Unscented kalman filter for combined longitudinal and lateral vehicle dynamics

verfasst von

M. Wielitzka, S. Eicke, A. Busch, M. Dagen, T. Ortmaier

Abstract

Numerous advanced driver assistance systems rely on parametric models describing the vehicle’s behavior and stability. Accurate knowledge about the vehicle’s dynamics in terms of specific states has the potential to improve these systems’ performance drastically. Since several crucial states, e.g. side slip angle, cannot be measured directly, estimation methodologies were developed. In this paper a highly nonlinear model of the vehicle’s longitudinal and lateral dynamics, containing 13 states and 8 measurements is used within an unscented Kalman Filter. The estimation results are compared to high precision measurements taken using a Genesys ADMA inertial measurement unit on a Volkswagen Golf GTE Plug-In Hybrid during three different maneuvers.

Organisationseinheit(en)

Institut für Mechatronische Systeme

Typ

Aufsatz in Konferenzband

Seiten

515-520

Anzahl der Seiten

6

Publikationsdatum

2017

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Fahrzeugbau, Maschinenbau, Angewandte Informatik