The problem of tyre model coefficients identification using minimum test data is studied in this work. To obtain these tyre model parameters, an intense research effort by the automotive community has been made and there are different methods to fit the values of these parameters. This problem is addressed in this work through a coevolutionary algorithm that interactively searches the optimum tyre model parameters and new test data in disagreement with the tyre model. The algorithm is composed of two stages: the estimation phase, which finds out the tyre model parameters which can predict actual tyre test data, and the exploration phase, which finds out new test data which have the most disagreement with the response of the current model. The feasibility of the methodology is demonstrated comparing the obtained results with other known techniques.
