改进量子遗传算法在无人机航迹规划中的应用

Improved Quantum Genetic Algorithm for UAV Route Planning and Simulation

研究无人机航迹规划优化问题,有效地规避威胁,可提高无人机的生存能力。但传统量子遗传算法在航迹规划方面局部寻优精度较低、稳定性差。为解决上述问题,提出改进量子遗传算法的无人机航迹规划方法。首先上述算法采用一维编码表示航迹,并对影响有效规避威胁的适应度构造代价模型和惩罚策略;针对量子遗传算法初始种群的单一性,引入关于概率划分的小生境协同进化策略,并对各种群采用动态量子旋转角,并借鉴狼群分配原则对种群进行更新,提高收敛速度;利用精英选择运算,创建精华种群,保留父代中最佳个体。仿真结果表明,上述算法的无人机航迹规划效率高,稳定性好,能够获得平滑的低代价航迹,是一种有效可行的航迹规划算法,且具有一定的推广意义。

In order to improve the survivability of unmanned aircraft vehicle （UAV） , and overcome the poor abil- ity at local searching precision and stabilization of simple quantum genetic algorithm, this paper proposed an improved quantum genetic algorithm for route planning. Firstly, the route was denoted by one-dimension coding, and the fit- ness function of the route planning problem was constructed. For a single quantum genetic algorithm initial popula- tion, evolutionary strategy with niche was introduced based on probability partition to increase the diversity of popula- tion. Then dynamic quantum rotating angles were used various groups, and the principle of distribution from wolves was used for reference to update the population for improving convergence speed; essence population was created u- sing elitist selection operation, and the best individual in the parent was reserved. The simulation shows the path planning of UAV based on the improved algorithm has higher efficiency and stabilization, and it can also gain gently and low cost route. So the algorithm is an effective and feasible route planning algorithm, with certain significance for popularization.