基于改进蚁群算法的森林防火预警无人机路径规划

Path Planning Approach Based on Improved Ant Colony Optimization for Forest Fire Warning UAV

针对无人机应用于森林防火预警问题,提出了基于蚁群算法的固定点位巡逻路径规划方法。原始蚁群算法对已选路径采用信息素累积的反馈方式对其他蚂蚁产生影响,这使算法容易过早陷入局部最优,为此,提出了一种基于信息素递减更新的蚁群算法,信息素浓度减少值于该路径长度成正相关,从而提高新鲜路径信息素的相对值,增大算法全局搜索的概率。仿真结果显示,与基于信息素累积的蚁群算法相比,改进的算法有更快的收敛速度,且可以得到更好的路径规划结果。

Aiming at the application of unmanned aerial vehicles to forest fire warning,a fixed point patrol path planning method based on ant colony algorithm is proposed.The classical ant colony algorithm adopts the feedback method of pheromone accumulation on the selected path to affect other ants,which makes the algorithm easy to fall into the local optimum prematurely.For this reason,an ant colony algorithm based on pheromone decreasing update is proposed.The reduced value of the element concentration is positively correlated with the path length,thereby increasing the relative value of the fresh path pheromone and increasing the probability of the algorithm's global search.The simulation results show that compared with the ant colony algorithm based on pheromone accumulation,the improved algorithm has a faster convergence rate and can obtain better path planning results.