基于改进ACO算法的多UAV协同航路规划

针对无人机(Unmanned Aerial Vehicle,UAV)在执行任务过程中遇到的诸如敌方防空火力、地形障碍及恶略天气等各类威胁源,采用威胁源概率分布的方法进行威胁的量化处理,构建任务空间的威胁概率密度分布图,有效消除了威胁源的差异性。根据UAV在任务飞行过程中的性能约束与时、空协同约束,同时考虑任务过程中UAV的损毁概率最小、任务航程最短,构建了相应的综合任务航路代价最优化目标函数。结合传统蚁群优化算法(Ant Colony Optimization,ACO)在解决此类问题中的不足,给出了相应的改进策略,提出采用协同多种群ACO进化策略来实现多UAV在满足时、空协同约束下的协同航路规划。通过相应的仿真计算表明,改进后的ACO协同多种群进化策略算法更适用于多UAV协同任务航路规划问题,具有一定的实用性。从而为多UAV协同任务航路规划问题的求解提供了科学的决策依据。

Cooperative Path Planning for Multi-UAVs Based on Improved ACO Algorithm

In this paper,a cooperative path planning method by utilizing improved ant colony optimization (ACO) is presented.In the military domain UAVs usually require the intelligence to safely maneuver along a task path to an intended target,avoiding obstacles such as enemy threats,terrain or bad weather.The method of probability distribution is adopted to deal with such obstacles is adopted.By constructing the obstacles probability density distribution map of the task area the individual diversity of the obstacles effectively can be eliminated.Then according to the flying performance,spatial and temporal constraints of the UAVs,a path cost optimization objective function is proposed by balancing the damage probability of UAV and shortest voyage.For the deficiency of standard ACO algorithm,an improved ACO algorithm is brought that some improvement strategies is put forward for such optimization problems,also a co-evolution multi-ant colony algorithm is proposed to solving the cooperative path planning problems for multi-UAVs.Simulation results show that the improved ACO algorithm can solve the problem effectively,and compared with standard ACO algorithm,it is also more efficiency.