JP2015222537A - Road traffic demand prediction device and road traffic demand prediction method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the accuracy of road traffic demand prediction by integrating future behaviors of general vehicles that are important factors for road traffic conditions in the road traffic demand prediction.SOLUTION: A road traffic demand prediction device 1 comprises a road traffic demand prediction unit 13 including an antenna 14. The antenna 14 acquires road traffic demand data relating to a road traffic demand from a road traffic demand database 2, and acquires a global context 101 that is information relating to an environment common to a plurality of people or to a specific region. The road traffic demand prediction unit 13 receives input predicted action data 104 that relates to a travel action and that is predicted from a personal context that is information relating to an environment of an individual or to a specific organization when the individual drives a vehicle, and predicts the road traffic demand using the road traffic demand data, the global context 101, and the predicted action data 104.

Description

  The present invention relates to a road traffic demand prediction apparatus and a road traffic demand prediction method.

  Conventionally, in order to predict road traffic demand, there are those that use a global context that represents the environmental situation such as weather and events, or those that use business / public vehicle operation plans together with the road traffic demand database. . As a known technique of such a technique, for example, there is a path calculation device described in Patent Document 1.

JP 2006-146889 A

  However, the invention described in Patent Document 1 described above is based on past road traffic data, future global context (information common to a plurality of personnel, information on a relatively wide specific area, etc.), business vehicles and public vehicles. The demand for road traffic demand is calculated in consideration of the operation plan (business / public vehicle operation plan). However, the invention described in Patent Document 1 does not consider the future behavior of ordinary vehicles. The future behavior of ordinary vehicles is an important factor that determines the state of road traffic.

In other words, road traffic demand for general vehicles fluctuates in the same global context, depending on the environment common to a smaller number of personnel than those related to the global context, or the personal context related to organizations with a smaller number of personnel than regions related to the global context. To do. Even with small changes in road traffic demand, traffic congestion occurs when the overall road traffic volume reaches saturation. Known road traffic demand prediction does not take into account fluctuations in road traffic related to general vehicles as a factor, and therefore cannot provide sufficient accuracy for prediction of occurrence of traffic jams.
The present invention has been made in view of the above points, and incorporates the future behavior of a general vehicle, which is an important factor in the state of road traffic, into road traffic demand prediction, and improves the accuracy of road traffic demand prediction. Objective.

  In order to solve the above problems, a road traffic demand prediction apparatus according to the present invention includes a road traffic demand prediction unit. The road traffic demand forecasting unit acquires road traffic demand data related to road traffic demand from the road traffic demand database, and acquires a global context that is a global context that is information related to a common environment or a specific region for multiple personnel. Then, a personal context that is information related to the personal environment or a specific organization is acquired. And the driving | running | working action at the time of a person driving a vehicle from a personal context is estimated. The road traffic demand is predicted using the road traffic demand data, the global context, and the prediction behavior data indicating the prediction behavior.

  The present invention predicts a driving action when an individual drives a vehicle from a personal context that is information related to the personal environment or a specific organization. The present invention predicts road traffic demand using predicted behavior data estimated from personal context together with road traffic demand data and global context. Therefore, the present invention can incorporate the future behavior of a general vehicle, which is an important factor of the road traffic state, into the road traffic demand prediction, and can improve the accuracy of the road traffic demand prediction.

