RBF Neural Network-Based Adaptive Fault-Tolerant Control of Electromechanical Servo System with LuGre Friction

An adaptive control scheme was put forward for electromechanical servo system with LuGre friction, with a view to mitigating the detrimental effect on the system by friction nonlinearity. Initially, the LuGre friction model was presented for the servo system. Through incorporation of sliding mode function, selection of Lyapunov function and development of adaptive law based on a radial-basis-function (RBF) neural network (RBFNN), we designed an adaptive fault-tolerant controller that used RBFNN for electromechanical servo system for the first time, to the best of our knowledge, and assessed its stability. Simulation results showed that our control scheme outperformed the single RBFNN control in both speed and position tracking.