Cooperative predictive control for arbitrarily mixed vehicle platoons with guaranteed global optimality

This paper studies the cooperative control problem of a mixed vehicle platoon, which is composed connected autonomous vehicles (CAVs) and human-driven (HDVs) in an arbitrary order. An alternating direction method multipliers (ADMM) based distributed model predictive (DMPC) algorithm proposed for CAVs to lead platoon travelling string with anticipated inter-vehicle spacing desired velocity. First, divided into multiple interrelated sub-platoons any two adjacent having common CAV, then generic constructed each sub-platoon on intelligent driver HDV kinematic respectively. Second, local MPC controller designed optimize inputs objective minimizing position velocity errors all sub-platoon, ADMM utilized obtain global optimal solution among controllers sub-platoons. Finally, numerical simulations experiments are carried out verify effectiveness DMPC algorithm, results reveal that it can reduce computation cost significantly ensure performance comparable centralized algorithm.