四川轻化工大学学报：自然科学版

针对多约束下多无人机三维路径规划搜索效率低、规划路径质量欠佳的问题，提出一种融合多策略改进的蛇优化算法（ISO）。首先，通过减法平均优化器重构蛇群在勘探阶段的位置更新规则，克服因方向性缺失导致的效率损失；其次，结合边界选择思想构建理想域，利用优质信息引导雌性和雄性种群进行双模差分变异，实现广域搜索与精细化搜索的柔性切换；最后，设计动态边界博弈机制，保留精英信息的同时使算法识别并逃离局部最优。在6个基准测试函数上验证了改进算法的寻优能力。将该算法应用于不同复杂度场景下的多无人机路径规划任务，结果表明，改进算法在路径长度、协同约束、适应度值等方面均优于对比算法，在两个不同场景中，其优化后的适应度值较标准蛇优化算法分别降低了80.8%和71.7%。

Aiming at the problems of low search efficiency and poor planning path quality in 3D path planning of multiple UAVs under multiple constraints， an improved snake optimizer ( ISO ) integrating multiple strategies is proposed. Firstly， the position update rule of the snake group in the exploration phase is reconstructed through a subtractive optimizer to overcome the efficiency loss caused by the lack of directionality. Secondly， the ideal domain is constructed by combining the boundary selection idea， and the high-quality information is used to guide the male and female populations to perform a dual-mode differential mutation mechanism to achieve flexible switching between wide-area search and refined search. Finally， a dynamic boundary game mechanism is designed to retain elite information while enabling the algorithm to identify and escape from the local optimum. The optimization ability of the improved algorithm is verified on six benchmark test functions. This algorithm is applied to multi-UAV path planning tasks in scenarios of different complexity. The results show that the improved algorithm is superior to the comparison algorithms in terms of path length， collaborative constraints， and fitness value. In two different scenarios， its optimized fitness value is reduced by 80.8% and 71.7% respectively compared with the standard snake optimization algorithm.