摘要
针对制导控制一体化(IGC)模型中的不确定性难以进行估计补偿的问题,提出了基于神经网络的IGC反演设计方法。首先,根据弹目相对运动关系以及导弹自动驾驶仪模型建立了三维空间中的IGC模型。其次,针对由目标机动引起的模型不确定性,提出应用高阶滑模微分器(SMD)对导弹导引头获得的弹目相对运动信息进行微分,从而估计出目标加速度的方法,然后考虑导弹自身由于参数摄动以及未建模动态引起的模型不确定性,应用SMD和神经网络模型进行在线逼近补偿,基于反演控制理论设计了带有SMD和神经网络模型的IGC算法,应用李雅普诺夫稳定性理论对所设计的控制算法进行了稳定性证明。最后,进行了导弹六自由度仿真,验证了所设计控制算法的有效性。
In allusion to the problem that the uncertainty in the integrated guidance and control （IGC） model is difficult to estimate and compensate, a novel IGO algorithm is designed based on neural network and back-stepping control theory. Firstly, the three space dimensional IGO model is constructed with the relative motion of the missile and target as well as the missile autopilot model. Secondly, aimed at the model uncertainty caused by the target maneuvers, a novel method is proposed featuring the estimation of target acceleration by introducing the high order sliding mode differentiator （SMD） which differen-tiates the target and missile relative motion information obtained from the missile active homing seeker. Then, the uncertainties caused by possible parameter perturbations and unmodelled dynamics are considered. With the SMD and neural network performing an online estimation and compensation, a novel IGO algorithm with SMD and neural network is designed based on the back-stepping control. The stability of the algorithm is precisely derived and analyzed. Finally, the missile＇s six degree of freedom simulation is carried out to verify the effectiveness of the algorithm.
出处
《航空学报》
EI
CAS
CSCD
北大核心
2015年第5期1661-1672,共12页
Acta Aeronautica et Astronautica Sinica
基金
航空科学基金(20130196004)~~
关键词
制导控制一体化
不确定性
高阶滑模微分器
神经网络
反演控制
integrated guidance and control
uncertainty
high order sliding mode differentiator
neural network
back-stepping control
