Comparative Study between the Magic Formula and the Neural Network Tire Model Based on Genetic Algorithm

In this study, we will investigate and compare the performance of some appropriating methods for tyre force under combined representation of longitudinal and lateral force, which the appropriating methods performance has been introduced with two models: the Magic Formula and Artificial Neural Networks (ANN). The identification of the values of these two models parameters with starting up from a set of experimental available data, results quite difficult. The present work shows a genetic algorithm based methodology for the determination of these coefficients. After a description of the general genetic algorithm procedures, the applications to the case of the model of the tyre are presented. Moreover, we conducted a comparison based on actual date sets. The simulation results obtained have shown that ANN performed better than the other.