基于自然驾驶数据的高速公路出口区换道风险模型

Lane-change Risk Model of Freeway Off-ramp Areas Based on Naturalistic Driving Data

摘要:

依托上海自然驾驶实验数据，提取高速公路出口区的交通流特性数据和驾驶行为样本.以单向四车道高速公路出口为例，考虑不同道路交通环境和出匝换道路径因素，基于比例优势模型构建高速公路出口区换道风险模型.在该模型基础上生成换道风险图谱，解析高速公路出口区换道风险分布特征，为车流控制管理提供理论决策支持.

Abstract:

Based on the data from Shanghai naturalistic driving study (SH-NDS), traffic flow characteristics data and driving behavior samples in freeway off-ramp areas were extracted. Illustrated by the example of 8-lane freeway off-ramp, considering the road traffic environment and exiting lane-change route for modeling purpose, a proportional odds model was adopted to fit the lane-change risk model of off-ramp areas. Based on the model, the lane-change risk graphs were drawn to visually describe the lane-change risk distribution characteristics, providing a theoretical decision support for traffic flow management.