UKF-based State and tire slip estimation for a 4WD electric vehicle. Issue 10 (2nd October 2020)

• Record Type: Journal Article
• Authors: Heidfeld, Hannes; Schünemann, Martin; Kasper, Roland
• Is Part Of: Vehicle system dynamics. Volume 58:Issue 10(2020)
• Page Start: 1479
• Page End: 1496
• Publication Date: 2020-10-02
• Subjects: Unscented Kalman Filter -- state estimation -- electric vehicles -- vehicle dynamics control; Motor vehicles -- Dynamics -- Periodicals; Electronic journals; 629.231
• DOI: 10.1080/00423114.2019.1648836
• Languages: English
• ISSNs: 0042-3114

Abstract

Abstract : For performance improvement of modern vehicle dynamic control systems, exact knowledge of the vehicle motion state and the road conditions is essential. Addressing this problem, the development and validation of a state observer for a four-wheel drive electric vehicle is examined in this article. In particular, a reliable and accurate estimation of the tire slip in longitudinal and lateral direction at each wheel is desired. For this, an Unscented Filter is used to estimate the vehicle chassis and wheel dynamics. At the same time, the estimator is able to adapt the tire model according to the current road conditions by using wheel-individual tire-road friction coefficients. In the beginning, details on the used vehicle model as well as the estimator tuning and implementation procedure are given. The validation takes place first in a vehicle simulation environment to point out the performance of the observer especially in severe driving situations, like a sudden change of road friction conditions during a slalom manoeuver. Afterwards, driving tests with an electric four-wheel drive vehicle are carried out to proof the presented observer concept under real-world conditions. All performed driving manoeuvers are described and the test results are presented and evaluated.