A method for identifying a key bottleneck of urban road section based on congestion propagation

Abstract

The application discloses a kind of urban road section traffic key bottleneck identification method based on congestion propagation, fusion taxi GPS trajectory data and network car GPS trajectory data, by extracting congestion event and congestion activation event, find congestion propagation link;By defining a congestion cost function, the influence caused by road congestion is quantitatively evaluated, and the key congestion bottleneck is found according to the influence degree.The whole method flow step is clear, easy to operate, and has good application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of urban traffic management, and particularly relates to a method for identifying a key bottleneck of urban road section passing based on congestion propagation. BACKGROUND

[0002] With the continuous development of the city, the load of urban roads is becoming more and more serious. In order to improve the efficiency of road operation, the problem of congestion needs to be solved. The degree of road network smoothness is directly restricted by the bottleneck of road section passing. Serious passing bottleneck will cause the efficiency of road network passing to decrease sharply. Identifying the key bottleneck of road section passing is very important for supporting traffic congestion management decision making and improving the efficiency of road operation of the whole city. Unlike the traditional research idea of only considering the local congestion road section, the method defines congestion events and congestion activation events, extracts complete propagation links for analysis. In addition, the existing research method lacks a quantitative evaluation method for the loss cost of road network operation caused by congestion links. The method defines a congestion cost function to quantify the total cost formed by the congestion links as an important basis for determining the keyness of the congestion links. Finally, the extracted congestion propagation link results are analyzed to find the starting road section of the most critical congestion link, and the road section passing bottleneck is identified. SUMMARY

[0003] Therefore, one of the purposes of the present application is to provide a method for identifying a key bottleneck of urban road section passing based on congestion propagation.

[0004] The purpose of the present application is realized by the following technical solutions.

[0005] A method for identifying a key bottleneck of urban road section passing based on congestion propagation, comprising the following steps:

[0006] Step S1: data acquisition: collecting vehicle GPS trajectory data, including taxi trajectory data and online car GPS trajectory data, for representing road conditions, and collecting road network data;

[0007] Step S2: data fusion: fusing the collected online car GPS trajectory and taxi GPS trajectory data to unify the data dimensions;

[0008] Step S3: map matching: matching the GPS trajectory to the road section;

[0009] Step S4: road congestion state determination: calculating the passing speed of the road section, analyzing the speed-time distribution law of the road section, and determining the saturation critical speed of the road section, i.e. the speed determination threshold when the road section is congested;

[0010] Step S5: extracting congestion events: calculating the average speed of each road section at each time slice, and regarding the set of continuous N time slices from the formation of congestion to the dissipation of congestion of a road section as a congestion event;

[0011] Step S6: extracting congestion activation event: if two adjacent road segments enter congestion state successively, and the congestion state of the downstream road segment lasts after the congestion of the upstream road segment is formed, it is determined that the congestion is propagated from the downstream road segment to the upstream road segment;

[0012] Step S7: forming congestion propagation link: combining the congestion event and the congestion activation event, the complete propagation link is formed by connecting the congestion events through the congestion activation events;

[0013] Step S8: quantitatively evaluating the congestion cost of the congestion propagation link:

[0014] Step S9: identifying road segment traffic key bottleneck: analyzing the obtained congestion link-congestion cost distribution diagram, extracting the congestion links greater than the congestion cost threshold, and analyzing the time distribution of the congestion links to extract the key bottleneck road segment.

[0015] Further, in step S2, the original GPS track points are processed by using a hidden Markov model map matching method.

[0016] Further, in step S8, in order to quantitatively reflect the loss degree of the congestion propagation link to the road network operation, the length of the congestion road segment, the number of lanes of the congestion road segment and the congestion duration are considered as factors, and the total cost calculation formula of the congestion link is as follows:

[0017]

[0018] wherein Cost k is the congestion cost of the kth congestion propagation link, D k is the set of congestion events corresponding to the kth congestion propagation link, Len p is the length of the road segment r corresponding to the congestion event p, CDS p is the number of lanes of the road segment r corresponding to the congestion event p, and T p is the congestion duration of the road segment r corresponding to the congestion event p. k k r r p

[0019] The second object of the present application is to provide a computer device comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the method as described above when executing the computer program.

