Traffic situation prediction device and traffic situation prediction method

Abstract

To predict a traffic situation with high accuracy even when a road is in an irregular scene.SOLUTION: The processor is configured to determine, for each of the sections, whether a possibility that a traffic situation indicated by the traffic situation prediction data occurs in terms of traffic engineering is equal to or greater than a predetermined threshold based on the past traffic situation data, and correct the traffic situation prediction data such that the possibility that the traffic situation indicated by the traffic situation prediction data occurs in terms of traffic engineering is equal to or greater than the predetermined threshold when the possibility is less than the predetermined threshold.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] An embodiment of the present invention relates to a traffic condition prediction device and a traffic condition prediction method. [Background technology]

[0002] For roads such as expressways, predicting future traffic conditions such as congestion has long been important for road users (vehicle drivers) and traffic controllers who perform traffic control duties. Currently, for example, traffic condition information on expressways is provided to road users and traffic controllers by processing and determining the current traffic conditions using a traffic control system. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2024-4031 [Patent Document 2] Japanese Patent Application Laid-Open No. 2007-179348 [Patent Document 3] Japanese Patent Application Laid-Open No. 2010-191614 [Patent Document 4] Patent Publication No. 2021-71995 Summary of the Invention [Problem to be solved by the invention]

[0004] Traffic conditions on expressways change every moment, but current systems are unable to provide information on congestion, travel times, etc. that takes future traffic conditions into account. There are several possible ways to predict future traffic conditions, but in the case of machine learning prediction models, for example, the logic is a black box, so the output (prediction) may be poorly reproducible from a traffic engineering perspective.

[0005] To give a specific example, on a route with no inflow or outflow, the traffic volume forecast results for point A may be output as exceeding the traffic capacity of point B, which is located immediately upstream. This tendency is expected to be particularly pronounced in unusual traffic conditions, such as lane restrictions, road closures, or speed restrictions, which are not conditions that would lead to natural congestion. Such irregular situations can be caused by frequent accidents, construction work, broken-down vehicles, as well as heavy snowfall and natural disasters.

[0006] Therefore, an embodiment of the present invention provides a traffic condition prediction device and a traffic condition prediction method that can predict traffic conditions with high accuracy even when the road is in an irregular state. [Means for solving the problem]

[0007] A traffic condition prediction device according to an embodiment includes a prediction unit that predicts future traffic conditions for each section of a road on which a vehicle travels based on past traffic condition data and a predetermined prediction model, and outputs traffic condition prediction data; and a filter processing unit that determines, for each section, based on the past traffic condition data, whether the traffic condition indicated by the traffic condition prediction data is likely to occur from a traffic engineering perspective, equal to or greater than a predetermined threshold, and if the likelihood is less than the predetermined threshold, corrects the traffic condition prediction data so that the traffic condition indicated by the traffic condition prediction data is likely to occur from a traffic engineering perspective, equal to or greater than the predetermined threshold. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram schematically illustrating a road according to an embodiment. [Figure 2] FIG. 2 is a functional block diagram schematically illustrating the overall configuration of the traffic control system according to the embodiment. [Figure 3] FIG. 3 is a flowchart showing the processing performed when learning a prediction model. [Figure 4] FIG. 4 is a flowchart showing the process for predicting traffic conditions. [Figure 5]FIG. 5 is a flowchart showing the process of step S12 in FIG. [Figure 6] FIG. 6 is a flowchart showing the process of step S13 in FIG. [Figure 7] FIG. 7 is a diagram showing an example of a QV diagram. [Figure 8] FIG. 8 is a graph showing an example of the relationship between speed and time. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the traffic condition prediction device and traffic condition prediction method of the present invention will be described with reference to the drawings. Note that the configurations of the embodiments described below, and the actions and results (effects) brought about by the configurations, are examples, and the present invention is not limited to the contents described below.

