基于多准则交互膜进化算法的UAV三维航迹规划

针对智能优化算法在无人机(unmanned aerial vehicle, UAV)三维航迹优化中搜索复杂度较高、容易陷入局部最优的问题,提出一种基于嵌套式细胞膜结构的多准则交互式多目标进化算法。以建立的多目标航迹评价模型来克服航迹评价加权求和的不足;同时在应用降维离散缩减寻优空间的基础上,采用萤火虫算法和人工蜂群算法作为不同膜内优化准则,利用膜系统计算的并行性和膜内信息交互优势提高算法性能;并对膜内进化规则进行非支配排序、搜索加权等改进,实现了UAV三维多目标航迹寻优。仿真实验表明,所提方法在有无威胁两种环境下均能快速搜索到不同侧重目标的相对最优航迹,证明了该方法的有效性。

Three dimensional route planning of UAV based on the multi-criterion interactive membrane evolutionary algorithm

Aiming at the problem of the high complexity and easy to fall into the local optimal for the intelligent optimization algorithms in solving unmanned aerial vehicle(UAV) three dimensional route planning, a multi-criterion interactive multi-objective evolutionary algorithm based on the nested membrane structure is proposed. The multi-objective evaluation model is established to overcome the deficiency of weighted sum of route planning evaluation. Meanwhile, based on the application of the dimensionality reduction discrete to reduce the optimal space, firefly algorithm and artificial bee colony algorithm are used as intra-membrane optimization rules, takes advantage of the parallelism of membranes structure and the information interaction within the membrane to improve the performance of the algorithm. And the evolution rules within the membrane are improved by the method of non-dominated sorting and search weighting to realize the three dimensional multi-target route planning optimization. The simulation experiments show that the proposed algorithm can quickly find the relative optimal trajectory with different focuses under both threat and non-threat environments, which demonstrate the efficiently of the proposed algorithm.