Sliding mode observers for fault reconstruction and estimation problem

This book is addressed to fault reconstruction and estimation (FRE) issue for nonlinear systems subject to actuator and sensor faults. We present two FRE schemes based on sliding mode observers under relaxed assumptions. The first FRE approach is based on an adaptive descriptor sliding mode observer whose design combines sliding mode concept, singular observer synthesis and adaptive control for a class of nonlinear systems with actuator faults in the dynamics and sensor faults in outputs equations. The second FRE method is based on an auxiliary high gain sliding mode observer whose design is based on high gain techniques and sliding mode theory. The latter FRE scheme is addressed to nonlinear systems such that both observer matching condition and minimum phase assumption are weakened and that the total number of faults may be larger than the number of outputs. The mathematical demonstrations of convergence for the two developed FRE Schemes are based on Lyapunov theory. The different theoretical results are validated via numerical simulations with Matlab/Simulink software by considering two applications based on single link robot arms.