Integrated MPC Framework With Distributed Longitudinal and Robust Lateral Control for 2D Vehicular Platoons

This paper proposes an integrated model predictive control (MPC) framework for two-dimensional vehicular platoons operating under a leader with unknown and time-varying inputs and arbitrarily assigned lateral targets. The vehicle kinematics are decoupled into longitudinal and lateral subsystems to enable targeted control. For longitudinal motion, a distributed MPC scheme is employed, featuring a tailored terminal control law and cost to track the leader's speed while maintaining fixed inter-vehicle spacing. For lateral motion, a tube-based MPC strategy is developed to address system uncertainties arising from longitudinal velocity variations, which directly affect lateral kinematics. An artificial target and an extended terminal region are introduced to decouple the target and terminal design, enabling flexible lane changes with arbitrarily given lateral targets. The proposed framework ensures recursive feasibility and closed-loop stability, with rigorous theoretical guarantees. Numerical simulations demonstrate the effectiveness and robustness of the method across various scenarios involving time-varying leader input and arbitrary lateral targets.