It is a figure which shows the structure of the system containing the road traffic demand prediction apparatus of 1st Embodiment of this invention. It is a figure which shows the example which applied the road traffic demand prediction apparatus shown in FIG. 1 to the structure which uses the schedule of the member which belongs to an organization and an organization as a personal context. It is a figure which shows the flowchart for demonstrating the process for producing | generating the prediction action information performed with the road traffic demand prediction apparatus shown in FIG. It is the figure which showed the flowchart of the process of the road traffic demand prediction which the road traffic demand prediction part shown in FIG. 1, FIG. 2 performs. It is a figure for demonstrating the modification 1 of the road traffic demand prediction apparatus of 1st Embodiment of this invention. It is a figure which shows the flowchart for demonstrating the process for producing | generating the prediction action information of the modification 1 shown in FIG. It is a figure for demonstrating a part of flowchart shown to FIGS. 6-1 in detail. It is a figure for demonstrating the modification 2 of the road traffic demand prediction apparatus of 1st Embodiment of this invention. It is the figure which showed the system containing the road traffic demand prediction apparatus of 2nd Embodiment of this invention. It is a figure for demonstrating that the road traffic demand prediction part classified by zone shown in FIG. 8 divides | segments a target area into a travel zone. It is the figure which showed the flowchart of the process of the road traffic demand prediction which the road traffic demand prediction part according to zone of the road traffic demand prediction apparatus shown in FIG. 8 performs. It is a table | surface for demonstrating the method by which the road traffic demand prediction part according to zone of the road traffic demand prediction apparatus shown in FIG. 8 evaluates the magnitude | size which the subject's road traffic demand has on the whole demand for every driving | running | working zone. . It is the figure which showed the system containing the road traffic demand prediction apparatus of 3rd Embodiment of this invention. FIG. 13 is a diagram illustrating a flowchart of road traffic demand / congestion prediction processing executed by a road zone traffic demand / congestion prediction unit shown in FIG. 12. It is a flowchart for demonstrating the procedure in which the road traffic demand and congestion prediction part according to driving | running | working zone shown in FIG. 12 produces an average road traffic demand margin database. It is the figure which showed the system containing the road traffic demand prediction apparatus of 4th Embodiment of this invention. It is the figure which showed the flowchart of the process of the road traffic demand prediction which the road traffic demand prediction part by road link shown in FIG. 15 performs. It is a table | surface for demonstrating the method for which the road traffic demand prediction part according to road link shown in FIG. 15 evaluates the magnitude | size of the influence which a subject's road traffic demand has on the whole demand for every road link. It is the figure which showed the system containing the road traffic demand prediction apparatus of 5th Embodiment of this invention. It is the figure which showed the flowchart of the process of the road traffic demand and traffic jam prediction which the road traffic demand and traffic jam prediction part according to road link shown in FIG. 18 performs. It is a flowchart for demonstrating the procedure in which the road traffic demand and traffic congestion prediction part according to road links shown in FIG. 18 produces the average road traffic demand margin database for every road link. It is the figure which showed the system containing the road traffic demand prediction apparatus of 6th Embodiment of this invention. It is a flowchart for demonstrating the process performed by the road traffic demand prediction part of 6th Embodiment shown in FIG. It is a flowchart for demonstrating the process sequence which divides | segments the object area of 6th Embodiment of this invention into a travel zone. It is a figure for demonstrating in detail a part of flowchart shown to FIGS. It is a figure for demonstrating the travel zone subdivided by the process of the flowchart shown to FIGS. It is a schematic diagram which shows the state which subdivided the traveling zone of the rectangular grid | lattice form of 6th Embodiment of this invention. It is a flowchart for demonstrating the process which determines the variable shown to step S252 from step S237 of FIG. It is a flowchart for demonstrating the subdivision of the travel zone of 7th Embodiment of this invention. It is the schematic diagram showing the object area where the subdivision of the travel zone of 7th Embodiment of this invention was performed. It is the figure which overlapped and showed the position of facilities etc. on the run zone subdivided by the method of a 7th embodiment of the present invention. It is a flowchart for demonstrating the subdivision of the target area of 8th Embodiment of this invention. It is a schematic diagram which shows the state which subdivided the object area in 8th Embodiment of this invention. It is the figure which overlapped and showed the position of facilities etc. on the run zone subdivided by the method of a 7th embodiment of the present invention.

Hereinafter, embodiments of the road traffic demand prediction apparatus and the road traffic demand prediction method of the present invention will be described.
[First Embodiment]
1.1 Configuration of Road Traffic Demand Prediction Device FIG. 1 is a diagram illustrating a configuration of a system including a road traffic demand prediction device according to the first embodiment. The system shown in FIG. 1 includes a road traffic demand prediction device 1 and a road traffic demand database 2. The road traffic demand database 2 stores road traffic demand data 103 relating to road traffic demand.
The road traffic demand prediction apparatus 1 includes a personal context input unit 11, a travel behavior prediction unit 12, and a road traffic demand prediction unit 13. The road traffic demand prediction unit 13 includes an antenna 14 that acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103.

