A Mesoscopic Traffic Simulation Model Construction Method for Mixed Traffic Flow Operation Scenarios

Abstract

The invention provides a method for constructing a mesoscopic traffic simulation model oriented to a mixed traffic flow operation scene. The method constructs a mixed traffic flow driving model of ordinary vehicles and public transport vehicles, a bus vehicle entry and exit model, a mixed traffic flow point queuing model, and a mixed traffic flow point The conversion model can support the mesoscale simulation of many types of typical scenarios under the operation of mixed traffic flow; restore the vehicle trajectory information through AVI data, and obtain the actual travel time of vehicles entering and leaving the road section on each road section, as the calibration information set of the simulation, and then, to realize Guided by the equivalence of the travel time of the vehicle entering and leaving the road section, the corresponding calibration method is designed to restore the road section level information during the vehicle travel process to the greatest extent while highly simplifying the mesoscopic simulation model.

Description

Technical field [0001] The invention relates to the field of intelligent transportation, in particular to a method for constructing a meso-traffic simulation model for a mixed traffic flow running scene. technical background [0002] Traffic simulation is an important means to support the accurate simulation, quantitative analysis and feedback adjustment of traffic system running state. In the aspect of traffic system simulation model, the macro model started earlier, and the simulation of large-scale network can get satisfactory aggregation results, but the scale of coarse particles can't simulate the travel process of individual vehicles. Furthermore, it is a micro-simulation model with vehicles as the description unit, which can describe the car-following motion among vehicles in detail, but inevitably involves a lot of tedious and time-consuming parameter calibration work. As the advantages of both macro and micro models are integrated, meso-traffic simulation has gradually b...

AI Technical Summary

Problems solved by technology

[0005] (2) Existing mesoscopic simulation models, such as MesoTS, DynasTIM, Mezzo, etc., basically use traditional cross-section aggregate data or low sampling rate individual data as calibration information, and only a few scholars have discussed the use of vehicle identity perception data in Innovative role in visual simulation modeling and calibration

Due to problems such as the inability of the cross-section aggregate detector to track travel trajectory information and the low penetration rate of vehicles sampled and detected, it is impossible to fully obtain the traffic travel demand and operating status, and it is difficult to support accurate simulation model parameter calibration.

Images