Neural Network Observer-Based Adaptive Fault-Tolerant Control of Gantry Synchronization Systems Under Actuator and Sensor Faults

Synchronization control is a key technology in dual-axis linear motor gantry synchronous (DALMGS) systems, since it achieves high acceleration and accuracy for large inertia equipment. However, during long operating periods, linear motors may be affected by a variety of faults, such as demagnetization, wear, or nonlinear drift, among others. These faults can seriously affect the performance of DALMGS systems, and even lead to the deformation of the mechanical structure. This article focuses on the development of an adaptive fault-tolerant controller based on a neural network (NN) observer for DALMGS systems under actuator and sensor faults. First, based on a radial basis function NN, an observer is designed to approximately fit and compensate for uncertain system parameters, actuator faults, and external disturbances. Then, a suitable adaptive controller is designed to estimate sensor faults and NN weights. The stability of the closed-loop system is demonstrated through Lyapunov functions. The effectiveness of the proposed method compared with other State-of-the-Art ones is verified through several experiments on a real DALMGS system.