Research On Variable Speed Limit Control Strategy Considering Microscopic Driving Behavior In Intelligent Network Environment

With the constantly development of China’s social and economic level,the government is vigorously developing the construction of transportation infrastructure in order to meet the increasing traffic needs,in which the road traffic gradually formed a road transportation system with expressways as the skeleton.Expressways can provide high efficient transportation due to the characteristic of ‘high speed’,however,the accompanying increase in traffic accident problem is increasingly serious.Variable Speed Limit(VSL)control system,as a representative technical means to improve traffic safety in intelligent transportation system,,has been widely concerned in the field of traffic safety research.The arrival of the information era has promoted the development of the transportation system to the direction of digitization,networking and intellectualization.The assumptions of a new generation of transportation system is to build a fully connected intelligent driving system with complete information interaction.With the development of intelligent transportation system,it is inevitable that there will be a long period of mixed driving between connected and automated vehicles and manual driven vehicles in the near future.How to effectively combine the active safety control technology such as VSL with the advanced connected and automated driving technology,then further improve the expressway traffic safety level under the mixed flow scenario is a research hotspot.Therefore,in this thesis,the VSL control strategy is deeply studied considering microscopic driving behavior for the mixed flow of connected and automated vehicles and manually driven vehicles.The main research achievements of this thesis include the following aspects:(1)A micro-simulation platform for car-following scenario was first built based on MATLAB software,then the parameters of the car-following model were calibrated using the real microscopic trajectory data.Furthermore,three basic longitudinal driving behaviors were combined to construct simulation scenarios.Besides,the relationship between vehicle behavior and longitudinal rear-end collision risk was established by applying surrogate safety measures.The influence of individual heterogeneity on freeway longitudinal safety was investigated by several simulation experiments,and the effect of individual heterogeneity was discussed from two aspects.(2)A VSL control system based on CAV was proposed.Firstly,the influence of human driver’s heterogeneity on VSL control performance was analyzed.Extensive simulation experiments were conducted for different CAV market penetration rates and different driver heterogeneity rates to test the effectiveness of the proposed CAV-VSL control system in onedimensional vehicle dynamics scenarios.(3)A VSL control strategy based on microscopic driving behavior was proposed,the Python software was adopted to build a micro-simulation platform,and a simulation scene facing the weaving area at off-ramp terminal is constructed.Based on the strategy,the CAV-VSL control system proposed in Part 2 is improved,and the effectiveness of the proposed strategy on system safety improvement in a two-dimensional vehicle dynamics scenario is verified.There are 28 figures,6 tables,and 132 references...