[0010] FIG. 1 is a diagram schematically illustrating a road R in an embodiment. The road R is, for example, a highway. The road R is divided into sections 1, 2, 3, ... as management units. A vehicle detector 1 and a CCTV camera 2 are installed in each section. In the following, the reference number for road R will be omitted and the road will be referred to as "road."

[0011] 2 is a functional block diagram showing a schematic overall configuration of a traffic control system S according to an embodiment. The traffic control system S includes a vehicle detector 1, a CCTV camera 2, a traffic control device 3 (traffic condition prediction device), an information output device 4, and a weather meter 9.

[0012] The weather meter 9 measures temperature, humidity, wind direction, wind speed, precipitation, snowfall, etc., and transmits the measurement information as weather data to the traffic control device 3. Note that the traffic control device 3 may receive weather data from a weather data management company or the like instead of the weather meter 9.

[0013] The traffic control device 3 is a computer device, and includes a processing unit 31, a storage unit 32, an input unit 33, a display unit 34, a communication unit 35, and a traffic control information providing system 36.

[0014] The processing unit 31 includes, for example, a central processing unit (CPU), a read-only memory (ROM), and a random access memory (RAM), and executes various processes.

[0015] The CPU comprehensively controls the operation of the traffic control device 3. The COM is a storage medium for storing various programs and data. The RAM is a storage medium for temporarily storing various programs and rewriting various data.

[0016] The MPU uses the RAM as a work area to execute programs stored in the ROM, storage unit 32, etc. Details of the processing unit 31 will be described later.

[0017] The storage unit 32 is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and stores various types of information. The storage unit 32 stores, for example, accident form data 321 and a prediction model 322.

[0018] The accident report data 321 is report data that records details of accidents that have occurred on roads in the past. The accident report data 321 includes various information such as the location of the accident, the cause of the accident, accident detection, dispatch information of emergency vehicles (ambulances, accident response vehicles, police cars, etc.), the time until the accident response is completed, lane closure status, information on injured people, the presence or absence of fallen objects, and information on whether the accident vehicle can continue driving.

[0019] The prediction model 322 is a learning model for predicting traffic conditions including the scale of congestion that is expected to occur on a road, time periods, etc. The prediction model 322 is set for each section. The prediction model 322 may also be set for each time period.

[0020] The prediction model 322 can be realized by, for example, a machine learning model. In addition to a normal time prediction model used in normal times, an irregular time prediction model used in irregular situations is set as the prediction model 322. The irregular time prediction models are prepared for, for example, accidents, construction work, broken-down vehicles, heavy snow, disasters, etc., and are trained using the respective training data.

[0021] The memory unit 32 also stores various information such as road information, such as road sections, number of lanes, interchanges, and locations of parking areas, sensor information obtained from the vehicle detector 1, video data obtained from the CCTV camera 2, weather data obtained from the weather meter 9, and various calculation processing results by the processing unit 31.

[0022] The input unit 33 is an input device that accepts user operations on the traffic control device 3, and is, for example, a keyboard, a mouse, or the like.

[0023] The display unit 34 is a display unit located near the input unit 33 and is provided for each traffic controller. The display unit 34 is realized by, for example, a liquid crystal display device (LCD (Liquid Crystal Display)), an organic EL (Electro-Luminescence) display device, or the like.

[0024] The communication unit 35 is a communication interface for communicating with external devices (such as the vehicle detector 1, the CCTV camera 2, the information output device 4, and the weather meter 9).

[0025] The traffic control information providing system 36 is a system that controls the traffic control of expressways and related roads, and displays information to be provided about target road sections on a large display device 36M.

[0026] The processing unit 31 includes, as functional components, an acquisition unit 311, a prediction unit 312, a filter processing unit 313, a creation processing unit 314, and a control unit 315.

