基于效用选择的换车道模型

通过论述当前换车道模型的常见实现方式,指出在换车道需求产生过程中广泛应用的PLC方法存在下列不足:(1)不能反映微观特性;(2)不具有普适性;(3)不能准确反应运行环境变化的影响;并从理论分析的角度指出换车道行为是驾驶员在不同车道行驶状态下满意程度的选择结果,其需求产生的过程可以用随机效用理论进行描述。通过引入随机效用理论,建立基于效用选择的换车道模型,并以其作为核心子模型之一,编写仿真程序并对复杂交通流进行验证,得到较好的效果。

A Lane Changing Model Based on Random Utility Theory

Possibilities of lane changing(PLC)have been used extensively in the field of the traffic microscopic simulation to determine the demand of lane changing activities.However,existing approaches used for determining the actions of the lane changing are mostly ad hoc and suffer from some limitations:(i)unable to characterize realistic driving behaviors;(ii)restricted by the the calibration data set;(iii)insensitive in different traffic situations.In this paper,a random utility method is proposed to describe the decision-making process of a lanechanging action maneuver for an individual.This methodology,which is based on the utility maximization principle,links the lane changing decision of an individual to different traffic characteristics by a series of utility functions.An empirical case study is conducted to demonstrate the capacity of the proposed methodology by using field data set from the Third Ring Expressway in Beijing.Furthermore,the models are validated using a traffic flow simulation.