[0020] The third object of the present application is to provide a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement the method as described above.

[0021] ​​​​​The method for identifying a key traffic bottleneck of the application has the advantages that the method for identifying a key traffic bottleneck is easy to operate, the GPS trajectory data of a taxi and the GPS trajectory data of a network car are fused, a congestion propagation link is found by extracting a congestion event and a congestion activation event, a congestion cost function is defined to quantitatively evaluate the influence caused by a road congestion, and a key congestion bottleneck is found according to the influence degree. The whole method process is clear, easy to operate, and has good application value.

[0022] Additional advantages, objects, and features of the application will be set forth in part by the description that follows, and in part will become apparent to those skilled in the art upon examination of the following specification or can be learned by practice of the application. The objects and other advantages of the application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to make the objects, technical solutions and advantages of the application clearer, the following will further describe the application in detail with reference to the drawings, in which:

[0024] Figure 1 is a flow chart of the method for identifying a key traffic bottleneck of the application

[0025] Figure 2 is the road network data of a part of the selected area road section in the embodiment of the application

[0026] Figure 3 is the road operation level grade division of the Urban Road Traffic Operation Evaluation Standard referred to by the application

[0027] Figure 4 is the congestion cost distribution diagram of the congestion link in the embodiment of the application

[0028] Figure 5 is the frequency histogram of the congestion bottleneck in two typical time periods in the embodiment of the application in a week

[0029] Figure 6 is the distribution diagram of the key bottleneck road section from 6 to 10 o'clock in the embodiment of the application

[0030] Figure 7 is the distribution diagram of the key bottleneck road section from 14 to 17 o'clock in the embodiment of the application DETAILED DESCRIPTION

[0031] The preferred embodiments of the application will be described in detail below with reference to the drawings. It should be understood that the preferred embodiments are only for illustrating the application, and are not intended to limit the protection scope of the application.

[0032] As Figure 1As shown, a congestion propagation-based urban road section traffic key bottleneck identification method of the present application comprises the following steps:

[0033] Step S1: data collection: collect vehicle GPS trajectory data, including taxi trajectory data and online car GPS trajectory data, for representing road conditions, and collect road network data; considering that taxis and online cars are operating vehicles, they have long running time, running time regularity, and extensive location distribution, and can better reflect urban road conditions, so the present method collects taxi GPS trajectory and online car GPS trajectory as research data;

[0034] Step S2: data fusion: fuse the collected online car GPS trajectory and taxi GPS trajectory data to unify the data dimension;

[0035] Step S3: map matching: matching the GPS trajectory to the road section;

[0036] Step S4: road congestion state determination: calculate the traffic speed of the road section, analyze the speed-time distribution law of the road section, and determine the saturation critical speed of the road section, i.e. the speed determination threshold when the road section is congested;

[0037] Step S5: extract congestion events: calculate the average speed of each road section at each time slice, and the set of continuous N time slices from the formation of congestion to the dissipation of congestion of a road section is called a congestion event;

[0038] Step S6: extract congestion activation events: if two adjacent road sections enter the congestion state successively, and the congestion state of the downstream road section lasts after the congestion of the upstream road section is formed, it is determined that the congestion is propagated from the downstream road section to the upstream road section;

[0039] Step S7: form congestion propagation links: combine the congestion events and the congestion activation events, and the complete propagation link is formed by continuously connecting the congestion events through the congestion activation events;

[0040] Step S8: quantitatively evaluate the congestion cost of the congestion propagation link:

[0041] Step S9: identify road section traffic key bottlenecks: analyze the obtained congestion link-congestion cost distribution diagram, extract congestion links greater than the congestion cost threshold, and analyze the time distribution of these congestion links to extract key bottleneck road sections.

