Dynamic Relationship based Data-driven Formation Control for Connected Vehicles

This paper investigates the longitudinal formation control problem of vehicles by learning from the connected autonomous vehicle systems. Firstly, a linear data model formulating the dynamic relationships between a vehicle and all its neighbor vehicles is established. Subsequently, based on the linear data model, a data-driven formation control method that encompasses a control law and a parameter updating law is devised. The proposed method is data driven and can learn the protocols of neighbor vehicles. Finally, the validity of the proposed scheme is substantiated via simulations.