Choi, Junhee (Seoul National University), Kim, Dong-Kyu (Seoul National University)

Vehicle Trajectory Reconstruction with Estimation of Shockwave Speed Using Traffic Sensor and Probe Vehicle Data

Scheduled for presentation during the Regular Session "S01b-Data-Driven Simulation and Modeling for Smart Mobility Systems" (WE-EA-T1), Wednesday, November 19, 2025, 14:50−15:30, Southport 1

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 19, 2025

Abstract

Vehicle trajectories provide both spatiotemporal and attribute information for traffic management. Since traffic sensors and probe vehicles cannot cover all road segments, gathering all observed vehicle trajectories is challenging. Although some studies have proposed data-fusion methods, errors still arise during congestion propagation due to the driver’s reaction times and varying shockwave speeds under different conditions. This study reconstructs vehicle trajectories with shockwave speed estimation. We assume an environment with a virtual traffic sensor and randomly distributed probe vehicles. Our method estimates shockwave speed by detecting jerks in speed profiles of probe vehicles and interpolates shockwave speed using a radial basis function. Trajectories are reconstructed via speed correction and data assimilation grounded in a car-following model. The result of shockwave speed estimation shows the trend between shockwave speed and traffic state. The shockwave speed is higher when the traffic state is under transition flow with deceleration. We evaluate our approach on the Next-Generation Simulation (NGSIM) dataset and demonstrate lower errors than other methods, performing a mean absolute error of 24.76 m in position and 2.44 m/s in speed. This method can support more accurate trajectory reconstruction under various traffic environments.