[0042] Embodiment

[0043] In this embodiment, the data collected in step S1 comes from taxi GPS trajectory data and online car-hailing GPS trajectory data in Chongqing, and the data records include vehicle ID, location, time, instantaneous speed, azimuth, positioning identifier, etc., wherein the time includes year, month, date, hour, minute, and second, and the location includes longitude and latitude. In addition, part of the road network data of a certain urban area in Chongqing is collected, which includes the length of the road, the number of lanes, the level of the road, and the upstream and downstream topological relationship. Figure 2

[0044] In step S2, the collected online car-hailing GPS trajectory and taxi GPS trajectory data are fused, and the data dimensions are unified, only keeping the vehicle ID, time, longitude and latitude, instantaneous speed, and azimuth fields. In addition, since the collected road network data is generally obtained by dividing the road network nodes, the length of the road section is generally between 100 meters and 200 meters, but the expressway may be several kilometers, which is unevenly distributed, and needs to be further divided to make the length of each road section about 100 meters to avoid the inability to find congestion existing in the middle part of the road section due to the excessive length.

[0045] In step S3, for the fused trajectory data, i.e., the vehicle ID, time, longitude and latitude, instantaneous speed, and azimuth field data, and the re-divided road network data, an improved hidden Markov model map matching method is used to match the trajectory data to the divided road sections.

[0046] Firstly, abnormal trajectory data needs to be removed, including the following cases: ① removing trajectory points with instantaneous speed exceeding 28 m / s; ② removing trajectory points with azimuth outside [0, 360]; ③ removing trajectory points with the same vehicle ID, instantaneous speed and azimuth of adjacent trajectory points being 0, but longitude and latitude and time changing. Then, the coordinates are converted to WGS-84 coordinate system.

[0047] Then, a 50-meter buffer zone is set for the road network data to obtain candidate state points within the buffer zone. These candidate state points and road network data are used as inputs of the improved hidden Markov model to calculate the observation probability and transition probability of all candidate state points. Then, the observation probability is used as the node and the transition probability is used as the edge to construct a directed weighted graph, and the candidate state point set with the maximum cumulative sum of observation probability and transition probability is obtained. The matching point set is obtained by backtracking the point set, which is the optimal matching trajectory.

[0048] In step S4, 5-minute time slices are used to filter out trajectory points matched to the same road section and belonging to the same time slice, and the harmonic mean speed is calculated as the running speed of the road section in that time slice. The calculation formula is as follows:

[0049] ​

[0050] In step S5, the road congestion state is determined, and the road running level grade division table in the "Urban Road Traffic Running Evaluation Standard" is referred to Figure 3 It is easy to find that the congestion determination threshold of different levels of roads is 40% of the free flow speed, and the congestion determination threshold is 25% of the free flow speed. Generally, the free flow speed is mainly related to the level of the road, the number of lanes, the length of the road, etc., but there is no standard corresponding relationship, so here we need to draw a speed-time distribution graph for each road section, and take 40% and 25% of the highest four instantaneous speed averages in a day as the congestion critical speed and the congestion critical speed of the road section, i.e. the speed determination threshold when the road section is congested.

[0051] According to the congestion critical speed of each road section obtained, the harmonic mean speed of all road sections at each time slice is determined.

[0052] In step S6, according to the congestion state determination result of step 5, the congestion event of each road section is extracted. Starting from the first time slice, when a time slice with a congestion determination result of 1 is encountered, the start time of the current congestion event is recorded, and the subsequent time slices are continuously traversed. If the congestion determination result of the subsequent time slice is also 1, it is the same congestion event, otherwise the last time slice with a congestion state of 1 that has been traversed is recorded as the end time of the current congestion event. Repeat the above steps until all congestion events of all road sections are extracted, forming a three-tuple record of (road section, congestion start time, congestion end time). The algorithm process is as follows:

[0053]

[0054] In step S7, the congestion propagation of the road section needs to meet the rules of spatial adjacency and time intersection. Assuming that road_1 and road_2 are adjacent road sections, road_1 is the downstream road section of road_2, then the vehicles will flow from road_2 to road_1, and the congestion will propagate from road_1 to road_2. Assuming that the congestion state of the downstream road section road_1 continues after the congestion of the upstream road section road_2 is formed, it is judged that the congestion is propagated from the downstream road section road_1 to the upstream road section road_2.

