Application of Driving Simulation Technology in Calibration of Traffic Simulation Parameters
To address the insufficient accuracy in traffic simulation modeling due to the lack of in-depth consideration of complex driving behaviors,a calibration method for traffic simulation parameters based on driving simulation technology is proposed.The reconstruction and expansion project of Shenzhen Bao'an International Airport Expressway is selected as the case.Using VISSIM simulation software,a comprehensive traffic simulation model of the entire route is constructed,and UC-winRoad software is employed to create the highly realistic driving simulation scenarios.Driving simulation experiments are conducted to extract the typical driving behavior characteristics in complex scenarios.Calibration functions for simulation parameters are established by using the car-following behavior indicators such as headway and driving speed.To address the efficiency of the calibration algorithm,an improved genetic particle swarm optimization algorithm is employed to optimize the calibration parameters of Wiedemann99 car-following model to enhance the analytical precision and application effectiveness of the microscopic traffic simulation model.The results show that,compared to the orthogonal experiments and iterative optimization based on traffic perception data,the proposed calibration method based on driving simulation technology reduces the error in parameter calibration values by 47.2％and 9.1％,respectively,and aligns more closely with the localized driving behaviors,which significantly enhances the credibility of the traffic simulation model.Compared to the conventional genetic algorithms and particle swarm algorithms,the proposed genetic particle swarm algorithm can not only obtain the global optimum but also converge faster by 17 rounds,which can meet the effectiveness and efficiency requirements for the algorithm.
traffic simulationdriving simulationparameter calibrationgenetic & particle swarmcar following behavior
NETL
NSTL
万方数据
