Fuel efficient quasi optimal adaptive cruise control by control identification

Abstract

Adaptive cruise control (ACC) systems are designed basically to improve comfort, safety and traffic smoothness. The expected diffusion of vehicle-to-vehicle (V2V) communication systems allows now adding additional information on the expected behavior of nearby vehicles, thus offering the possibility to extend the targets of ACCs to include fuel efficiency. In earlier works, it has been shown both by simulations and measurements that the corresponding potential can be quite high. However, the on-line implementation of the method can have prohibitive costs. Against this background, this paper suggests a different solution based on the approximation of the results of an optimization problem. This is done by using a Hammerstein-Wiener model combined with additional control loops to ensure safety related constraints. The method is shown to provide almost the same performance as the original problem while requiring only a fraction of the computational effort.