Implementing kinematic wave theory to reconstruct vehicle trajectories from fixed and probe sensor data

A data fusion framework is examined to reproduce vehicle trajectories on urban arterials by combining probe and fixed sensor data, and signal timing parameters. The methodology is based on the kinematic wave theory and employs the variational theory for the solution method. However, the original methodology cannot deal with the vehicles coming in and out in the middle of the study section despite the frequent existence of such vehicles in the real world. Therefore, the methodology is extended so as to incorporate the vehicles coming in and out. The proposed method is then applied to real world data and its robustness is confirmed by changing the input data characteristics.