Tire-road Friction Coefficient Estimation Based on Longitudinal Measurements

Juqi Hu,Subhash Rakheja,Youmin Zhang
DOI: https://doi.org/10.1109/icamechs.2017.8316533
2017-01-01
Abstract:This paper investigates a new tire-road friction coefficient (TRFC) estimation method based on longitudinal measurements (only wheel angular velocity and vehicle longitudinal acceleration are required). Based on these measurements, vehicle longitudinal velocity is firstly estimated from a Kalman filter (KF), which is responsible for tire slip ratio calculation. Longitudinal tire force and normal tire force can also be calculated based on longitudinal dynamics and a static force model, respectively. Considering the problem of nonlinearity and slow convergence rate, a constrained unscented Kalman filter (C-UKF) method is designed based on the brush tire model to identify the TRFC. In order to evaluate the effectiveness of the proposed method, vehicle accelerating straightly on two different roads are simulated using CarSim and Matlab/Simulink. The results show a robust tracking perfomance and good stability properties. In addition, the computational complexity is substantially reduced since only one parameter is required to be identified. This method may enable a cost-effective and easy implementation for real vehicle applications.
What problem does this paper attempt to address?