The tire slip angle estimation algorithm based on intelligent tire technology

Abstract

Tire slip angle is a very important parameter in tire/vehicle dynamics control. This paper proposes a tire slip angle estimation method that combines intelligent tire technology and machine learning. Firstly, a finite element model of a 205/55/R16 radial tire was established by ABAQUS software, and the tire finite element model is verified by the radial stiffness experiment and dynamics experiment. Secondly, the curves of lateral acceleration and the curves of the lateral displacement obtained by five virtual tri-axial accelerometers installed on the inner line of the finite element tire under different slip angles, tire pressures, loads, and speeds were analyzed. Finally, combined with linear correlation analysis method, the promising input eigenvalues were determined, and three slip angle prediction models were trained based on the same set of train sets to predict the same set of test sets. The prediction results showed that the slip angle prediction curve of BP model has the highest degree of coincidence with the actual curve, and the mean absolute percentage error is 3.55%, indicating that the slip angle estimation algorithm proposed in this paper is feasible, which is very important for the stability control of the vehicle.