[0027] The acquisition unit 311 acquires various types of information from external devices. For example, the acquisition unit 311 acquires sensor information (traffic condition data) from the vehicle detector 1. Then, for example, the acquisition unit 311 creates information such as traffic volume, average speed, and occupancy rate in one-minute or five-minute increments based on the sensor information. The acquisition unit 311 also acquires road video data from the CCTV camera 2. The acquisition unit 311 also acquires weather data from the weather meter 9.

[0028] The prediction unit 312 executes various prediction processes. For example, the prediction unit 312 predicts future traffic conditions for each section of a road on which a vehicle travels based on predetermined parameters (for example, traffic capacity, etc.) used for predicting traffic conditions, past traffic condition data, and a normal state prediction model (a predetermined prediction model), and outputs traffic condition prediction data.

[0029] Furthermore, the prediction unit 312 determines whether or not an irregular situation that will affect the traffic situation has occurred for each section. This determination may be made based on information from the vehicle detector 1, the CCTV camera 2, and the weather meter 9, for example, or on other information (for example, information input by a traffic controller via the input unit 33 regarding construction work, a broken-down vehicle, a disaster, or the like when such a situation occurs).

[0030] The prediction unit 312 may perform at least one of the following for an irregular scene section: using an irregular scene prediction model instead of a normal scene prediction model; and changing predetermined parameters according to the irregular scene.

[0031] When the prediction unit 312 uses the irregular situation prediction model instead of the normal situation prediction model for an interval that is an irregular scene, for example, the normal situation prediction model is a machine learning model, and the irregular situation prediction model is any one of a machine learning model, a simulation model, and a rule-based model. This will be described in detail.

[0032] In order to predict traffic conditions for a target section under various circumstances, the prediction model to be used is switched depending on the situation. The switching conditions here may be not only due to traffic conditions such as lane closures and road closures, but also heavy snow, tourist season, etc. Furthermore, the prediction model for irregular situations is not necessarily limited to a machine learning model. For example, a simulation model may be used when a road closure occurs due to heavy snow. Furthermore, for example, a rule-based model may be used when an accident occurs at a high-accident location and lane restrictions are imposed. Note that when a machine learning model is used, it is preferable to use new data (for example, data from the most recent few years) as learning data, since old data may not be suitable as learning data.

[0033] Furthermore, when the prediction unit 312 changes a predetermined parameter for an irregular section in accordance with the irregular situation, for example, the predetermined parameter is traffic capacity, and the prediction unit 312 changes the traffic capacity in accordance with the irregular situation. This will be described in detail.

[0034] In order to predict traffic conditions under various circumstances for a target section, not only the prediction model but also the input information is changed according to the situation. For example, if the traffic capacity for each section is input into the prediction model, this value is normally constant (fixed), but if a broken-down vehicle occurs and lane restrictions are implemented, traffic capacity will temporarily decrease. In preparation for such changes in conditions due to external factors, the input information is parameterized and changed according to the situation. In other words, rather than simply inputting data on observed conditions, like traffic data or weather data, a mechanism is provided to change the input information (parameters) according to the conditions.

[0035] The filter processing unit 313 determines, for each section, based on the past traffic condition data, whether the possibility that the traffic conditions indicated by the traffic condition prediction data will occur from a traffic engineering perspective is equal to or greater than a predetermined threshold, and if the possibility is less than the predetermined threshold, corrects the traffic condition prediction data so that the possibility that the traffic conditions indicated by the traffic condition prediction data will occur from a traffic engineering perspective is equal to or greater than the predetermined threshold.

[0036] Here, an example of the processing performed by the filter processing unit 313 will be described in detail with reference to FIGS. Figure 7 shows an example of a QV diagram. The driving conditions of a group of vehicles on a highway are determined to some extent by the range of possible actual values ​​(measured values) for each section, depending on conditions such as the road alignment and speed limit. When predicting traffic conditions, corrections are made for values ​​that fall outside this range of possible values. In other words, predicted values ​​with low reproducibility from a traffic engineering perspective are corrected to predicted values ​​with high reproducibility from a traffic engineering perspective.

