JP7372058B2 - Traffic flow prediction support device, traffic flow prediction support method, and traffic flow prediction support program - Google Patents

Description

The present invention relates to technology that supports traffic flow prediction.

BACKGROUND ART Conventionally, a technique for predicting traffic flow by executing a traffic flow simulation based on past OD data is known. In recent years, the accuracy of predicting traffic flow has been improved by correcting the traffic volume at observation points based on real-time information such as vehicle detectors placed on the road.

For example, a technique described in Patent Document 1 is known as a technique for predicting traffic flow using observation information from a vehicle sensor. In the technology of Patent Document 1, the traffic volume and route selection parameters are adjusted to minimize the difference between the traffic volume based on observation information and the traffic volume determined by traffic flow simulation, and the traffic volume and route selection parameters are The adjustment and traffic flow simulation are repeatedly executed until the result of executing the traffic flow simulation using the input as input satisfies a predetermined termination condition.

JP2013-235326A

The technique disclosed in Patent Document 1 uses observation information from a vehicle sensor and takes into consideration the traffic volume at the observation point. Therefore, the actual movements of individual vehicles are not taken into consideration, and movement information of individual vehicles cannot be reflected, resulting in a problem in terms of accuracy. Further, in the technique disclosed in Patent Document 1, it is necessary to repeatedly execute the adjustment and traffic flow simulation until the termination condition is satisfied, which causes problems in that the processing load is high and the responsiveness is poor.

The present invention has been made in view of the above circumstances, and its purpose is to provide a technique that allows traffic flow to be predicted with high accuracy.

In order to achieve the above object, a traffic flow prediction support device according to one aspect includes an acquisition unit that acquires movement information of a plurality of vehicle models based on past OD data, and a communication unit that is estimated to exist in a predetermined actual vehicle. A vehicle selection unit that selects a compatible vehicle model that is a vehicle model that is compatible with an actual vehicle based on movement information of a terminal, and a vehicle selection unit that selects a compatible vehicle model that is a vehicle model that is compatible with an actual vehicle, and a vehicle selection unit that selects a compatible vehicle model that is a vehicle model that is compatible with an actual vehicle. An OD data generation unit that generates OD data based on the information.

According to the present invention, traffic flow can be predicted with high accuracy.

FIG. 1 is an overall configuration diagram of a traffic flow prediction system according to an embodiment. FIG. 2 is a configuration diagram of an example of OD data according to an embodiment. FIG. 3 is a hardware configuration diagram of a traffic flow prediction support device according to an embodiment. FIG. 4 is a flowchart of OD data generation processing according to one embodiment. FIG. 5 is a flowchart of real-time correction processing according to one embodiment.

Embodiments will be described with reference to the drawings. The embodiments described below do not limit the claimed invention, and all of the elements and combinations thereof described in the embodiments are essential to the solution of the invention. is not limited.

FIG. 1 is an overall configuration diagram of a traffic flow prediction system according to an embodiment.

The traffic flow prediction system 1 includes a past OD data storage section 2, a mobile data storage section 3, a traffic flow prediction support device 10, a corrected OD data storage section 4, a traffic flow simulation device 5, a display device 6, It includes a bus operation information storage section 7 and a corrected operation information storage section 8.

The past OD data storage section 2, the mobile data storage section 3, the corrected OD data storage section 4, the bus operation information storage section 7, and the corrected operation information storage section 8 are included in the traffic flow prediction support device 10. However, in the example of FIG. 1, the explanation will be made assuming that it is stored in another computer, etc. (not shown).

The past OD data storage unit 2 stores OD data collected in the past. The past OD data may be, for example, OD data collected in a national road traffic census.

