超机动飞行非线性鲁棒自适应控制系统研究.pdfVIP

30 4 V ol . 30 No. 4 2008 4 Sy stems Engineer ing and Electr onics A pr. 2008 : 1001506X 2008) 0407 1005 周 丽, 姜长生, 文 杰 南京航空航天大学自动化学院, 江苏 南京2100 16) : 针对飞机模型中存在气动参数不确定性以及外界干扰等 响因素, 设计了 一种超机动飞行的非线性 鲁棒自适应控制系统控制系统设计过程中, 模型不确定性和外界干扰由RBF 神经网络在线补偿, 控制律及神 经网络权值自适应律由反步法得到解决了系统中控制增益矩阵未知, 同时存在外界干扰情况下的鲁棒飞行控 制系统设计, 并证明了闭环系统所有信号有界, 系统跟踪误差和神经网络权值估计误差指数收敛到有界紧集内 对所研究的飞行控制系统进行了过失速H erbst 机动仿真, 结果验证了该系统在过失速机动条件下具有良好的控 制性能 : 飞行控制系统; 干扰抑制; 反步法; R BF 神经网络; 超机动 : V 249. 1; T P 273 : A Research on robust and adaptive nonlinear control system of supermaneuverable f ight ZH OU L i, JIA N G Changsheng, WEN Jie ( Coll. of A utomation Eng ineering, N anj ing Univ . of A er onautics and A str onautics, N anj ing 210016 , China) Abstract: A r obust and adaptive flight contr ol sy stem is de signed for a nonlinear air cr aft model w ith param et er uncert aint ies and unknow n distur bance. Dur ing the design of contr ol sy stem, par am eter uncert aint ies and distur bance ar e compensat ed for online by RBF neur al netw or ks. Contr ol law and adapt ive law s of neural net w ork s ar e achieved t hrough backstepping m ethod. As a re sult, t he de sign problem of r obust flight contr ol sy s t em is r esolved under th e condit ion of unknow n cont rol gain mat r ix and disturbance. It is prov ed t hat all signals in the clo sedloop syst em are bounded. M or eover , t racking err or of the sy stem and estimat ion err or of neur al netw or ks w eight s ar e r emained in t he com pact set s. H erbst m aneuv er sim ulation re sult s dem onst rate th at the designed contr ol sy stem has good perform ance under condit ions of