Adaptive Switching Control for Nonlinear Uncertain Systems With Sensor Faults and Quantization

In this article, global adaptive quantized control problem of a class of nonlinear uncertain systems with sensor faults is addressed, and a novel non-identification adaptive control method is proposed based on a switching mechanism. One advantage of the proposed method is that continuous multiplicative sensor faults are tolerated and handled in a generalized Lyapunov matrix inequality; another is its lower computational complexity compared to the traditional adaptive backstepping method (avoiding the 'differential explosion' problem). In addition, a new hysteresis logarithmic-type quantizer is developed to prevent chattering. The quantization error is dealt with by using sector bounded property, where the compensation of multiplicative deviation in quantization is similar to that of multiplicative sensor faults, while additive bias is treated as a disturbance (quantization dead-zone size determines the magnitude of the disturbance). Especially, the relationship between quantization dead-zone and control accuracy is clearly characterized. Also, it is shown that the proposed control method is robust to mismatched disturbances, and the solutions of the closed-loop system can be guaranteed to converge to the arbitrarily small neighborhood of the origin as long as the dead-zone of quantizer and mismatched disturbances are sufficiently small.