The mobile data storage unit 3 stores mobile data (an example of movement information) including position information and movement speed transmitted from various communication terminals. Mobile data can be obtained, for example, from various communication carriers or collected from each communication terminal. In this embodiment, the mobile data includes an ID that can identify data from the same communication terminal as being the same, and location information of the communication terminal. Here, the ID may be an ID that can directly identify the communication terminal itself, or an ID that cannot directly identify the communication terminal itself. The location information of the communication terminal may be, for example, coordinates (latitude and longitude) determined by GPS (Global Positioning System). In this embodiment, the communication terminal includes a mobile phone or smartphone carried by an individual, a communication terminal mounted on a vehicle, and the like.

The corrected OD data storage unit 4 stores OD data generated by the traffic flow prediction support device 10. The data format of the OD data may be the same as that of the OD data in the past OD data storage section 2, for example.

The bus operation information storage unit 7 stores bus operation information including stopping points for buses among vehicles, travel time to each stopping point, and the like. The corrected operation information storage unit 8 stores information (corrected operation information) regarding the difference between the bus operation information and the actual bus operation information.

The traffic flow simulation device 5 predicts traffic flow by inputting OD data, corrected operation information, etc. and executing a traffic flow simulation. The display device 6 displays traffic flow information that is a simulation result by the traffic flow simulation device 5.

The traffic flow prediction support device 10 includes a route/traffic volume allocation calculator 11 , a transportation mode estimator 12 , a vehicle adaptor 13 , and a monitor 14 . Here, the acquisition unit, vehicle selection unit, OD data generation unit, and operation determination unit correspond to the vehicle adaptor 13, the deviation determination unit corresponds to the monitor 14, and the vehicle estimation unit corresponds to the transportation mode estimator. Corresponds to 12.

The route/traffic distribution calculator 11 inputs the past OD data stored in the past OD data storage unit 2 and calculates the position of the model (vehicle model) corresponding to each vehicle in the OD data, the travel route to the destination, Information such as travel time (statistical data estimation information: travel information) is calculated. Note that in the calculated statistical data estimation information, each vehicle model is managed by an identifiable ID (vehicle model ID). For buses included in the vehicles in the OD data, the route/traffic distribution calculator 11 calculates the position, travel route, and movement of a model (bus model) corresponding to this bus according to the bus operation information in the bus operation information storage section 7. Calculate information such as time (statistical data estimation information).

The transportation means estimating unit 12 receives the mobile data stored in the mobile data storage unit 3 as input, identifies a communication terminal that is estimated to exist in the vehicle, and uses the mobile data of the communication terminal that is estimated to exist in the vehicle as input. It is transmitted to the adaptor 13. For example, it is possible to estimate whether a communication terminal is owned by a pedestrian, mounted on a vehicle, or owned by a person riding in a vehicle, depending on the moving speed based on mobile data. I can do it. Furthermore, if multiple communication terminals are located close to each other and are moving relatively fast, and at the same or almost the same speed, the multiple communication terminals may be in the possession of a person riding the bus. It can be estimated that In this case, mobile data from any one communication terminal owned by a person estimated to be on the bus may be passed to the vehicle adaptor 13 as mobile data about the bus.

The vehicle adaptor 13 acquires statistical data estimation information of a plurality of vehicle models based on past OD data from the route/traffic volume distribution calculator 11, and calculates the actual vehicle (actual vehicle) in which the communication terminal that transmitted the mobile data exists. A vehicle model (compatible vehicle model) that satisfies predetermined conditions is selected for the vehicle (vehicle), and is managed in association with the vehicle model. For example, the predetermined conditions include being within a predetermined distance from the actual vehicle position at an assumed time (current time in real-time processing), and having a vehicle speed difference within a predetermined speed. In addition, to manage the correspondence between the compatible vehicle model and the actual vehicle, the vehicle model ID of the compatible vehicle model and the ID of the mobile data may be managed in correspondence.

The vehicle adaptor 13 generates OD data based on the movement information of mobile data of a communication terminal estimated to exist in the actual vehicle to which the adaptive vehicle model is adapted, instead of the movement information of the adaptive vehicle model. . In addition, the vehicle adaptor 13 generates a vehicle model corresponding to the movement of the actual vehicle when there is no compatible vehicle model that satisfies predetermined conditions, and stores the OD data including the movement information of the generated vehicle model. generate.