  The global context 101 is a context (a group of information) related to an environment common to a plurality of personnel or a specific region. In the first embodiment, for example, information about weather or information about an event is used as the global context. In the first embodiment, the global context related to the event relates to an environment common to a relatively large number of personnel such as the content of the event, the date and time of the event, and the day of the week. In addition, the global context regarding weather relates to a relatively wide specific area such as a prefecture or a municipality.

  The business / public vehicle operation plan information 102 is information related to the business vehicle and public vehicle operation plan. In the first embodiment, the business vehicle is a so-called business vehicle, for example, a vehicle used for transporting goods. In the first embodiment, in a so-called commercial vehicle, for example, a vehicle used for transporting passengers such as a route bus is referred to as a public vehicle. The road traffic demand data 103 includes average road traffic demand data that is an average road traffic demand of a plurality of personnel, and personal road traffic demand data that is an average road traffic demand for each individual.

  The personal context is a context related to a smaller number of personnel than the personnel related to the global context, or a region smaller than the region related to the global context. In the first embodiment, as personal context, for example, information regarding an environment common to an organization such as a company, a group, and a local government, or a member (member) belonging to the organization is used.

As the personal context, for example, it is possible to use at least one of information of an individual who has accepted the provision of information, or information on an organization that has accepted the provision of information, and public information that is available without consent. At this time, the personal context input unit 11 can acquire, for example, at least one of the schedule of the organization and the schedule of the members belonging to the organization as the personal context. Also, the personal context input unit 11 can acquire, for example, at least one of the information provision service member schedule and the history of information exchanged by the member on the Internet as a personal context. Furthermore, the personal context input unit 11 can acquire, for example, information distributed on the Internet cloud service as a personal context.
The driving behavior prediction unit 12 predicts the driving behavior when the individual drives the vehicle from the personal context. The road traffic demand prediction unit 13 predicts road traffic demand using the road traffic demand data, the global context 101 and the prediction behavior information 104 indicating the prediction behavior predicted by the travel behavior prediction unit 12.

  FIG. 2 shows an example in which the road traffic demand prediction apparatus 1 shown in FIG. 1 is applied to a configuration in which an organization and a schedule of members belonging to the organization are used as a personal context. The road traffic demand prediction apparatus 3 shown in FIG. 2 includes a movement schedule extraction unit 21, a travel route / time prediction unit 22, and a road traffic demand prediction unit 13. 2 acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103 by the antenna 14, as shown in FIG. (The illustration is omitted in FIG. 2).

  In the road traffic demand prediction apparatus 3 shown in FIG. 2, the movement schedule extraction unit 21 extracts the movement schedule information 202 from the schedule of the organization and the schedule of the members of the organization that have previously agreed to provide information. At this time, the movement schedule extraction unit 21 refers to the outlook 201 on the intranet of the organization and generates member movement schedule information 203. The member movement schedule information 203 is output to the travel route / time prediction unit 22.

  The travel route / time prediction unit 22 acquires address information 207 regarding the addresses of the members of the organization from the address database (D / B) 25. Further, the travel route / time prediction unit 22 is connected to the member movement history D / B 26 and acquires the member movement history information 204. Note that the movement history information 204 includes information on the departure place, destination, and means of transportation from the departure place to the destination (transport mode: car, train, airplane, walking, etc.) in the member's schedule. ing.

  Further, the travel route / time prediction unit 22 acquires the network information 205 related to the road network from the road network data D · B 23 and the location information 206 related to the place from the POI (Point Of Interest) D · B 24. The POID / B24 may store, for example, information on a place arbitrarily selected by a member. Moreover, the member may accumulate information regarding places that have been set as departure points, destinations, and waypoints in the past.

