Lane-changing Behavior Modeling And Traffic Flow Characteristics Analysis Based On Cellular Automaton

Traffic congestion has become an important factor restricting social and economic development.In order to effectively alleviate traffic congestion,it is necessary to comprehensively explore the driving behavior characteristics of drivers and deeply understand the evolution laws of traffic systems in different environments.But the traffic system is an extremely complex giant system,with individual interactions and constraints within it,making related research extremely challenging.Lane changing is a common driving behavior in road traffic systems and one of the incentives for the complex characteristics of traffic flow.In order to clarify the specific impact of lane changing behavior on the traffic system,this paper,based on relevant knowledge such as system methodology and system evolution theory,simulates and studies the impact and mechanism of different lane changing behavior on the evolution process of traffic flow in specific scenarios,in order to summarize some universal conclusions and provide theoretical basis for traffic management and control.The main research work of this paper is as follows:(1)In order to explore the impact of lane changing process on the traffic flow in a two-lane road traffic system,this paper establishes a two-lane lane changing cellular automaton traffic flow model based on the analysis of lane changing trajectory data,and studies the general laws and effects of lane changing behavior.Preprocess real trajectory data,eliminate abnormal data,and extract lane changing trajectory data.Considering the presence of significant noise in the data,wavelet denoising methods are used to denoise the data and make it smoother.Use the processed data to statistically analyze the lane changing angle and lane changing duration of vehicles,summarize their correlation,and establish an expression for the relationship between vehicle lateral displacement,speed,and lane changing angle.Furthermore,a two-lane cellular automaton traffic flow model(TCA-H model)considering lane changing process and unsuccessful lane changing behavior is proposed.The simulation results are used to study the fundamental diagram,lane changing frequency,lane changing duration,proportion of unsuccessful lane changing,and relevant results under mixed traffic flow conditions,in a two-lane road scenario.The reliability of the results is verified with actual data.(2)In order to explore the characteristics of driving behavior in a two-lane road traffic system at a deeper level and improve the TCA-H model accordingly,this paper uses machine learning models to predict car following and lane changing behavior,and embeds the prediction results into the rules of the cellular automata model to improve the accuracy of rules and simulation results.Establish long short term memory models and support vector machine models that can predict the next driving situation of vehicles,to explore the characteristics of car following and lane changing behavior.Train and test the model using real vehicle trajectory data,and embed the trained model into the cellular automata model to improve the update rules.Then,construct a lane changing model based on prediction rules(P-TCA model)as an improvement of the TCA-H model.Finally,compare the simulation results with actual data to verify the effectiveness of the model,and study the fundamental diagram,lane changing frequency,lane changing duration,and the proportion of unsuccessful lane changing.(3)To investigate the merging behavior in the on-ramp system and the impact of different collaborative merging strategies on the evolution process of traffic flow,this paper establishes updating rules for cellular automata models that consider the merging process of vehicles and collaborative merging strategies,and studies the results of speed and flow under different conditions.Summarize three typical vehicle collaborative merging strategies in an interconnected environment,analyze the collaborative merging process between main road vehicles and merging vehicles,and propose an merging model.Through simulation,compare and analyze the average speed of vehicles and the average flow of different road sections under the implementation of merging strategy and without merging strategy.Set the different values of merging safety distance parameters and study its impact on the average flow.(4)To investigate the impact of overtaking behavior on the evolution of traffic flow in a two-lane two-way road traffic system,a cellular automaton traffic flow model considering overtaking process was established based on the analysis of overtaking behavior and deceleration behavior of opposite vehicle.And the simulation was carried to study the impact of overtaking behavior on traffic flow under different parameter settings.By analyzing overtaking scenarios and overtaking intentions,a framework and method for characterizing the overtaking process of vehicles are proposed.Vehicle entry rules,forward rules,overtaking rules,unsuccessful overtaking rules,and deceleration rules of opposite vehicle are formulated,respectively.A cellular automaton traffic flow model considering the overtaking process is constructed.Simulates and studies the average travel time,flow diagram,spatiotemporal diagram,overtaking frequency,and proportion of unsuccessful overtaking vehicle under different truck penetration rates,deceleration probabilities of opposite vehicle,overtaking duration(representing time pressure),and safe return distance.