[0037] There are various possible cases for correction by the filter processing unit 313, but for example, correction is performed when a value that exceeds the upper or lower limit of the actual value for each section is found. The actual value here is, for example, a combination of predicted values ​​of traffic volume (vertical axis) and speed (horizontal axis), as shown in Fig. 7. The actual value is the accumulation of past traffic data, but traffic conditions change depending on social conditions, road alignments, etc., and old data cannot be said to reflect the most recent traffic conditions, so it is preferable to use newer data (for example, data from the most recent few years).

[0038] In FIG. 7, the multiple round points M represent measured values. The multiple triangular points P represent predicted values. Line L1 is a line that approximates the outer boundary of point M with a quadratic curve. In this case, areas A1 and A2 are areas that include the predicted value P to be corrected. For the predicted value P to be corrected, if the most recent predicted value is inside line L1, the filter processing unit 313 corrects the predicted value P to be corrected by, for example, connecting the most recent predicted value and the predicted value P to be corrected with a straight line and moving the predicted value P to the intersection of the straight line and line L1. In addition, the low reproducibility from the perspective of traffic engineering may be determined using, for example, the so-called Drake equation.

[0039] Figure 8 is a graph showing an example of the relationship between speed and time. Here, as an example, consider a case where the average speed value for a target section is predicted and output every minute. In this case, it is unlikely that the predicted value will change instantly from 30 km / h to 80 km / h (low reproducibility from a traffic engineering perspective), so it is preferable to correct the predicted value.

[0040] In FIG. 8, multiple circles represent predicted values. Line L2 represents the upper limit of the predicted values. Line L3 represents the lower limit of the predicted values. In this case, the circle surrounded by areas A3 and A4 extends beyond the area between lines L2 and L3 and represents the predicted value to be corrected. For example, if the most recent predicted value is within the area between lines L2 and L3, the filter processing unit 313 corrects the predicted value to be corrected by connecting the most recent predicted value and the predicted value to be corrected with a straight line and moving the predicted value to be corrected to the intersection of the straight line and line L2 or line L3.

[0041] Returning to Figure 2, the creation processing unit 314 performs processing such as calculating the required travel time for a specified section based on the traffic condition prediction data predicted by the prediction unit 312 and data corrected by the filter processing unit 313 regarding the traffic condition prediction data, and creates information to be provided.

[0042] The control unit 315 executes processes other than those executed by the units 311 to 314. The control unit 315 transmits the information to be provided created by the creation processing unit 314 to the information output device 4, for example.

[0043] The information output device 4 includes, for example, a road information board 42, a highway information terminal 44, and a communication device 45. The control unit 315 of the traffic control device 3 transmits the provided information and the like to the traffic control information providing system 36, the information output device 4, the terminal device 51, and the car navigation system 52.

[0044] The road information board 42 is installed, for example, on the shoulder of the road including the target section or above the road in a state spanning the road. The road information board 42 displays the provided information mainly in text.

[0045] The highway information terminal 44 is an information providing device installed in service areas, parking areas, etc., and displays a wide range of information, for example, centered on service areas and parking areas, in accordance with the operation of road users, and outputs related audio information.

[0046] Furthermore, the information output device 4 transmits the provided information, etc. to the terminal device 51 via the mobile communication network N. The terminal device 51 is a mobile phone, smartphone, tablet, etc. that can be used by road users, and displays the provided information, etc. in accordance with the operation of the road users and outputs related audio information by accessing a predetermined web page, for example.

[0047] Furthermore, the information output device 4 transmits the provided information, etc. to a car navigation system 52 via a communication device 45 such as a roadside antenna. The car navigation system 52 is installed in a vehicle in which a road user rides, and acquires provided information, etc. about surrounding roads based on the vehicle's position identified by a GPS (Global Positioning System), and provides the information to the road user.

[0048] Next, the processing by the traffic control device 3 will be described with reference to FIGS. FIG. 3 is a flowchart showing the processing performed when learning a prediction model.

