Title: Vehicle lateral dynamics modelling by subspace identification methods and tyre cornering stiffness estimation

Authors: Tengyue Ba; Xiqiang Guan; Jianwu Zhang

Addresses: School of Mechanical Engineering, State Key Laboratory for Mechanical Systems and Vibration, Shanghai Jiao Tong University, 800 Dong Chuan Road, Minhang District, Shanghai 200240, China ' School of Mechanical Engineering, State Key Laboratory for Mechanical Systems and Vibration, Shanghai Jiao Tong University, 800 Dong Chuan Road, Minhang District, Shanghai 200240, China ' School of Mechanical Engineering, State Key Laboratory for Mechanical Systems and Vibration, Shanghai Jiao Tong University, 800 Dong Chuan Road, Minhang District, Shanghai 200240, China

Abstract: In this paper, subspace identification method is used for vehicle lateral dynamics modelling, which is based on input and output data and describes the vehicle behaviour more accurately. By standard vehicle road tests, the effects of three traditional subspace methods are validated in vehicle modelling. The results show that the MOESP and N4SID methods are effective, and the CVA models have the highest estimated precision. Furthermore, the CVA models from different tests are evaluated in order to obtain a proper vehicle model. By contrast, the CVA-step model and CVA-DLC model provide better estimations on yaw rate and side-slip angle, respectively. Moreover, a tyre cornering stiffness estimation method based on identified model is proposed, which is derived on condition that the vehicle system poles are constant. The result shows that the estimation of Kf and Kr based on CVA models is more accurate.

Keywords: vehicle lateral dynamics; dynamic modelling; subspace identification; road test validation; tyre cornering stiffness; stiffness estimation; yaw rate; side-slip angle; standard road tests; vehicle modelling.

DOI: 10.1504/IJVSMT.2015.073040

International Journal of Vehicle Systems Modelling and Testing, 2015 Vol.10 No.4, pp.340 - 355

Available online: 12 Nov 2015 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article