Low-cost adaptive fuzzy neural prescribed performance control of strict-feedback systems considering full-state and input constraints

A low-cost adaptive neural prescribed performance control (LAFN-PPC) scheme of strict-feedback systems considering asymmetric full-state and input constraints is developed in this paper. In the controller design procedure, one-to-one nonlinear transformation technique is employed to handle the full-state constraints and prescribed performance requirement. The Nussbaum gain technique is introduced for solving the unknown control direction and the input constraint nonlinearity simultaneously. Furthermore, a fuzzy wavelet neural network (FWNN) is utilized to approximate the unknown nonlinearities. Compared with traditional approximation-based backstepping schemes, the constructed controller can not only overcome the so-called "explosion of complexity" (EOC) problem through command filter, but also reduce filter errors by error compensation mechanism. Moreover, by constructing a virtual parameter, only one parameter is required to be updated online without considering the order of system and the dimension of system parameters, which significantly reduces the computational cost. Based on the Lyapunov stability theory, the presented controller can ensure that all the closed-loop signals are ultimate boundedness, and all state variables and tracking error are restricted in the prespecified regions. Finally, the simulation results of comparison study verify the effectiveness of the constructed controller.