基于混合多种群自适应蚁群算法的无人机航路规划

Route Planning of UAV Based on Hybrid of Multi-Population and Adaptive Ant Colony Algorithm

摘要:

针对基本蚁群算法在航路规划中易于过早陷入局部最优解,对蚁群算法进行了改进；提出了具有多种群的蚁群算法,并将导引因子引入到状态转移策略中,减少蚂蚁局部搜索的盲目性,确保蚂蚁最终完成航路搜索；当算法陷入局部收敛时,通过交换各种群的信息素,并对每个种群的挥发系数进行自适应调整,从而扩大了搜索空间,提高了搜索全局性；最后在代价函数简化后的栅格图中对改进算法进行了仿真；仿真结果表明,该方法可以有效防止算法陷入局部最优,是一种有效的航路规划方法。

Abstract:

The prominent shortcoming of the basic ant colony algorithm in route planning is easily trapped into local optimal solution. In the paper, the original ant colony algorithm is improved. A hybrid population ant colony algorithm is proposed to solve the problem. And a guidance factor was included into the state transition strategy, and it can reduce the blindness of local search and ensure ants to reach the destination. When the algorithm trapped into local convergence solution, through exchange each population's pheromone and the adaptive adjustments of the volatile coefficient can expand the search space and improve the overall searching workspace .Finally the improved algorithm is simulated in the grid map after the simplification of cost function. Simulation results show that the algorithm improves the convergence greatly, and can effectively prevent algorithm is easily trapped into local optimal solution,and is effective for the route planning problem.