Adaptive Neural Network Constrained Fault Tolerant Control for Nonlinear Systems With Actuator Failures and Saturation

This article studies a neural network (NN)-based adaptive fault-tolerant control (FTC) scheme to solve the actuator failure problem of nonlinear systems with input saturation and time-varying state constraints. The introduction of asymmetric Barrier-Lyapunov function (BLF) makes controller design more difficult due to the occurrence of actuator saturation and failure. Therefore, the method proposed in this article proposes a saturation fault-tolerant controller with constraint compensation information under the backstepping control design framework to solve the state constraint asymmetry problem. By design-improved asymmetric time to change the BLF, the design of the state constraint controller will become more realistic and constraints will be weakened. Actuator failure in this article considers deviation failure and loss of effectiveness. Based on the nature of the index function, the improved BLF can make the boundaries of state constraints smaller and smaller, and the boundaries of constraints can change with the expected trajectory when satisfying the input saturation. Therefore, this article solves the FTC problem of non -linear strict feedback systems, overcoming the impact of non -symmetrical state constraints and input saturation on system performance. Simulation verified the feasibility of this control method. Note to Practitioners —This work implements asymmetrically constrained fault-tolerant control under actuator sensor failure for a nonlinear system with input saturation. Input saturation and state-constrained control asymmetries caused by actuator failures are still rarely studied. This article proposes an improved BLF to solve the asymmetry problem under actuator failure. A fault-tolerant control method is designed considering multivariable conditions and control input constraints, which covers information on actuator faults and amplitude saturation. Based on the properties of the exponential function, the improved BLF can make the bounds of the state constraints smaller and smaller, and the bounds of the constraints can change with the desired trajectory while satisfying input saturation. The proposed method has a general structure and has the potential to be implemented in real systems.