Dynamic predictor-based adaptive cruise control

Abstract

We study the stabilization problem of a platoon of Adaptive Cruise Control (ACC) vehicles in the presence of input-delay. We use a dynamic predictor for input-delay compensation, a filtered version of the standard finite spectrum assignment method that overcomes robustness issues, in particular those raised by the approximation of distributed time-delay terms. Each vehicle must achieve the velocity of the preceding vehicle while ensuring a safe inter-vehicular distance established by a time headway-based spacing policy. To this end, a proportional-integral type controller combined with a dynamic predictor is added to each vehicle in the platoon that guarantees stability and zero steady-state error. String stability property of the closed-loop system, i.e., the platoon’s ability to attenuate fluctuations arising in the motion of the leading vehicle, is analyzed using a frequency domain framework. The effectiveness of the proposed control scheme is illustrated with simulation results of a platoon of five vehicles.