Trajectory Data-Driven Dynamic Platoon Dispersion Model Considering Platoon Internal Diffusion Behaviors

Connected vehicle (CV) trajectory data provides an intuitive reflection of arterial traffic progression, holding significant potential in capturing platoon dispersion. This study proposes a trajectory data-driven platoon dispersion model considering platoon internal diffusion behaviors. The movement of a platoon is separated into the forward transports and internal diffusion behaviors. A cellular transmission model (CTM)-based method is used to simulate the progression of a platoon. Historical trajectory data is used to predict the platoon speeds. Real-time trajectory data is used to calibrate the diffusion coefficient and predict the progression behavior over a future time range. Simulation studies validate the advantages of the proposed model over the CTM and Robertson’s model. Results of sensitivity analysis indicate that (1) collecting trajectory data for more than 25 s, including the first 5 s of the green interval, is recommended, and (2) the model is applicable under the partially CV environment with a satisfactory performance.