The travel route / time prediction unit 22 predicts the travel route / time using the member movement schedule information 203 together with the information acquired from each database. The travel route / time prediction unit 22 generates information on the travel route / time prediction result using the prediction result, and outputs the information as predicted behavior information 104.
At this time, the travel route / time prediction unit 22 serves as a center function that aggregates information of one or more organizations that have accepted the provision of the personal context and makes a road traffic demand prediction. For this reason, from the viewpoint of information security of the organization, the first embodiment may place a computer functioning as the travel route / time prediction unit 22 inside the organization. Then, the travel route / time prediction unit 22 performs a process of generating the prediction behavior information 104 indicating the travel route / time prediction result inside the organization, and only the travel route / time prediction result is transmitted to the road traffic demand prediction unit 13. It is possible to output.
In the road traffic demand prediction apparatus 3 described above, the travel route / time prediction unit 22 corresponds to a personal context acquisition unit. The travel route / time prediction unit 22 corresponds to a travel behavior prediction unit. The road traffic demand prediction unit 13 performs road traffic demand prediction using the predicted behavior information 104.

1.2 Road Traffic Demand Prediction Method FIG. 3 is a diagram illustrating a flowchart for explaining a process for generating the predicted behavior information 104 performed by the road traffic demand prediction apparatus 1 shown in FIG. The program for executing the flowchart shown in FIG. 3 operates on a general-purpose computer and executes road traffic demand prediction processing. Moreover, a program may operate | move on one computer and may be installed in several different computers for every function, and may perform road traffic demand prediction jointly.

  The movement schedule extraction unit 21 illustrated in FIG. 2 inputs information of the member scheduler of the Outlook 201 as a personal context. The travel route / time prediction unit 22 reads the start time, end time, location, and contents of the future (for example, tomorrow) schedule of each member (individual) recorded in the member scheduler (step S31). Then, the event location in the read schedule is set as the destination (step S32).

  The travel route / time prediction unit 22 determines whether or not the schedule read in step S31 is the first schedule of the day (step S33). As a result of the determination, if it is the first schedule of the day (step S33: Yes), the departure place is set to the home (step S34). If the read schedule is not the first schedule as a result of the determination in step S33 (step S33: No), the destination of the schedule immediately before the read schedule is set as the departure place (step S35).

  Next, the travel route / time prediction unit 22 searches the member movement history D · B26 (step S36). Then, the same OD pair (match) as the set OD pair that is the combination (pair) of the destination (D) set in step S32 and the departure point (O) set in step S34 or S35 in the member movement history D · B26 It is determined whether there is an OD pair) (step S37). As a result of the determination in step S37, if there is no matching OD pair in the member movement history D / B26 (step S37: No), the member movement history D / B26 is searched (step S38), and the departure point, destination, movement It is determined whether there is a similar OD pair having a similar movement status such as a time zone (step S39). When there is no similar OD pair in the member movement history D · B26 (step S39: No), the travel route / time prediction unit 22 is configured to extract the processes from step S31 to step S42 by the planned movement extraction unit 21. It is determined whether or not all employees have been performed (step S43).

  If it is determined in step S37 that there is a matching OD pair (step S37: Yes), or if it is determined in step S39 that there is a similar OD pair (step S39: Yes), the travel route / time prediction unit 22 matches. It is determined whether or not the road traffic mode corresponding to the OD pair or the similar OD pair is a car (step S40). If the road traffic mode is not an automobile as a result of the determination in step S40 (step S40: No), the travel route / time prediction unit 22 is configured to extract the processing from step S31 to step S42 by the planned movement extraction unit 21. It is determined whether or not all employees have been performed (step S43).

  If it is determined in step S40 that the road traffic mode corresponding to the matching OD pair or the similar OD pair is a car (step S40: Yes), the travel route / time prediction unit 22 determines the start time of the schedule and a predetermined margin time. And the movement start time and end time are set from the average movement possession time required for moving between the departure place and the destination (step S41). The travel route / time prediction unit 22 then sets the departure point, destination, movement start time, movement end time, history of the route between the matching OD pair or similar OD pair (in the figure, “history pair route”). Is output as predicted behavior information 104 (step S42).