In addition, when it is determined that the movement information of the compatible vehicle model and the movement information of the actual vehicle to which the compatible vehicle model is matched differ, the vehicle adaptor 13 generates a new vehicle that matches the actual vehicle. Select the model as the new compatible vehicle model.

In addition, when the type of the actual vehicle is estimated to be a bus, the vehicle adaptor 13 selects a bus model that is compatible with the bus, and uses movement information based on the operation information of the bus model and a bus model that is compatible with the bus model. Based on the determined bus movement information, information on differences from the actual bus operation information (for example, delay time, deviation distance, etc.) is identified and output to the corrected operation information storage unit 8. do.

The monitor 14 determines whether there is a discrepancy between the movement information of the compatible vehicle model and the movement information of the actual vehicle to which the compatible vehicle model is matched. Here, a discrepancy in the movement information between the compatible vehicle model and the actual vehicle means that the location of the compatible vehicle model and the actual vehicle is more than a predetermined distance apart, or the route that the compatible vehicle model and the actual vehicle pass. There are cases where the values are different.

Next, OD data will be explained.

FIG. 2 is a configuration diagram of an example of OD data according to an embodiment. The OD data shown in FIG. 2 shows the OD data stored in the corrected OD data storage section 4, but the OD data stored in the past OD data storage section 2 also has a similar configuration.

The corrected OD data storage unit 4 stores OD data 41, 42, 43, etc. for each of a plurality of time points. The OD data 41, 42, 43, etc. are a table summarizing the number of vehicles for each starting point and ending point at the corresponding time. According to the OD data 41, when the difference from the reference time (for example, the current time) is -1 hour and 20 minutes, there are 50 vehicles starting at observation point 1 and ending at observation point 1, and 50 vehicles starting at observation point 1 and ending at observation point 2. It can be seen that there are eight vehicles that have the starting point as the starting point and the ending point as the ending point 2.

Next, a hardware configuration diagram of the traffic flow prediction support device 10 will be explained.

FIG. 3 is a hardware configuration diagram of a traffic flow prediction support device according to an embodiment.

The traffic flow prediction support device 10 is configured by a computer such as a PC (Personal Computer), and includes a communication interface (communication I/F) 101, a processor 102, a storage device 103, and a memory 104. . Communication I/F 101, processor 102, storage device 103, and memory 104 are communicably connected via bus 105.

The communication I/F 101 is, for example, an interface such as a wired LAN card or a wireless LAN card, and communicates with other devices via a network (not shown).

Processor 102 executes various processes according to programs stored in memory 104 and/or storage device 103.

The memory 104 is, for example, a RAM, and stores a program (traffic flow prediction support program 106) executed by the processor 102 and necessary information.

The storage device 103 is, for example, an HDD (hard disk drive) or an SSD (solid state disk), and stores programs executed by the processor 102 and data used by the processor 102.

In this embodiment, the processor 102 executes the traffic flow prediction support program 106 to configure the route/traffic volume allocation calculator 11, the transportation mode estimator 12, the vehicle adaptor 13, and the monitor 14. Ru.

Next, processing operations in the traffic flow prediction support device 10 will be explained.

FIG. 4 is a flowchart of OD data generation processing according to one embodiment.

The OD data generation process is executed, for example, at predetermined time intervals. First, the vehicle adaptor 13 acquires statistical data estimation information for all vehicle models from the route/traffic volume allocation calculator 11 (step S10).

Next, the transportation estimator 12 acquires mobile data of the communication terminal from the mobile data storage unit 3 (step S11).

Next, the transportation means estimator 12 estimates the communication terminals present in the vehicle based on the acquired mobile data of the communication terminals, based on the positions, speeds, density, etc. of the communication terminals, and uses the mobile data of the communication terminals as Information (real-time estimation information) such as the position and speed of the vehicle (actual vehicle) in which this terminal is estimated to exist is estimated (step S12). Note that the mobile data targeted for this processing is mobile data obtained from communication terminals within the range targeted by the OD data processed by the route/traffic volume allocation calculator 11.

