A convex and robust distributed model predictive control for heterogeneous vehicle platoons

Abstract

The roll out of connected and autonomous vehicle (CAV) technologies can be beneficial for road traffic in terms of road safety, traffic and energy efficiency. This paper addresses the platooning problem of heterogeneous CAVs with consideration of a time-varying leader speed and multi-dimensional uncertainties that include modeling uncertainties and local measurement disturbances. Resorting to a spatial domain modeling approach with appropriate coordination changes and the relaxation of nonconvex constraints, the traditional nonlinear optimal control problem formulation is convexified for improved computational efficiency and ease of implementation. Then, a convex and tube-based distributed model predictive control algorithm (DMPC) utilizing a predecessor-following communication topology is designed with certified theoretical properties, which can be boiled down to DMPC parameter tuning criteria. Finally, numerical results and comparisons against nominal and nonlinear DMPC-based methods are carried out to verify the performance and computational efficiency of the proposed method under different driving scenarios.