A Hierarchical Approach for Integrating Merging Sequencing and Trajectory Optimization for Connected and Automated Vehicles

This paper presents a hierarchical tactical merging optimization (HTMO) approach for connected and automated vehicles (CAV) at freeway merging segments. The proposed approach comprises two layers: a merging sequencing layer and a trajectory optimization layer, which are coupled by a hierarchical model that utilizes sequence set and vehicle state variables. In the merging sequencing layer, a sequence set variable is introduced to simplify the sequence space and identify the optimal merging sequence using a customized tabu search algorithm. In the trajectory planning layer, we formulate a two-point-boundary optimal control model for CAV trajectory planning, incorporating a platoon formation strategy to further enhance travel efficiency. To handle outliers caused by variations in the preceding vehicle’s trajectory, we have developed a heuristic trajectory optimization algorithm to ensure the generation of feasible trajectories with predetermined optimal acceleration values as proposed. Numerical experiments conducted demonstrate the robust convergence performance and computational efficiency of the HTMO approach across different arrival flow scenarios and parameters setting, thanks to its utilization of a rolling horizon strategy. Additionally, when combined with the platoon formation strategy, the schedule produced by HTMO significantly reduces total travel time and delay, as evidenced by our findings.