Descriptor Observer based Approach for Adaptive Reconstruction of Measurement Noises and Sensor Faults in Uncertain Nonlinear Systems

This paper addresses the problem of reconstructing measurement noises and sensor faults in uncertain nonlinear dynamic systems under Lipschitz condition. A descriptor observer based approach is proposed, where an auxiliary state vector consisting of measurement noise and sensor fault is introduced to build an augmented descriptor system. Based on the descriptor system, a descriptor observer is designed for the state estimation and noise/sensor fault reconstruction. Using the linear matrix inequality technique, estimation convergence condition is established. It is shown that by selecting the observer parameters properly, desired reconstruction precision can be achieved. A numerical example is presented to illustrate the design procedures, and the simulation results show the satisfactory estimation performance.