Optimal Connected Automated Vehicle Control in Freeway Merge Segments through Distributed Coordination

Recent advancements in connected automated vehicle (CAV) technologies promise significant improvements in traffic management, particularly in complex roadways such as freeway merge segments. However, achieving these improvements requires the implementation of a systematic control framework to coordinate CAV operations effectively. This paper presents a distributed cooperative optimization algorithm specifically designed to refine the trajectory and lane-changing decisions of CAVs. A vehicle-level mixed-integer nonlinear program is introduced, optimizing discrete lane-changing decisions and continuous lateral and longitudinal acceleration of CAVs. The optimization approach uses a hybrid solution technique that combines linearization with a receding horizon framework. This reduces computational complexity while ensuring adaptability to the traffic system’s dynamics. The algorithm is evaluated using a case study, and it significantly improves traffic flow efficiency. The results showed reductions of up to 93.6% in average delay, 50.0% in speed variation, and 47.6% in fuel consumption. Sensitivity analysis revealed the algorithm’s robustness across varying speed limits, demand levels, and lane configurations. For instance, while higher demand rates severely degrade traffic performance in simulation runs, the optimization consistently maintains low delays and high speeds. This shows the algorithm’s ability to adapt to challenging traffic conditions. In addition, sensitivity tests indicate that design features, such as longer acceleration lanes, reduce speed variations and improve merging efficiency. These results highlight the algorithm’s capability to deliver reliable and efficient traffic management under diverse operational scenarios.