Adaptive observers for a class of nonlinear systems with application to induction motor

This paper presents an adaptive observer for a class of multi-input multi-output (MIMO) nonlinear systems. These systems are linear in unknown parameters, and system nonlinearities satisfy Lipschitz conditions. The implementation of observer does not require any coordinate transformation. However, the estimation convergence analysis is dependent on a nonlinear filtered coordinate transformation. Sufficient conditions for stability are derived in terms of H-infinity like matrix Riccati equations. Further, the estimated states in original coordinates and estimated parameters converge to true values with persistency of excitation. In the end, the usefulness of the proposed observer is shown by an example. In this example, an adaptive observer is designed for induction motor to estimate its five states, rotor resistance and load torque by measuring only stator currents and input voltages.