Microscopic Modeling And Properties Analyzing Of Mixed Traffic Flow Considering Driver-vehicle-road Interaction

Mixed traffic flow is still one of the major characteristics of urban traffic in China. Compared to homogeneous traffic, conflicts among different travel modes are more frequent and complex and the traffic flow characteristics as well as operational conditions are seriously changed. However, owing to the complexity of mixed traffic scenarios, some specific driving behaviors in the microscopic modeling process are more likely to be neglected and the understanding of vehicle characteristics as well as road conditions is also insufficient, which will further cause deviations between the simulation results and actual traffic conditions. In this situation, the objective that urban traffic problems will be solved scientifically is even more difficult to achieve. Therefore, by considering driver-vehicle-road interaction mechanism and their impacts on driving behavior, modeling and simulation of mixed traffic flow are of great significance and value to depict the vehicle behavioral dynamics, understand the internal interference mechanism of traffic flow, enhance the knowledge of microscopic modeling system as well as reproduce the specific mixed traffic characteristics of China.Inspired by the related research on the microscopic modeling of mixed traffic flow, and taking into account the effect of driver-vehicle-road coupling mechanism on establishing appropriate traffic flow models and understanding the natural properties of traffic flow, this dissertation focuses on the influence of some special microscopic behaviors, physical characteristics as well as road traffic environment factors. Then several typical mixed transportation systems are chosen and studied through cellular automata modeling and simulation. The main research work of this dissertation can be summarized as follows:(1) A two-dimensional cellular automaton model is established to study typical characteristics of mixed bicycle traffic induced by lateral drift and accompanying variable lateral clearance maintaining behavior.Based on derived positive correlation between passing speed and drift speed through survey, the occurrence probability of lateral drift and the degree of maintained lateral clearance are both introduced in accordance with the variance of passing speed. Thus, the model enriches the descriptions of the bicycle dynamic characteristics and gets ride of using the typical lane-based modeling framework. Simulation results show that the lateral drift manner will increase the interactions between bicycles, increasing lateral movement frequency in low density region, while reducing traffic flux and passing events in intermediate and high density region. In addition, complex impacts of lateral drift behavior on the temporal-spatial dynamics of mixed bicycle flow are observed, i.e., it causes speed fluctuation in low density while alleviating the speed fluctuation in relatively high density. It is further verified that speed distortions in intermediate and high density region induced by assumption of constant lateral clearance requirements can be revised by introducing its variability characteristics. Then, firm conformity between simulation results and actual survey data is reached in whole density region.(2) A mixed bicycle flow cellular automaton model considering desired speed variation is established, which is based on an analysis method that desired speed distribution characteristics can be obtained from the statistical analysis of samples with free speeds and a specified slow down probability value. Calibration and validation results verify that the introduction of desired speed diversity into the proposed model will be helpful to improve the simulation accuracy and the model generality effectively, as firm conformities between simulated speed values and actual survey data in varying road traffic conditions, including road width and ratio of electronic bicycle, are observed. Numerical simulation results further prove that for each type of passing events except passive passing, its frequency is reduced as ratio of electronic bicycle increases from each bicycle type’s perspective, while opposite trend is observed in total frequency if bicycle types are not distinguished. It reveals that the major treatment method that regards conventional bicycle and electronic bicycle as the same kind of traffic service object will result in the distortion of mixed bicycle level-of-service(LOS) estimation. Results of this study can provide some theoretical guidance for determining bicyclist’s LOS in a mixed bicycle traffic environment.(3) Considering illegal lane-changing behavior of bicycles on road sections with motorized vehicle and bicycle lanes separated by traffic markings, a new multi-lane cellular automaton model is proposed to investigate impacts of bicycle characteristics and vehicle’s honk effect on mixed traffic dynamics. Numerical simulation results indicate that, as a type of backward stimulation, honk can effectively decrease the likelihood of illegal lane-changing behavior of bicycles and create impacts on mixed traffic in intermediate density region. In other density ranges, no obvious effects can be observed. By considering part of bicycle traffic characteristics which can intensify the conflicts between bicycles, including lateral drift behavior and the mixture with electornic bicycles, the conflicts between bicycles and motorized vehicles are also observed to become more serious.(4) By incorporating the typical lateral shift behavior of vehicles towards roadway centerline, a new multi-lane CA model is established to analyze traffic characteristics on a bidirectional two-lane road segment with on-street parking. Then, traffic phenomenon that through vehicles are apt to keep appropriate lateral distances to parked vehicles to support desired speeds is reproduced. Numerical simulation results indicate that this phenomenon can help reduce the negative effect of on-street parking obviously. In addition, as lateral drift is restricted by opposite traffic situation, compared to one-way parking, the negative effect of two-way face-to-face parking might be more serious. Subject traffic flow in condition that opposite parking area locates downstream the subject one is higher than those in upstream and face-to-face conditions. The above achievements may help to offer scientific guidance for on-street parking arrangements and operations.(5) The combined effect of the U-turn intersection and its nearby bus stop is investigated by using a multi-lane CA model. Four cases that the bus stop locates pstream or downstream U-turn intersection, and ones with the same moving direction or not are considered. Simulation results show that if the distance between the bus stop and the U-turn intersection is lower than a critical value, combined bottleneck has significant negative effect on the performance of straight and U-turn traffic. Comparisons among four cases reveal that, in both of the scenarios that buses and U-turn vehicles are in the same moving direction with low opposite vehicle entering probability and buses and U-turn vehicles are in the opposite moving direction, the downstream bus stop always performs better than the upstream one. Thus, in traffic engineering, according to the specified simulation results based on certain traffic environment factors including U-turn flow, ratio of bus and so on, negative effect of the combined bottleneck can be minimized by adjusting both the relative location and the distance of two specified transportation facilities. These findings may be useful to offer scientific guidance for the management and design of traffic networks.