基于自适应蚁群算法的无人机最优路径规划

Optimal Path Planning of Unmanned Aerial Vehicle Based on Adaptive Ant Colony Algorithm

为了提高无人机执行突防侦察任务时的成功率,在敌人防御区域内执行侦察任务前必须规划设计出高效的无人机飞行路径,以保证无人机能够以最小的被发现概率及最优路径到达目标点.针对这一问题,根据敌人防御区域内的威胁源建立威胁代价模型,并且在基本蚁群算法的基础上进行了分析讨论,通过调整信息素挥发因子大小,提出了一种适用于路径规划的自适应蚁群算法,并对算法得到的最优路径进行了仿真计算.仿真结果表明,采用自适应蚁群算法可以得到17.5km的最优路径长度,算法收敛时间为7.2s,只需迭代8次;采用基本蚁群算法可以得到18.4km的最优路径长度,算法收敛时间为15.6s,需要迭代54次.对比可知自适应蚁群算法具有明显的优势.

To improve the success rate of UAV （unmanned aerial vehicle） in penetration reconnaissance mission, a highly effective flight path for the UAV should be planned before the execution of the reconnaissance mission into enemy defense area to ensure that the UAV reach its destination by following an optimal path and have the lowest probability to be sighted. To deal with this problem, a threat cost model is set up according to the threat sources of enemy defense area and, based on ant colony algorithm, an analysis and discussion is presented. By adjusting the pheromone volatilization coefficient, an adaptive ant colony algorithm is proposed and applied to path planning. Simulation calculations for the optimal path derived from two algorithms are presented. The simulation results show that using adaptive ant colony algorithm can get optimal path length of 17.5km, convergence time of 7.2s and it needs 8 times iterations. Using ant colony algorithm can get optimal path length of 18.4km, convergence time of 15.6s and it needs 54 times iterations. Comparing the results, the adaptive ant colony algorithm has an obvious advantage