Predictive Cruise Control of Connected and Autonomous Vehicles: An Adaptive Dynamic Programming Approach

This paper proposes a novel predictive cruise control method for a platoon of connected and autonomous vehicles. The main objective is to minimize idle time of vehicular platoon using information from the traffic lights. First, the reference velocity is determined for each vehicle in the platoon. Second, a data-driven learning strategy, named adaptive dynamic programming (ADP), is employed to develop an optimal state-feedback controller without any prior knowledge of the platooning system dynamics. This resultant controller regulates the headway, velocity and acceleration of each vehicle to accommodate both safety and trip time reduction goals. A numerical simulation is demonstrated to ascertain the effectiveness of the proposed approach.