Modeling Of Traffic Flow Theory Based On Driver’s Behavior

Driver’s behavior is one of the most important part of traffic flow theory. The study of traffic flow model considering driver’s behavior is a supplement for traffic flow theory. Taking driver’s behavior analysis as the research breakthrough point, headway is the most important parameter in car following theory, which directly reports the mental activity. The relationship between flow, density and speed, is the fundamental of traffic flow theory, performing the cluster behavior of individual drivers.Based on the analysis of driver’s behavior, this paper treats headway distribution and speed-density relationship as the main objects of the research, and takes mechanism analysis, model construction, calibration validation as research process. Based on model mechanism and function form, we classify the existing models, discuss the origin, development and construction of these models, and summarize function form, related parameters and typical features. We advise paying more attention to four aspects in modeling, i.e. the design of model structure, parameter calibration, the driver’s behavior and the characteristics of field data, which provides the basis for the following modeling research.In order to analyze the driver’s car-following behavior, we introduce safety-sufficiency penalty mechanism to analyze the formation and change of headway, and propose a family of headway distribution models, which effectively describe driver’s following safety and driving efficiency. Each parameter in the proposed headway distribution model can characterize the effects of safety and efficiency and the model matches the field data well. In model comparison, we propose the KS test pass rate and analyze the influence of sample size and shifted headway. The proposed model performs satisfactorily in comparison with other models. It indicates that the safety-sufficiency penalty mechanism is effective to describe driver’s behavior.To further characterize the speed-density relationship, we make a basic assumption that energy conservation exists between the psychological potential and the vehicle’s kinetic energy in the driver’s psychological field based on the driver’s mental activities. A virtual spring is used to describe the storage and release of psychological potential energy. Under the above conditions, we establish a macroscopic traffic flow model with conservation law. Each parameter in the new model is physically meaningful and explicit. Additionally, we observe section data and collect fixed location data to validate the versatility and generality of the proposed model. The model can fit field data consistently well both in free-flow and congested situations, which offers scientific evidence to the validity of the model and the rationality of the energy conservation concept.Considering the actual acquisition method of fixed-location data and section data, we explain the difference between the measured data and theoretical demand of speed-density relationship, and analyze the effective application of Logistic speed-density curve. Starting from the initial concept of logistic model, we define the rate of speed change, apply traffic flow state to modify the original assumption, and derive a family of logistic related speed-density models in simple mathematical forms with physically meaningful parameters.This dissertation introduces safety-sufficiency penalty mechanism, energy conservation theory and logistic origin assumption to analyze driver’s behavior, and proposes a series of traffic flow models, which improves the understanding of traffic flow, suggests the selection of the models, and provides a new theoretical foundation.