Estimation of tyre forces using smart tyre sensors and artificial intelligence

Abstract

In-tyre strain measurements from a smart tyre sensor system are analysed using two artificial neural network types, in order to estimate tyre forces. A tyre finite element model is used to calculate in-tyre strain (inputs) and tyre forces developed at the wheel centre (outputs) for use in the neural networks. Neural networks are trained on pure slip conditions and tested on combined slip events with the goal of accurately predicting tyre longitudinal and lateral forces and the aligning moment. The large mapping function is fitted using multilayer perceptron and radial basis function networks. Results from the best radial basis function network design are excellent, with training times under one minute, testing times of milli-seconds and calculated tyre forces within 1%. The conclusion is that radial basis function networks can be used effectively for real time analysis of strain sensor measurements in a smart tyre sensor system.