Tire-Road Friction Coefficient Estimation Based on Fusion of Model- and Data-Based Methods

Abstract

The tire-road interaction generates vehicle driving forces, which affect vehicle performance such as maximum acceleration and stability. Sequential extended Kalman filter (S-EKF) integrated with a slope method has been used for tire-road friction coefficient estimation with its own limitations, along with several “cause-based” and “effect-based” methods. This research proposes a new stochastic-based evaluation criterion using existing vehicle sensor signals with the help of the data-driven Kriging model. The proposed estimation method is validated by both CarSim™ simulation and experimental studies, respectively, under different road conditions. The results show that the proposed novel criterion has a strong correlation with the road friction coefficient and provide an improved tire-road friction coefficient estimation. A signal fusion estimation scheme based on both S-EKF and proposed evaluations is developed to improve estimation robustness.