Abstract: For some deficiency of general path planning methods such as weak applicability, long time-cost and easily planning-fail, a new fast path planning method is proposed. A new method of multi-danger-value matrix is used as the modeling carrier. The ameliorative PSO method is used for seeking the cursory path, after that some optimization is needed to be made respectively for the cursory path. The path will become the global-best path. The simulation result indicates that the method plays well in complex circumstance. The method is universally available, fast and precise.

Key words: Particle Swarm Optimization（PSO）, respective optimization, path planning

摘要： 针对一般航迹规划算法在复杂环境下适用性差、耗时长、容易失效的问题，提出了一种复合快速航迹生成算法。该算法采用新型的多逻辑威胁值网点法为环境建模形式，在此基础上将加以改进后的粒子群算法引入到航迹近似最优解的快速求取过程中，对所得的航迹近似解进行分段局部优化以及可飞性修正处理，将近似解变为最优解。仿真结果显示该复合算法通用性好、速度快、精度高，能较好地应用于复杂环境下无人机航迹的快速求解。

关键词: 粒子群, 分段优化, 航迹规划