基于改进SOS算法的UCAV鲁棒机动决策研究

Robust optimization based on modified SOS for UCAV maneuvering decision

针对无人作战飞机自主空战机动决策问题,提出了一种鲁棒机动决策方法。设计了反映空战态势的鲁棒隶属函数,并基于此设计鲁棒多目标决策函数;针对动作库在机动决策中的不完备性与传统优化方法求解时效性缺陷,运用基于自适应和精英反向学习策略改进的共生生物算法,对控制量进行优化进而完成机动决策;仿真结果表明,鲁棒机动决策结果更具优势且改进算法求解具有实时性,满足机动决策需求。

To solve the problem of Unmanned Combat Aerial Vehicles(UCAV)maneuvering decision during the automatic air combat, a new kind of robust maneuvering decision method is proposed. Firstly, robust membership functions that reflect the air combat situation are proposed in order to construct the Robust Multi-objective Optimization(RMO)functions. Then, in view of the incompleteness of maneuvering library and the timeliness limitation of traditional optimization method, a modified Symbiotic Organisms Search(SOS)algorithm which is based on adaptive and elite opposition-based learning strategy is used to optimize the control variables, so the maneuvering decision is completed. The simulation result shows that promising results can be obtained with the help of RMO, and the modified SOS has a better real-time performance which can satisfy the real-time requirement of maneuvering decision.