Abstract:

A meta-cellular automaton model for the intertwined region was constructed to analyze the complex vehicle lane change behavior in the weaving area of urban expressways considering the frequency distribution of lane change based on vehicle micro-trajectory data. The overhead video was used to extract the full sample of vehicle lane change information in the weaving area. Considering the actual decision time and safety priority awareness of drivers, the vehicle lane change prediction and lane change spacing in the weaving area were combined, and the lane change motivation and lane change timing rules were established respectively. Subsequently, the lane change timing decisions can be made step by step. The model was evaluated with parameters such as flow rates, density, velocities and channel change frequency distribution. The simulation results show that the constructed meta-cellular automata model is more effective, and the relative errors of flow, density and speed are 0.7%, 1.4% and 1.6%, respectively, and the errors of the number of lane changes in different directions are 2.97%~22.98%, and the distribution of lane changes is basically consistent with the measured data, which can effectively simulate the bottleneck phenomenon in the interweaving area, and reflect the real demand for lane changes and describe the actual operation state.

Key words: transportation planning and management, weaving area, frequency of lane change, cellular automaton