基于自适应RBF神经网络的超空泡航行体反演控制

针对超空泡航行体姿轨控制普遍存在的模型不确定性问题进行相关研究.为此, 首先对其动力学特性进行分析, 并建立了超空泡航行体的动力学名义模型, 随后将其改写为不确定反馈系统, 然后利用反演控制方法设计超空泡航行体姿轨控制器, 针对模型中的未知函数利用径向基函数(Radial basis function, RBF)神经网络进行逼近并补偿, 由基于Lyapunov稳定理论设计的自适应方法计算神经网络的权重, 并给出稳定性证明.仿真研究验证了控制器设计的有效性.

Adaptive RBF Neural Network Based Backsteppting Control for Supercavitating Vehicles

This paper is proposed for the problems of model uncertainty such as the control of supercavitating vehicles. Firstly, the nominal model of supercavitating vehicles is built based on the analysis of the vehicle dynamic characteristics. Then we rewrite it as the uncertainty feedback system, and an orbit and attitude controller is designed via the backstepping control theory. The radial basis function (RBF) neural networks are presented to approximate and compensate the unknown functions, otherwise, the weights of the neural networks are designed by the adaptive method based on the Lyapunov theory, and the stability proof is also proposed. Finally, the simulations prove the effectiveness of the above controllers.