Design of robust fault detection filters for MIMO uncertain plants with quantized information

This article addresses the robust (in the disturbance decoupling sense) design problem of fault detection (FD) filters for uncertain multi-input-multi-output (MIMO) sampled-data systems with fixed-point quantisation. The only available signal is the output variable. A design technique for a deterministic FD observer coupled with a stabilising controller is proposed such that ultimate boundedness (asymptotic vanishing if quantisation is absent) of closed-loop trajectories and estimation error is guaranteed, robustly with respect to disturbances affecting the continuous-time system. A test-function is proposed for detecting the occurrence of actuator faults in the plant. A simulation study on a vehicle suspension system is also reported.