Adaptive Unscented Kalman Filter-Based State Estimation With Applications To SI Engines

With a rapid increase in the vehicular industry to ensure maximum efficiency and lower fuel consumption in the vehicles, advanced control strategies are required. Such strategies need precise information of intake manifold pressure, engine speed, and fuel flow rate. Due to the engine placement and operation, sensors can not be placed to measure all the variables. This issue is addressed in this paper and a new adaptive Unscented Kalman filter (UKF) is proposed to estimate these variables. The novelty lies in the new adaptive laws that are designed to update the process noise and measurement noise covariances within the constrained augmented state-based UKF (CASUKF). Another contribution lies in the new combination of the novel adaptive laws, and CASUKF. In the literature, variants of the UKF either adapt the process noise and measurement noise covariances on the standard UKF or implement CASUKF with constant values of the process noise and measurement noise covariances. The algorithm is implemented for the nonlinear mean value spark-ignition engine model. Simulation results are shown to provide the effectiveness of the algorithm in comparison to other variants of the UKF.