Analysis Of OD Experience Trajectory Based On Taxi Trajectory Data

Route planning is an important daily activity.Due to its extensive application,it has been deeply studied by many scholars.The path planning problem in road networks can be regarded as the shortest path problem in graph theory in most cases.With the development of floating car technology,the collection of traffic information is more convenient,and the taxi location data is more convenient and accurate.In current research in the field of transportation,the shortest path obtained by the shortest path model in the past is only the "shortest path of theory" obtained by the model algorithm,and it is still different from the actual path of individual subjective judgment.As an important participant in urban transport,taxi drivers have a wealth of driving experience and mileage,and they have their own unique insights into the routes and road conditions.In addition to taking into account the distance and time of the route,taxi drivers also effectively reflect other external factors such as road width,number of lanes,number of traffic lights,and the number of corner crossings,and effectively resolve these factors and obtain drivers.The best path during actual driving.However,these external factors are difficult to integrate into the existing path planning system.If the taxi driver's driving experience can be extracted and applied to the existing route planning method,it can greatly help other motorist drivers find the best route and help the relevant departments to alleviate traffic problems.If a taxi passes frequently on a road section,it may be considered that this road section is favored by the drivers.The chance of passing through this road section is very high.Therefore,the concept of road section frequency can be used to indicate the popularity of a road section.In some previous studies,the calculation of the road segment frequency did not consider the constraints of the starting point(O)and the destination(D).Therefore,under the specific OD conditions,the global road segment frequency contains many "impurities" and needs further "purification".The OD-restricted road section frequency can more accurately capture the taxi driver's experience trajectory at a specific OD.The main tasks of this article are:(1)Data preprocessing,classifying the original data,screening out bad values,and matching the original trajectories to enable accurate visualization on Google Maps.(2)Through analyzing and excavating historical trajectory data of taxis in June in Beijing,the space-time distribution law of taxis carrying passengers wasexplored and summarized,and the collection of hotspots with frequent OD trips was obtained.(3)Consider the traffic conditions in different time periods,divide the day into 12 time periods,and categorize the trajectory data sets by time period.For the sections in different time segments,the global traffic flow and the traffic flow under the OD direction constraint are respectively counted as the global frequency and OD frequency of the road section.A local empirical subnet is established based on the section where the trajectory passes within the OD interval,and an empirical trajectory is generated within this empirical road network.The results show that the generated experience trajectory can fully reflect the driver's choice.It is also proved that the OD limited trajectory frequency is more suitable than the global frequency to reflect the taxi driver experience.(4)Analyze the relationship and difference between the OD experience trajectory and the global experience trajectory,the shortest length and the time trajectory.