[0049] In step S1, the control unit 315 records input information and output information for each of a plurality of prediction models to be used depending on traffic conditions, lane restrictions, and other circumstances, and performs learning at regular intervals.

[0050] Next, in step S2, the control unit 315 updates the prediction model after a certain period of time has elapsed.

[0051] 4 is a flowchart showing the process of predicting traffic conditions. In step S11, the acquisition unit 311 acquires various information from external devices (vehicle detector 1, CCTV camera 2, weather meter 9, etc.).

[0052] Next, in step S12, the prediction unit 312 executes a traffic situation prediction process. FIG. 5 is a flowchart showing the process of step S12 in FIG.

[0053] In step S121, the prediction unit 312 selects a prediction model (a normal time prediction model, an irregular time prediction model) according to the situation.

[0054] Next, in step S122, the prediction unit 312 changes the input information (for example, traffic capacity) depending on the situation.

[0055] Next, in step S123, the prediction unit 312 predicts future traffic conditions based on the input information, past traffic condition data, and the selected prediction model, and outputs traffic condition prediction data.

[0056] Returning to FIG. 4, after step S12, in step S13, the filtering unit 313 executes filtering.

[0057] Here, Fig. 6 is a flowchart showing the process of step S13 in Fig. 4. In step S131, the filter processing unit 313 acquires traffic condition prediction data.

[0058] Next, in step S132, the filter processing unit 313 determines whether or not the traffic condition prediction data needs to be corrected, and if Yes, proceeds to step S133, and if No, ends the process. Specifically, in step S132, the filter processing unit 313 determines, for each section, based on the past traffic condition data, whether or not the traffic condition indicated by the traffic condition prediction data is more than or equal to a predetermined threshold value in terms of traffic engineering.

[0059] In step S133, the filter processing unit 313 corrects the traffic condition prediction data so that the traffic condition indicated by the traffic condition prediction data has a traffic engineering probability of occurring equal to or greater than a predetermined threshold.

[0060] Returning to FIG. 4, after step S13, in step S14, the creation processing unit 314 performs processing such as calculating the travel time required for a specified section based on the traffic condition prediction data predicted in step S12 and the data corrected for that traffic condition prediction data in step S13, and creates information to be provided.

[0061] Next, in step S15, the control unit 315 transmits the provided information to the information output device 4.

[0062] As described above, the traffic control device 3 of this embodiment can predict traffic conditions with high accuracy even when the road is in an irregular state. For example, traffic condition prediction data can be appropriately corrected from a traffic engineering perspective, and highly accurate traffic condition prediction data can be output. This makes it possible to predict future traffic conditions with high accuracy for various traffic conditions on expressways.

[0063] In addition, by switching the prediction model used for each section between normal and irregular traffic conditions, traffic conditions can be predicted with even greater accuracy.

[0064] Furthermore, for sections with irregular conditions, traffic conditions can be predicted with even higher accuracy by changing input information (parameters) such as traffic capacity according to the irregular conditions.

[0065] It is possible to simultaneously implement the correction of traffic condition prediction data from a traffic engineering perspective, the switching of prediction models between normal and irregular conditions, and the modification of input information such as traffic capacity in response to irregular situations, but this is not necessarily required. The effect can be achieved by implementing at least one of them.

[0066] The program executed by the CPU of the traffic control device 3 of this embodiment may be configured to be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a DVD (Digital Versatile Disk).

[0067] Furthermore, the program may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the program executed in this embodiment may be provided or distributed via a network such as the Internet.

[0068] Although several embodiments of the present invention have been described above, the above-described embodiments and modifications are merely examples and are not intended to limit the scope of the invention. The above-described embodiments can be implemented in various forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The above-described embodiments and modifications are included within the scope and spirit of the invention, and are also included in the inventions and their equivalents as set forth in the claims.