Next, the vehicle adaptor 13 and the monitor 14 execute real-time correction processing (see FIG. 5) for correcting vehicle model information based on the real-time estimation information (step S13).

Next, the vehicle adaptor 13 outputs the corrected OD data generated by the real-time correction process to the corrected OD data storage section 4. Note that, using the corrected OD data output to the corrected OD data storage unit 4, the traffic flow simulation device 5 predicts traffic flow by executing a traffic flow simulation, and outputs the prediction result to the display device 6. . Thereby, the traffic flow simulation device 5 can predict traffic flow in accordance with the actual situation with high accuracy.

FIG. 5 is a flowchart of real-time correction processing according to one embodiment.

First, the vehicle adaptor 13 acquires unprocessed real-time estimation information corresponding to one actual vehicle from the transportation means estimator 12 (step S21). In the following description of the process, the actual vehicle corresponding to the acquired real-time estimation information will be referred to as the target actual vehicle.

Next, the vehicle adaptor 13 determines whether a vehicle model compatible with the target actual vehicle has not been selected or whether there is a need to reselect a vehicle model compatible with the target actual vehicle (specifically, whether the vehicle model compatible with the target actual vehicle needs to be reselected). (Step S22).

As a result, if a vehicle model that matches the target actual vehicle has not been selected, or if it is necessary to reselect a vehicle model that matches the target actual vehicle (step S22: Yes), the vehicle adaptor 13 performs statistical analysis. From the base estimation information, a vehicle model (compatible vehicle model) that meets predetermined conditions is selected, and the target actual vehicle and the compatible vehicle model are associated (step S23). Thereby, the applicable vehicle model can be appropriately associated with the target actual vehicle. For example, if reselection is necessary, a new compatible vehicle model can be appropriately associated with the target actual vehicle.

Next, the vehicle adaptor 13 determines whether or not there is a compatible vehicle model (step S24), and if there is a compatible vehicle model (step S24: Yes), the process proceeds to step S28, while If there is no vehicle model (step S24: No), the vehicle model corresponding to the target actual vehicle is added to the statistical base estimation information (step S25), and the process proceeds to step S28. In this way, when there is no suitable vehicle model, the vehicle model corresponding to the target actual vehicle is added to the statistical-based estimation information, so that the statistical-based estimation information can be adapted to the actual situation.

On the other hand, in step S22, if the vehicle model compatible with the target actual vehicle is not unselected and there is no need to reselect the vehicle model compatible with the target actual vehicle (step S22: No), the monitor 14 It is determined whether there is a discrepancy in the movement information between the target actual vehicle and the vehicle model (step S26).

As a result, if there is a discrepancy in the movement information between the target actual vehicle and the vehicle model (step S26: Yes), the monitor 14 sets the vehicle adaptor 13 to perform reselection (step S27). , the process proceeds to step S28, while if there is no discrepancy between the movement information of the target actual vehicle and the vehicle model (step S26: No), the monitor 14 proceeds the process to step S28.

In step S28, the vehicle adaptor 13 determines whether a predetermined set time, that is, a time period for which the corrected OD data is compiled in a table format has elapsed. The set time may be, for example, in minutes or any arbitrary time unit.

As a result, if the predetermined set time has not elapsed (step S28: No), the vehicle adaptor 13 advances the process to step S21 and further executes the process on the unprocessed real-time estimation information.