[0055] In step 6, congestion events of all road segments are extracted. In step 7, congestion propagation events between road segments are extracted. Next, step S8 will be entered, i.e. combining congestion events with congestion activation events to connect single congestion events to form complete propagation links. Taking road_1 and road_2 in step 7 as an example, if there is a congestion propagation event from road_1 to road_2, the end time of a congestion event of road_1 is later than the start time of a congestion event of road_2, and the start time of the congestion event of road_1 is earlier than the start time of the congestion event of road_2, then the two congestion events are connected, and the above steps are repeatedly performed until no new congestion event can be connected.

[0056] Finally, the congestion cost of the congestion propagation link is quantitatively evaluated by step S9: in order to quantitatively reflect the influence degree of the congestion propagation link on the road network, the length of the congestion road segment, the number of lanes of the congestion road segment and the congestion duration are considered as factors, and the sum of the costs of all congestion events included in the link is calculated, and the total cost of the congestion link is calculated as follows:

[0057]

[0058] wherein Cost k is the congestion cost of the kth congestion propagation link, D k is the set of congestion events corresponding to the kth congestion propagation link, len r is the length of the road segment r corresponding to the congestion event p, CDS r is the number of lanes of the road segment r corresponding to the congestion event p, T p is the congestion duration of the road segment r corresponding to the congestion event p.

[0059] For each propagation link, the total cost of the congestion is calculated, and a congestion cost-propagation link distribution diagram is drawn, referring to Figure 4 , outliers are removed, and 100 is selected as the key bottleneck congestion cost threshold according to experience.

[0060] On the basis of the congestion link results obtained in step 9, the time distribution of the congestion link in a week is analyzed. First, a day is divided into several time periods, i.e. [00:00-06:00], [06:00-10:00], [10:00-12:00], [12:00-14:00], [14:00-17:00], [17:00-19:00], [19:00-21:00], [21:00-24:00]. The frequency histogram of the frequency of the congestion links with a congestion cost exceeding 100 in each time period in a week is drawn, Figure 5is the histogram of congestion link frequency of some road segments in a certain district of Chongqing during morning and evening rush hours in weekdays. If a link is a bottleneck at least 3 days in a week, it is called a frequently-occurring critical congestion link at that time period, referring to Figure 6 and Figure 7 .

[0061] It should be appreciated that embodiments of the present application can be realized by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable storage medium. The methods can be implemented in a computer program using standard programming techniques - including non-transitory computer-readable storage medium configured with a computer program where the storage medium so configured makes the computer operate in a specific and predefined manner according to the methods described in the detailed embodiments and drawings. Each program can be implemented in a high level procedural or object oriented programming language to be executed in a computer system. However, if required, the programs can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. Moreover, such programs can be run on a dedicated integrated circuit, or on a programmable computer.

[0062] Further, the operations of the processes described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes described herein (or variations and / or combinations thereof) can be performed under the control of one or more computer systems configured with executable instructions (e.g., computer programs, one or more computer programs, or one or more applications) to perform the processes, by hardware, or combinations thereof. The computer programs include processor-executable instructions that are stored in memory.

[0063] Further, the methods can be implemented in any suitable type of computing platform operably connected to, including but not limited to, a personal computer, a mini-computer, a mainframe, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and the like. Aspects of the present application can be implemented in machine readable code stored on a non-transitory storage medium or device, whether removable or integrated to the computing platform, such as a hard disk, optical read and / or write storage media, RAM, ROM, and the like, such that it is readable by a programmable computer and, when the storage medium or device is read by the computer, is used to configure and operate the computer to perform the processes described herein. Further, the machine readable code, or portions thereof, can be transmitted over a wired or wireless network. The present application described herein includes these and other different types of non-transitory computer readable storage media when such media include instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor. The present application also includes the computer itself when programmed according to the methods and techniques described in the present application.

