A Hierarchical Optimal Control for Vehicle Platoon Based on Bidirectional Error Prediction and Saturated Artificial Potential Fields

The Intelligent Vehicle Infrastructure Cooperative Systems (i-VICS) have much positive impact on the cooperative and adaptive platoon control for connected-automated vehicles( CAVs) based on the communication between vehicles and infrastructure. This paper proposes a novel hierarchical control architecture for CAVs platoon under i-VICS, where the upper layer calculates the time headway errors and the lower layer implements an optimal control. To ensure the information sharing among all vehicles in platoon, the bidirectional information flow topology under i-VICS is adopted to enhance the feasibility of cooperation. The sliding window is then introduced to predict the headway errors existing in the vehicle platoon while the distributed controller is designed to implement the optimal control for vehicle platoons based on gradient descent of a saturated artificial potential field. The parameters of optimal controllers are searched based on the proposed coarse-grained optimization. Finally four scenarios in simulation are selected to verify the proposed approach and analyse the results of its effectiveness.