Extended adaptive cruise control based on multiple target information

Abstract

Adaptive Cruise Control (ACC) is an improvement of regular cruise control that has been available in premium cars for over 10 years. With advanced sensing systems, based on for instance RADAR and camera sensors, ACC makes it possible for a vehicle to follow the preceding vehicle at a safe distance by automatically adjusting the velocity. In this thesis, it is investigated whether more information from the environment may be used to improve the ACC. By using information from vehicles in adjacent lanes and vehicles ahead in the same lane, the ACC is extended with two new functions; Lateral prediction and Multi-target control. Lateral prediction is used to predict the future position of vehicles in the adjacent lanes. When a vehicle from an adjacent lane is predicted to join the same lane, the ACC equipped vehicle can change target in the control law faster. Multi-target control is introduced by letting the ACC use information from more than one preceding vehicle in the control law. With the extra information it is possible to earlier detect and react to traffic disturbances further ahead in the lane. The extended ACC is evaluated with both simulations and in-vehicle tests and it is shown that it has advantages concerning driving comfort and safety over the standard ACC.