  Further, the travel route / time prediction unit 22 determines whether or not the processing from step S31 to step S42 has been performed for all the members extracted by the movement schedule extraction unit 21 (step S43). As a result of the determination, when the travel route / time prediction unit 22 determines that the process has been completed for all the members (step S43: Yes), the process of generating the predicted behavior information 104 ends. If it is determined that there is a member who has not been processed, the travel route / time prediction unit 22 reads the schedule of the unprocessed member in step S31 (step S31).

1.3 Road Traffic Demand Prediction Unit FIG. 4 is a diagram showing a flowchart of the road traffic demand prediction process executed by the road traffic demand prediction unit 13 shown in FIGS. The symbols or functions shown in FIG. 4 indicate the following factors.

なお、上記した因子のうち、以降、個人の将来道路交通需要を「個人の将来的な道路交通需要」と記し、個人の将来的な道路交通需要から推計して得られる将来道路交通需要と区別する。

Of the above factors, the individual future road traffic demand will be referred to as “individual future road traffic demand” and will be distinguished from the future road traffic demand obtained from the individual future road traffic demand. To do.

  The road traffic demand prediction unit 13 is based on the global context (for example, day of the week, time, weather, event contents, etc.) 101 shown in FIG. 1 and the road traffic demand data 103 stored in the road traffic demand D / B2. The average road traffic demand in the environment indicated by the global context (hereinafter referred to as “in the global context”) is extracted (step S51). The average road traffic demand represents the past average road traffic demand in tomorrow's global context to be estimated.

  Next, the road traffic demand prediction unit 13 adds up the individual future travel route / time prediction results to obtain a sum, and tabulates the individual future road traffic demand (step S52). The future road traffic demand of the individual represents the personal road traffic demand based on, for example, the day of the week, time, weather, event, etc., which are the global context of “Tomorrow”. Furthermore, the road traffic demand prediction part 13 extracts the average road traffic demand of the individual in global context from the road traffic demand D * B2 shown in FIG. 1 (step S53). The individual average road traffic demand represents the individual's average road traffic demand in the global context.

  The road traffic demand prediction unit 13 grasps in advance the relationship between changes in the average road traffic demand of individuals in the global context and changes in the average road traffic demand. In this way, the road traffic demand prediction unit 13 estimates the future road traffic demand in the global context from the individual future road traffic demand, the individual average road traffic demand, and the average road traffic demand (step S54). ). The function for obtaining the future road traffic demand is given by the following equation (1).

なお、式（１）中のαは、重み付け係数を示している。

In Expression (1), α represents a weighting coefficient.

[Modification]
The first embodiment of the present invention is not limited to the configuration described above. Hereinafter, the modification which acquires the personal context of 1st Embodiment is demonstrated. In the description of the modification, the same reference numerals are given to the same components as those described in the first embodiment, and a part of the description will be omitted.

・ Modification 1
FIG. 5 is a diagram for explaining a first modification of the first embodiment. The road traffic demand prediction device 4 according to the first modification shown in FIG. 5 includes a travel schedule extraction unit 21, a travel route / time prediction unit 22, a road traffic demand prediction unit 13, and a travel schedule estimation unit 58. The road traffic demand forecasting device 4 is a personal scheduler 55 or mail history, Internet search history, SNS history of members (members of the road traffic information providing service, etc., hereinafter referred to simply as “members”) who have previously agreed to provide information. A member's travel schedule is extracted from the history of the text, and driving behavior is predicted.
The movement schedule extraction unit 21 refers to a personal scheduler 55 on a personal information device (hereinafter referred to as “smart phone”) such as a smartphone in which a member service application is installed. Then, the member's movement schedule information 501 is extracted from the information recorded in the referenced personal scheduler 55.

