Signalized Intersection Operation Based On Floating Car Data State Assessment

With the development of society,traffic congestion has become a universal phenomenon in the world and a social problem to be solved.The use of floating vehicle detection methods and traffic big data to evaluate the traffic state at signalized intersections in cities is an effective method to gradually improve traffic congestion.In the development of intelligent transportation,GPS floating cars have become a powerful tool for people to study the urban road traffic status with various advantages such as low cost,wide range,and accurate data.They have fully utilized the characteristics of floating car data to conduct data mining and applied them.It has great significance in practice.Based on this,this paper studies the traffic state assessment of signalized intersections in cities through floating car data.Firstly,data collection and sorting were carried out by floating cars.Then the rough error elimination method and historical trend weighted estimation method were used to perform data cleaning and data repair on the problem data that affected data quality,and map matching and path finding were completed.Secondly,the actual state coefficient of the road is proposed,and various impedance factors of the road are calculated and stored in the attributes of the map road network.Combined with the collected data,the three traffic parameters of average speed,delay time and queue length were extracted and verified,and the extended evaluation index of lane separation was proposed for different lanes within a single intersection.Finally,combined with the fuzzy evaluation algorithm and the road network matrix model,this paper proposes a state assessment method for intersections that takes into account the actual state of the lanes.Classical fuzzy assessment,state estimation algorithms that take into account the actual state,and state assessment algorithms for lane extension indicators are used.The assessment and comparison of the selected intersections are carried out.After analyzing the data of the examples and comparing the results,the effectiveness of the proposed algorithm is verified with VB and MALAB.