On the other hand, if the predetermined elapsed time has elapsed (step S28: Yes), the vehicle adaptor 13 executes a process of adjusting the number of vehicle models included in the statistical-based estimation information (step S29). Specifically, for example, in a case where real-time estimation information for almost all vehicles can be obtained, the vehicle adaptor 13 may delete information on vehicle models that are not associated with actual vehicles. good. As a result, information about a vehicle model that does not actually exist at that time can be deleted from the statistical-based estimation data. Furthermore, when a new vehicle model is added in step S25, the vehicle adaptor 13 may delete information on vehicle models that are not associated with the same number of actual vehicles as the number of added vehicle models. good. In this way, the vehicle model can be made in accordance with the actual situation without changing the total number of vehicle models in the statistical-based estimation data. In addition, if the percentage of real vehicles for which real-time estimation information can be obtained (obtainable percentage) among real vehicles can be estimated, the ratio of the number of vehicle models associated with the real vehicle to the total number of vehicle models can be estimated. The number of vehicle models may be adjusted so that the ratio is the same as or close to the obtainable ratio. Note that the process of step S29 does not necessarily have to be executed.

Next, the vehicle adaptor 13 generates corrected OD data, here a table of corrected OD data for a certain time range, based on the statistical base estimation data and the matching relationship between the actual vehicle and the vehicle model (step S30). Specifically, the vehicle adaptor 13 uses movement information (position, etc.) according to real-time estimated information of the actual vehicle for a vehicle model that matches the actual vehicle in the vehicle model information of the statistical-based estimation data. , generates OD data. Thereby, information obtained from the actual vehicle can be included in the OD data instead of vehicle model data, and OD data that reflects the actual situation with high accuracy can be generated.

Note that the present invention is not limited to the above-described embodiments, and can be implemented with appropriate modifications without departing from the spirit of the present invention.

For example, in the above embodiment, the communication terminals for which mobile data is collected are various communication terminals such as mobile phones and smartphones carried by individuals, communication terminals installed in vehicles, etc. The present invention is not limited to this, and may be applied only to communication terminals mounted on vehicles. In this case, the mobile data from the vehicle may include vehicle destination information input into the vehicle's navigation system or the like. In this way, when only communication terminals mounted on vehicles are targeted, there is no need for the means of transportation estimator 12 to estimate whether or not the communication terminals are present in the vehicle.

In addition, when destination information is included in the mobile data, the vehicle adaptor 13 selects a vehicle model that is compatible with the vehicle equipped with the communication terminal, taking into consideration the destination of the destination information. You may also do so. In this way, it is possible to prevent the movement information between the vehicle and the vehicle model selected as compatible from being separated due to different destinations in the future, and to match the actual vehicle and the compatible vehicle model with high accuracy. Therefore, the accuracy of the corrected OD data can be further improved.

Furthermore, in the above embodiment, one compatible vehicle model is selected for one actual vehicle, and OD data is generated using the movement information of the actual vehicle instead of the movement information of the compatible vehicle model. However, the present invention is not limited to this. For example, a plurality of compatible vehicle models may be selected for one actual vehicle, and instead of movement information of each of the plurality of compatible vehicle models, a plurality of compatible vehicle models may be selected. The OD data may be generated based on movement information of an actual vehicle to which the vehicle model is determined to be suitable.

Further, in the above embodiment, the route/traffic flow distribution calculator 11, the transportation mode estimator 12, the vehicle adaptor 13, and the monitor 14 are provided in the same device, but the present invention However, these functional units may be distributed and provided in a plurality of devices, for example, a device including a route/traffic flow allocation calculator 11, a means of transportation estimator 12, a vehicle adaptor 13, and a monitoring unit. The device including the container 14 may be a separate device.

Further, in the above embodiment, the traffic flow prediction support device 10 and the traffic flow simulation device 5 are separate devices, but the present invention is not limited to this, and the traffic flow prediction support device 10 and the traffic flow simulation device 5 are separate devices. and may be configured by one device.

Furthermore, in the above embodiments, part or all of the processing performed by the processor 102 may be performed by a hardware circuit. Further, the program in the above embodiment may be installed from a program source. The program source may be a program distribution server or a storage medium (eg, a portable storage medium).

