Signal Timing Optimization Based on Trajectory Data

A fixed signal timing optimization method utilizing historical trajectory data at intersections is proposed. A model applicable to low-resolution trajectory data is developed to optimize fixed signal timing. The model takes the green split of each phase as decision variables, aiming to minimize the total trajectory delay. Based on the proportional principle of uniformly distributed randomly arriving trajectories under different signal timings, the constraints of the model are constructed from three aspects: signal constraints, trajectory signal constraints, and departure time constraints. The model solving is transformed into a mixed-integer linear programming problem. Simulation experiments using SUMO are conducted to validate the effectiveness of the proposed model.