Research On Optimization Of Urban Single Point Signal Control Scheme Based On Analysis Of Traffic Trajectory

With the development of satellite positioning technology,navigation and positioning has become a new way to obtain traffic data.At the same time,travel trajectory information has become a hot new type of traffic data.At present,most researchers estimate the average road speed and travel time based on the travel trajectory data,and use these data to evaluate the traffic state,and judge whether the intersection or road section is in the state of congestion.However,the literature research results of predicting the actual traffic flow are still relatively limited.For modern traffic management and control,traffic flow is an important basic parameter.From this point of view,this paper solves the flow direction of intersection,and then studies the timing optimization scheme.Firstly,based on travel trajectory data,this article first gives the definition of trajectory data and the source of trajectory data in this article,and introduces several common methods for matching trajectory data with real road networks.After analyzing the traffic flow characteristics of the trajectory data according to the results of map matching,it is determined that the traffic flow has three characteristics: randomness,similarity and periodicity.Based on these characteristics,it is believed that Markov chain can be used to predict traffic flow.Secondly,the traffic flow prediction model of this article is introduced.After determining the use of Markov chain to predict traffic flow and flow direction,it is found that the prediction effect of a simple Markov chain cannot meet the prediction needs of the real road network.For this reason,this paper introduces two correction methods for Markov chain,namely weighting Markov chain and using grey prediction model to correct the prediction result of weighted chain.Finally,comparing the three forecasting methods,it is concluded that the mixed Grey-Markov chain has a better forecasting effect.After judging the rationality of time allocation by using the flow scale data,this paper studies the traffic flow of vehicles at the intersection,predicts the queue length of traffic flow combined with the travel trajectory data of vehicles at the intersection,and solves the average delay in the process of vehicle traffic.Combined with the kinematics formula,the green light duration of each phase in the optimized timing scheme is solved.Finally,a complete optimized timing scheme is given.At the end of the paper,the real data of an intersection in Guilin is used to simulate,and the virtual intersection is established by SUMO to generate the timing scheme.The average delay of the intersection before and after optimization is solved,and the scheme comparison is made.The optimization degree is quantified,which reflects the superiority of the optimization scheme.It has a certain guiding role for the use of trajectory data,and provides a new idea for traffic control in the era of big data.