This paper shows an approach, how to implement a vehicle state estimator using a variant of the Kalman Filter, the so called Unscented Kalman Filter. This establishes the possibility to use non-linear elements, which are non-differentiable or whose derivative cannot be calculated with an acceptable computation effort. The described estimation is based on a two-track model of the vehicle, which simulates the tyre behaviour with the Magic Formula of Pacejka. The estimation designed that way, offers a significant improvement of the estimation accuracy, with a low increase of calculation complexity, compared to the standard Kalman Filter. These improvements, especially during high-dynamical manoeuvres with high sideslip angles, were demonstrated by means of a complex vehicle dynamics model.For adjusting the estimator design parameters and assessing the achieved estimation results, the estimator is integrated into a vehicle dynamics simulation environment. The basis for this simulation environment is formed by a MATLAB/Simulink full vehicle simulation model. Fundamentally, this model consists of the linear differential equations of motion of a 5 mass, 10 degree of freedom (DoF) system: translational and rotational motion of the body and vehicle respectively in three coordinate directions as well as translational motion of the 4 wheel masses. The following values are fed into the vehicle state estimator by the vehicle dynamics model: steering angle, lateral acceleration, yaw rate, longitudinal acceleration and longitudinal velocity. Significant advantages of the non-linear estimation are demonstrated in highly dynamic manoeuvres with large steering angles and high sideslip angles.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Vehicle state estimation using a non-linear two-track model and the unscented Kalman filter


    Contributors:


    Publication date :

    2009


    Size :

    11 Seiten, 7 Bilder, 18 Quellen


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    Vehicle state estimation using a non-linear two-track model and the Unscented Kalman Filter

    Saeger,M. / Reif,K. / Renner,K. et al. | Automotive engineering | 2009


    Unscented Kalman filter for vehicle state estimation

    Antonov, S. | Online Contents | 2011


    Unscented Kalman filter for vehicle state estimation

    Antonov,S. / Fehn,A. / Kugi,A. et al. | Automotive engineering | 2011


    Using the unscented Kalman filter and a non-linear two-track model for vehicle state estimation

    Reif, Konrad / Renner, Kerstin / Saeger, Martin | Tema Archive | 2008


    Unscented Kalman filter for vehicle state estimation

    Antonov, S. / Fehn, A. / Kugi, A. | Taylor & Francis Verlag | 2011