An Adaptive Driver Model for Driving Cycle Prediction in the Intelligent Truck

Abstract

Abstract Intelligent vehicles have significant potential to improve the worldwide traffic situation with regard to both safety and efficiency. Commercial vehicles are an ideal application of intelligent driver assistance systems because of their precisely defined operational limits and professional drivers. For many driver assistance systems, a prediction of the future driving cycle is necessary. This research presents an approach towards predicting the future velocity profile using a gain scheduled driver model together with a longitudinal vehicle model. The parameters of the driver model are estimated during vehicle operation using recursive least squares identification. Assuming repeated operation of the vehicle on the same route, the driver model is supplied with the desired velocity at a particular position and outputs the predicted velocity trajectory. In a case study, the benefit of prediction is shown in a hybrid hydraulic truck with predictive optimized energy management.