Adaptive Neural Fault Tolerant Control for Input-Delayed Stochastic Systems Subject to States and Input Quantization
针对带有输入延迟和状态、输入量化的随机系统，分三步设计自适应神经控制方案，用径向基神经网络逼近未知项，用Pade近似处理延迟，并解决传感器故障和量化不连续问题，仿真验证了有效性。
For the input-delayed stochastic systems with the states and input quantization, the adaptive stabilization problem is investigated in this article. The whole control scheme design process can be divided into three steps. First, the traditional adaptive neural control scheme is developed for the controlled system. Next, the effective control scheme is proposed for the system with the quantized states. Finally, the adaptive neural control method is developed for the considered system with the states and input quantization. The radial basis function neural network (RBFNN) is applied to approximate the unknown terms online, and the Pade approximation method is introduced to deal with the input-delayed problems. The adaptive neural fault control strategy is presented to address sensor faults and the discontinuity due to the quantized states. Under the constructed controllers, all the closed-loop signals remain semi-globally uniformly ultimately bounded (SGUUB) in mean square. The effectiveness and superiority of the presented control schemes are verified by some simulation results.