一类输出受限机械系统的鲁棒自适应容错跟踪

Robust Adaptive Fault-Tolerant Tracking for a Class of Mechanical Systems with Output Constraints

研究了一类输出受限机械系统的鲁棒自适应容错跟踪控制问题。借助动态面控制(DSC)技术和时变尺度函数,设计了一个基于神经网络的状态反馈控制器,确保系统在遭受非匹配扰动和非仿射非线性执行器故障的情况下,输出信号满足预设的时变约束,并且跟踪误差在预设的时刻之前进入原点的任意预设的小邻域,同时也解决了传统反步法中虚拟控制器需要被多次求导的问题。本文的创新点在于放宽了对时变约束函数的要求,并且在时变因素存在的情况下阐明了神经网络逼近的紧集和闭环信号的有界性之间的逻辑关系。最后通过仿真验证了控制策略的有效性。

This paper investigates the issue of robust adaptive fault-tolerant tracking control for a class of mechanical systems with output constraints.With the aid of the dynamic surface control(DSC)technology and scaling time-varying functions,a state feedback controller based on neural networks is presented.In the presence of unmatched disturbances and non-affine nonlinear actuator faults,the controller ensures that the system satisfies the predetermined time-varying output constraints and the tracking error enters any predetermined small neighborhood of the origin before the predetermined time,while solving the problem that virtual controllers need to be differentiated many times in traditional backstepping methods.The innovation lies in relaxing the requirement for the time-varying constraint functions and clarifying the logical relationship between the compactness of neural networks approximation and the boundedness of closed-loop signals in the presence of time-varying factors.Finally,the simulation is provided to demonstrate the efficiency of the proposed method.