[0064] A computer program can be applied to input data to perform the functionality described herein, thereby transforming the input data to generate output data that is stored to non-volatile memory. The output information can also be applied to one or more output devices, such as a display. In a preferred embodiment of the application, the transformed data represents a physical and tangible object, including a particular visual depiction of a physical and tangible object produced on a display.

[0065] Finally, it should be noted that the above-mentioned embodiments illustrate rather than limit the application, since the scope of the application is determined by the appended claims.

Claims

1. A congestion propagation based urban road segment traffic key bottleneck identification method, characterized in that: The method comprises the following steps: Step S1: data collection: collecting vehicle GPS trajectory data, including taxi trajectory data and online car GPS trajectory data, for representing road conditions, and collecting road network data; Step S2: data fusion: fusing the collected online car GPS trajectory and taxi GPS trajectory data, and unifying data dimensions; Step S3: map matching: matching the GPS trajectory to a road section; Step S4: road section congestion state determination: calculating the speed of the road section, analyzing the speed-time distribution law of the road section, and determining the saturation critical speed of the road section, i.e. the speed determination threshold when the road section is congested; Step S5: extracting congestion events: calculating the average speed of each road section in each time slice, according to the congestion state determination result of step S4, extracting congestion events for each road section, starting from the first time slice, when encountering a time slice with a congestion determination result of existing congestion, recording the start time of the current congestion event, and continuing to traverse the subsequent time slices, if the congestion determination result of the subsequent time slice of the road section also exists, it is the same congestion event, otherwise, the last time slice of the congestion state that has been traversed is recorded as the end time of the current congestion event; a set of N continuous time slices from the formation of congestion to the dissipation of congestion of a road section is referred to as a congestion event; Step S6: extracting congestion activation events: if two adjacent road sections enter the congestion state successively, and the congestion state of the downstream road section lasts after the congestion of the upstream road section is formed, it is determined that the congestion is propagated from the downstream road section to the upstream road section; Step S7: forming congestion propagation links: combining the congestion events and the congestion activation events, and the complete propagation link is formed by continuously connecting the congestion events through the congestion activation events; Step S8: quantitatively evaluating the congestion cost of the congestion propagation link: Step S9: identifying the key bottleneck of the road section: analyzing the obtained congestion link-congestion cost distribution diagram, extracting the congestion links greater than the congestion cost threshold, and analyzing the time distribution of the congestion links to extract the key bottleneck road section; In step S3, the original GPS trajectory points are processed by using a hidden Markov model map matching method: a 50-meter buffer zone is set for the road network data, and candidate state points of the GPS trajectory points within the buffer zone are obtained; the candidate state points and the road network data are taken as inputs of the hidden Markov model, and the observation probability and the transition probability of all candidate state points are calculated; the observation probability is taken as a node, and the transition probability is taken as an edge to construct a directed weighted graph; the candidate state point set with the maximum cumulative sum of the observation probability and the transition probability is obtained, the matching point set is obtained by backtracking the candidate state point set, and the matching trajectory is taken as the optimal matching trajectory, so as to match the GPS trajectory to the road section.

2. The method of claim 1, wherein the method is characterized by: In step S8, in order to quantitatively reflect the loss degree of the congestion propagation link to the road network operation, the length of the congestion road section, the number of lanes of the congestion road section, and the congestion duration are taken as consideration factors, and the total cost calculation formula of the congestion link is as follows: where Cost k is the congestion cost of the kth congestion propagation link, D k is the set of congestion events corresponding to the kth congestion propagation link, len r is the length of the road segment r corresponding to the congestion event p, CDS r is the number of lanes of the road segment r corresponding to the congestion event p, T p is the congestion duration of the road segment r corresponding to the congestion event p.

3. A computer device comprising a memory, a processor, and a computer program stored on the memory and capable of running on the processor, characterized in that: The processor executes the computer program to realize the method of any one of claims 1-2. The processor executes the computer program to realize the method of any one of claims 1-2.

4. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program, which when executed by the processor, implements the method of any of claims 1-2.

Images