1...Traffic flow prediction system, 2...Past OD data storage unit, 3...Mobile data storage unit, 4...Corrected OD data storage unit, 5...Traffic flow simulation device, 6...Display device, 10...Traffic flow prediction support device, 11...route/traffic volume allocation calculator, 12...transportation means estimator, 13...vehicle adaptor, 14...monitor

Claims (9)

an acquisition unit that acquires movement information of a plurality of vehicle models based on past OD data;
a vehicle selection unit that selects a compatible vehicle model that is a vehicle model compatible with the actual vehicle based on movement information of a first communication terminal that is estimated to exist in a predetermined actual vehicle;
For a vehicle model that does not match the actual vehicle, OD data is generated based on the movement information of the vehicle model, and for the compatible vehicle model, the OD data is generated based on the movement information of the vehicle model . an OD data generation unit that generates OD data based on movement information of the first communication terminal that is estimated to exist in the actual vehicle that is found to be compatible;
A traffic flow prediction support device equipped with The traffic flow prediction support device according to claim 1, wherein the vehicle selection unit selects a vehicle model that satisfies a predetermined condition with respect to the movement information of the first communication terminal as the compatible vehicle model that is compatible with the actual vehicle. . The traffic flow prediction support device according to claim 1, wherein the first communication terminal is a terminal mounted on an actual vehicle. The movement information of the first communication terminal includes destination information indicating a destination of the actual vehicle,
The traffic flow prediction support device according to claim 3, wherein the vehicle selection unit selects, as the compatible vehicle model, a vehicle model heading to the same destination as the destination indicated by the destination information of the actual vehicle. The vehicle selection unit specifies the type of the actual vehicle of the first communication terminal based on the presence density of the first communication terminal based on the movement information of the first communication terminal, and the vehicle selection unit specifies the type of the actual vehicle of the first communication terminal based on the type of the actual vehicle. The traffic flow prediction support device according to claim 1, wherein the traffic flow prediction support device selects a suitable vehicle model. The movement information of the plurality of vehicle models based on the past OD data includes movement information of the bus model,
The vehicle selection unit identifies communication terminals present on the bus based on movement information of the first communication terminal, and selects a bus model suitable for the bus;
Identifying a difference between the operation information of the bus based on movement information based on the operation information of the bus model and movement information of the first communication terminal present in the bus that is determined to be compatible with the bus model. The traffic flow prediction support device according to claim 5 , further comprising an operation determination section that outputs an output. The vehicle selection unit selects a plurality of compatible vehicle models for the actual vehicle,
The OD data generation unit is based on movement information of the first communication terminal that is estimated to exist in the actual vehicle to which the plurality of suitable vehicle models are applied, instead of movement information of the plurality of suitable vehicle models. The traffic flow prediction support device according to claim 1, wherein the traffic flow prediction support device generates OD data. A traffic flow prediction support method using a traffic flow prediction support device that generates OD data, the method comprising:
The traffic flow prediction support device acquires movement information of a plurality of vehicle models based on past OD data,
The traffic flow prediction support device selects a compatible vehicle model that is a vehicle model compatible with the actual vehicle based on movement information of a first communication terminal that is estimated to exist in a predetermined actual vehicle;
The traffic flow prediction support device generates OD data based on the movement information of the vehicle model for a vehicle model that is not compatible with the actual vehicle, and generates OD data for the compatible vehicle model based on the movement information of the vehicle model . Instead, a traffic flow prediction support method that generates OD data based on movement information of the first communication terminal that is estimated to exist in the actual vehicle to which the adaptive vehicle model is adapted. A traffic flow prediction support program that is executed by a computer that generates OD data,
The said calculator,
a vehicle selection unit that selects a compatible vehicle model that is a vehicle model compatible with the actual vehicle based on movement information of a first communication terminal that is estimated to exist in a predetermined actual vehicle;
For a vehicle model that does not match the actual vehicle, OD data is generated based on the movement information of the vehicle model, and for the compatible vehicle model, the OD data is generated based on the movement information of the vehicle model . an OD data generation unit that generates OD data based on movement information of the first communication terminal that is estimated to exist in the actual vehicle that is found to be compatible;
A traffic flow prediction support program that functions as a traffic flow prediction support program.