[0069] For example, the target roads are not limited to expressways, but may also be toll roads other than expressways or free general roads.

[0070] Furthermore, in FIG. 7, the line L1 is a line that approximates the outer boundary of the point M with a quadratic curve, but this is not limited to this. For example, the line may be a line that is slightly outside or inside the boundary, and the type of approximating line may be a curve other than a quadratic curve.

[0071] Furthermore, the input information that is changed in response to irregular situations is not limited to traffic capacity, and may be other input information such as an upper speed limit (for example, during heavy snowfall, etc.). [Explanation of symbols]

[0072] 1...vehicle detector, 2...CCTV camera, 3...traffic control device (traffic situation prediction device), 4...information output device, 31...processing unit, 32...storage unit, 33...input unit, 34...display unit, 35...communication unit, 36...traffic control information provision system, 42...road information board, 44...highway information terminal, 51...terminal device, 52...car navigation system, 311...acquisition unit, 312..., 313..., 314..., 315..., 321...accident report data, 322...prediction model, C...vehicle, R...road, S...traffic control system

Claims

1. a prediction unit that predicts future traffic conditions for each section of a road on which the vehicle travels based on past traffic condition data and a predetermined prediction model, and outputs traffic condition prediction data; a filter processing unit that determines, for each section, based on the past traffic condition data, whether or not a traffic condition indicated by the traffic condition prediction data is likely to occur from a traffic engineering perspective is equal to or greater than a predetermined threshold, and if the likelihood is less than the predetermined threshold, corrects the traffic condition prediction data so that the traffic condition indicated by the traffic condition prediction data is likely to occur from a traffic engineering perspective is equal to or greater than the predetermined threshold; A traffic situation prediction device comprising:

2. a prediction unit that predicts future traffic conditions for each section of a road on which the vehicle travels based on predetermined parameters used for predicting traffic conditions, past traffic condition data, and a predetermined prediction model, and outputs traffic condition prediction data; The prediction unit For each section, determine whether or not there is an irregular situation that will affect the traffic situation; A traffic situation prediction device that performs at least one of the following for the section in which the irregular situation occurs: using a prediction model for irregular situations instead of the prediction model; and changing the specified parameters in accordance with the irregular situation.

3. The prediction unit When the irregular situation prediction model is used instead of the prediction model for the section in which the irregular situation occurs, the predictive model is a machine learning model, The traffic condition prediction device according to claim 2 , wherein the irregular time prediction model is one of a machine learning model, a simulation model, and a rule-based model.

4. The prediction unit When the predetermined parameter is changed in accordance with the irregular scene for the section that is the irregular scene, the predetermined parameter is traffic capacity, The traffic condition prediction device according to claim 2 , wherein the traffic capacity is changed depending on the irregular situation.

5. a prediction step in which the prediction unit predicts future traffic conditions for each section of a road on which the vehicle travels based on past traffic condition data and a predetermined prediction model, and outputs traffic condition prediction data; a filtering processing step in which a filtering processor determines, for each section, based on the past traffic condition data, whether or not a traffic condition indicated by the traffic condition prediction data will occur from a traffic engineering perspective is equal to or greater than a predetermined threshold, and if the likelihood is less than the predetermined threshold, corrects the traffic condition prediction data so that the traffic condition indicated by the traffic condition prediction data will occur from a traffic engineering perspective is equal to or greater than the predetermined threshold; A traffic situation prediction method comprising:

6. a prediction step in which the prediction unit predicts future traffic conditions for each section of a road on which the vehicle travels based on predetermined parameters used for predicting traffic conditions, past traffic condition data, and a predetermined prediction model, and outputs traffic condition prediction data; In the prediction step, the prediction unit For each section, determine whether or not there is an irregular situation that will affect the traffic situation; A traffic situation prediction method that performs at least one of the following: for the section that is in the irregular situation, using a prediction model for irregular situations instead of the prediction model; and changing the specified parameters in accordance with the irregular situation.

Images