一种无人机路径规划的混沌遗传算法

A Chaotic Genetic Algorithm(CGA) for Path Planning of UAVs(Unmanned Air Vehicles)

提出采用基于混沌的遗传算法进行无人机路径优化问题的求解。算法利用极坐标描述战场中的威胁位置和航路点,缩短了路径编码长度,提高了搜索效率,并在遗传算法操作时加入混沌操作,扩大了搜索范围,提高了优化速度,有效地解决了解空间巨大带来遗传算法收敛速度慢和容易陷入局部最优的局限。实例仿真结果表明,文中的算法与标准遗传算法相比,优化效率显著提高,得到的优化解即优化航路更好地规避了威胁。

Purpose. We propose CGA to overcome the limitations unavoidable when using standard GA to do path planning of UAVs; these limitations, caused by the enormity of search space, include low convergence rate and frequently obtaining local optimum rather than the desired global optimum. We explain the CGA in detail in the full paper; in this abstract we just list the three topics of our explanation： （1）CGA; （2）path planning of UAVs, whose subtopics are gene code for path planning of UAVs （2.1）, fitness function （2.2）, and the eight steps of CGA （2.3）. As compared with standard GA, our CGA is better in three respects： （1）searching space is enlarged through chaotic search, （2）searching efficiency is improved, （3） global optimum is ensured. Finally we give a numerical simulation example for path planning of an UAV, whose results for both our CGA and standard GA are shown in Figs. 2 and 3 of the full paper; these results show preliminarily that our CGA improves the optimization efficiency and that the path planned by our CGA can avoid threats better.