Dead-zone effect on the performance of state estimators for hydraulic actuators

State estimation in hydraulic actuators is a fundamental technique for fault detection and it is also a valid tool useful to reduce the installation of sensors. The performance of the linear/linearization based techniques for the state estimation is strongly limited due to hard nonlinearities that characterize hydraulic actuator. One of the most common hard nonlinearities in hydraulic actuator is the dead-zone. This paper focuses on an alternative nonlinear estimation method that is able to fully take into account dead-zone hard nonlinearity and measurement noise. The estimator is based on the state-dependent-Riccati-equation (SDRE). A fifth order nonlinear model is derived and employed for the synthesis of the estimator. Several simulations have been conducted in order to analyse the effect of the dead-zone characteristic on the novel estimator performance, showing comparisons with the largely used extended Kalman filter (EKF). Numerical results demonstrate the effectiveness of SDRE based technique in applications characterized by extended dead-zone for which the EKF method provides poor results.