基于自适应遗传算法的无人机航路规划

Path Planning for UAV Based on Adaptive Genetic Algorithm

在规划区域内随机产生一系列威胁点和相应威胁半径来量化无人机(UAV)任务环境,通过纵向剖分目标区,将航路点的表示由二维缩减到一维,采用实值编码以提高运算精度.针对遗传算法(GA)早收敛和收敛慢的问题,在交叉和变异中设计了自适应算子.计算仿真表明该控制算法能使无人机在复杂环境中回避威胁,快速选择最短路径,提高了规划效率.

Series of stochastic points and its threat radius in the considered region are generated to simulate the task environment of UAV, then the region is equally divided into N part vertically, based on which the dimension of the expression of the UAV path decreases from 2-D to 1-D. Adaptive crossover and mutation operator is designed, by which premature and lower convergence problem can be prevented. The simulation result shows that the control algorithm is efficient in the obstacle prevention and shortest path choice in complicated environment.