Distributed Data-Driven Model Predictive Control for Heterogeneous Vehicular Platoon With Uncertain Dynamics

Abstract

To alleviate the adverse effects of heterogeneous vehicular platoon (HVP) with uncertain dynamics, a distributed data-driven model predictive control (DDMPC) strategy is proposed in this paper. A data-driven model is established with subspace identification using the input-output (I/O) vehicle trajectory. We integrate the data-driven model with the distributed model predictive control (MPC) algorithm to optimize the HVP control. Then, a DDMPC optimal scheme is designed with a target equilibrium and a pair of initial/terminal constraints. Its recursive feasibility and exponential stability are guaranteed by an I/O-to-state stability (IOSS) Lyapunov function and an optimal sum cost function. The string stability analysis of HVP also is provided. Finally, several experiments with heterogeneous vehicular platoon demonstrate the effectiveness of the proposed DDMPC strategy.