A Novel Hierarchical Flocking Control Framework for Connected and Automated Vehicles

Abstract

To resolve the issues of traffic fatality and congestion in modern transportation systems, this paper proposes a novel hierarchical flocking control framework for multiple connected and automated vehicles (CAVs) to integrate path planning, speed profile generation, and nonlinear vehicle dynamics control. Leveraged by the hierarchical configuration, the proposed flocking control framework can accommodate complicated vehicle dynamics with a self-organizing feature. In the high-level layer, applying a 2-dimensional (2D) flocking theory, the cooperative trajectories (paths and speeds) of multiple CAVs are produced via a distributed control based on shared information. In the low-level layer, the trajectory-tracking and vehicle orientation guidance of each vehicle are achieved by a feedforward and feedback control design. In particular, control allocation is established to take the merits of the over-actuated properties of each vehicle. Demonstrated by the one-way three-lane highway simulation results, five CAVs controlled by the proposed flocking control framework can successfully complete a 2D flocking coordination with velocity alignment, orientation control, collision avoidance, and multi-destination cohesion.