Vehicle navigation filter designs using adaptive constraint-filtering method

The conventional filter requires that all the vehicle dynamics and noise processes are completely known. As a practical fact this is usually impossible. To deal with such a problem, the adaptive constraint-filtering (ACF) method is proposed in this study. The CF method developed previously can accommodate the constraint in the filtering process for a non-linear dynamic system. However, the assumption that the modelling noise and the sensor noise are known may not be practical. Here, the fuzzy innovation adaptive estimation approach is proposed to determine the window size, which is assumed constant in the classical adaptive scheme. To assess the performance of the proposed algorithm, the Monte Carlo method is adopted. The performance of the various filters, such as the Kalman filter (KF), the adaptive KF (AKF), the CF and the adaptive CF ACF are then compared. The simulation results show that the ACF method is evidently better than the other filters. From dynamic experimental results, it is shown that the proposed methodology yields a successful algorithm to manage the ill-conditioned global positioning system positioning problem. The adaptation accuracy based on the proposed methodology is substantially improved.