Overtaking is a common driving maneuver and also the most complex one. Studying this maneuver is considered to be one of the toughest challenges in the development of autonomous vehicles. Here, a novel overtaking model based on adaptive neuro-fuzzy inference system is proposed. This model is designed for two vehicle classes: motorcycles and autos. The presented model is able to simulate and predict the trajectory of the overtaker vehicle in real traffic flow. In this model, important factors such distance, velocity, acceleration and the movement angle of the overtaker vehicle are considered. Using the field data, the performance of the model is validated and compared with the real traffic datasets. The results show very close compatibility between the field and model trajectory.