针对炮控系统随动模拟加载系统存在的摩擦、间隙、耦合等复杂非线性和参数时变等不确定性特征, 提出了一种自适应神经滑模控制策略。对于系统中存在参数时变等不确定性, 利用RBF神经网络自适应逼近不确定部分; 另外, 利用RBF神经网络动态调节切换函数的切换增益, 改善系统的动态品质。采用Lyapunov理论推导出自适应律, 在线估计神经网络权值和未知函数, 并证明了系统稳定性。仿真表明, 该控制策略能够较好地抑制干扰力矩, 响应快, 保证了系统静、动态的加载控制精度和鲁棒性。

To overcome the complicated nonlinearity such as friction, clearance and coupling and the uncertainties such as time-varying of parameter existed in the servo simulated loading system for Gun Control Systems (GCS), an adaptive sliding mode control strategy based on RBF neural network is proposed.To the uncertainties such as parameter time-varying in the system, RBF neural network is used to approach uncertain parts adaptively.Then, a dynamic adjustment approach of the switch gain RBF neural network is used to dynamically adjust the switching gain of the switching function, which enhances the dynamic performance of the system.An adaptive law is derived by using Lyapunov theory to estimate neural network weights and unknown functions online and ensure the stability of the system.Simulation results show that this control strategy can not only effectively suppress the external disturbances, but also has a rapid responding speed, which ensures the control precision and robustness when the system is loading in static or dynamic state.