【摘要】 无人机群航路规划的主要目标是为每架无人机生成一条连接起始点和目标点并满足约束条件的可飞行路径。传统方法主要以代价地图为基础，采用动态规划与几何运算等方法解决无人机群的路径规划问题。然而，这些方法难以解决无人机群的几何、物理与时间等复杂多约束问题，而且代价地图的构建十分耗时，使得无人机群难以应对复杂多变的实际环境。近年来，群体智能技术的出现为无人机群路径规划提供了新思路，该技术不但能够解决无人机一维静态的路径优化问题，同时为多维动态路径的优化提供更加精准、快速、有效的智能解决方案。此外，机载智能硬件的飞速发展也极大提升了无人机群的通信速率与运算效率，使得无人机群具备环境适应能力强、部署灵活和多功能集成等优势。基于以上背景，阐述了无人机群航路规划的环境感知建模、适应度函数、约束条件和障碍物规避四方面内容；综述了群体智能算法原理，并分析讨论了5种群体智能算法以及其应用于无人机群航路规划的优缺点；分析和展望了群体智能算法应用于无人机群航路规划与任务协同方向的发展趋势。更多还原

【Abstract】 UAV swarm path planning is to find a flight path from the starting point to the target point under complex constraints. Previous studies based on cost map mainly use dynamic programming and geometric algorithms for UAV path planning. However, it is difficult for these methods to solve the geometric, physical and time constraints of UAV swarm and the construction of cost maps is very time-consuming, which makes the UAV swarm unable to flexibly cope with complex and changing environments. In recent years, swarm intelligence methods have provided a new opportunity for UAV swarm path planning, which not only solves the one-dimensional static path optimization problem, but also provides more accurate, fast and effective multi-dimensional dynamic path optimization. In addition, there are also many breakthroughs in the hardware implementation of bionic intelligence methods, which greatly improves the communication speed and efficiency of UAV swarms and provides advantages such as strong environmental adaptability, flexible deployment, and functional integration for UAV swarms. The 3D environment perception modeling, fitness function, constraint function and obstacle avoiding of UAV swarm path planning are introduced. The principle of UAV swarm intelligence algorithm is summarized in detail, and five commonly used algorithms as well as their advantages and disadvantages in UAV swarm path planning are analyzed. Finally, the future research trend of swarm intelligence algorithm for UAV swarm path planning is discussed.更多还原