discrete particle swarm optimization algorithm not only speeds up the convergence, but also improves the optimality of the solution. In addition, the cubic spline interpolation algorithm is used to plan the UAV path. Finally, the improved algorithm is used in the three-dimensional space for the comparison simulation before and after the UAV failure. The results show that the designed algorithm has significant optimization effectiveness, which provides a theoretical basis for the problem of UAV minor failures and multi-UAV collaborative task planning.