针对一类含非参数不确定的受限非线性系统, 提出一种新的自适应迭代学习控制算法. 该算法利用神经网络(neural network)逼近系统中的非参数不确定性, 神经网络的权值通过迭代轴上的自适应算法进行更新. 控制器设计补偿项对神经网络逼近误差上界进行估计以此消除其影响. 采用一种新的障碍李雅普诺夫函数(barrier Lyapunov functions)并与组合能量函数(composite energy function)方法相结合, 当系统中同时存在非参数不确定性,随机初始误差,及非严格重复外部扰动时,能够保证系统同时满足输入和状态受限要求. 将所提出的算法应用于高速列车运行控制系统, 仿真结果验证了算法的有效性.

A new adaptive iterative learning control (AILC) is presented for a class of constrained nonlinear systems with non-parametric uncertainties. A neural network is utilized to approximate the non-parametric uncertainties in the system. The weights of the neural network are updated adaptively along the iteration axis. The approximation error of the neural network is also compensated by estimating its upper bound along the iteration axis. Composite energy function (CEF) with a new barrier Lyapunov function, together with a projection mechanism, is used to facilitate the analysis of tracking error convergence while satisfying the state and input constraints. Under non-parametric uncertainties, random initial conditions, and iteration-varying external disturbances, the asymptotic and pointwise convergence properties of the state errors are guaranteed and simultaneously the state and input constraints are not violated. The effectiveness of the proposed AILC is verified by simulations on a high-speed train operation system.