  The movement schedule estimation unit 58 uses at least one of the information provision service member schedule and the history of information exchanged by the member on the Internet as a personal context, and estimates a schedule related to the movement of the member from the personal context. In other words, the movement schedule estimation unit 58 performs data mining on movement behavior from the history information 502 of information 56 exchanged over the Internet, such as email, the Internet, and SNS, transmitted and received by a member using a smartphone, and the estimated movement schedule information 503. Is generated. The travel route / time prediction unit 22 is connected to the member movement history D / B 57 and acquires member movement history information 504. Note that the movement history information 504 includes the departure place, destination, and moving means (transport mode: car, train, airplane, walking, etc.) from the departure place to the destination in the member's schedule.

Further, the travel route / time prediction unit 22 acquires address information 207, network information 205, and location information 206. Then, the travel route / time prediction unit 22 predicts the travel route of the member, the movement start time and the end time, generates member travel route / time prediction result information, and sends the predicted behavior information 104 to the road traffic demand prediction unit 13. Output.
FIGS. 6A and 6B are flowcharts for explaining the process for generating the predicted behavior information 104 of the first modification. As illustrated in FIG. 6A, the movement schedule extraction unit 21 illustrated in FIG. 5 inputs the information of the personal scheduler 55 as a personal context. Moreover, the movement schedule estimation part 58 estimates a member's movement action from the historical information 502, and produces | generates the estimated movement schedule information 503 (step S71). Then, the location of the event in the estimated moving behavior is set as the destination (step S72).

  Here, the processing of the subroutine shown in step S71 will be described with reference to FIG. As illustrated in FIG. 6B, the movement schedule estimation unit 58 reads the start time, end time, location, and contents of each member's future (for example, tomorrow) schedule from the member's personal scheduler 55 (step S711). . Next, data mining regarding moving behavior is performed from the history information 502 of information exchanged by e-mail, search, SNS, etc. (step S712), and the moving behavior is estimated (step S713). The movement schedule estimation unit 58 sorts the member's personal schedule read in step S711 and the action schedule indicated by the estimated movement schedule information 503 estimated in step S713 in time series (step S714). The schedule or the like obtained as a result of sorting becomes the estimated movement schedule information 503 of the first modification.

  The travel route / time prediction unit 22 determines whether or not the schedule to be processed is the first schedule of the day from the estimated movement schedule information 503 (step S73), and if it is the first schedule of the day ( Step S73: Yes), the departure place is set to the home (Step S74). When the read schedule is not the first schedule as a result of the determination in step S73 (step S73: No), the travel route / time prediction unit 22 sets the destination of the schedule immediately before the read schedule as the departure place. Set (step S75).

  Next, the travel route / time prediction unit 22 searches the member movement history D · B22 (step S76). The member movement history D · B22 has the same matching OD pair as the set OD pair that is a combination (pair) of the destination (D) set in step S72 and the departure place (O) set in step S74 or step S75. Whether or not (step S77). If the result of determination in step S77 is that there is no matching OD pair in the member movement history D · B22 (step S77: No), the member movement history D · B22 is searched (step S78), and the departure point, destination, and movement time zone It is determined whether there is a similar OD pair having a similar movement status (step S79). If there is no similar OD pair in the member movement history D · B22 (step S79: No), the travel route / time prediction unit 22 performs the processing from step S72 to step S82 on the member extracted by the movement schedule extraction unit 21. It is determined whether or not all the processes have been performed (step S83).

  If it is determined in step S77 that there is a matching OD pair (step S77: Yes), or if it is determined in step S79 that there is a similar OD pair (step S79: Yes), the travel route / time prediction unit 22 matches. It is determined whether or not the road traffic mode corresponding to the OD pair or the similar OD pair is a car (step S80). As a result of the determination in step S80, if the road traffic mode is not an automobile (step S80: No), the travel route / time prediction unit 22 performs the processing from step S72 to step S82 on the members extracted by the movement schedule extraction unit 21. It is determined whether or not all of the above have been performed (step S83).

  If it is determined in step S80 that the road traffic mode corresponding to the matching OD pair or the similar OD pair is a car (step S80: Yes), the travel route / time prediction unit 22 determines the start time of the schedule and a predetermined margin time. And the movement start time and end time are set from the average movement possession time necessary for moving between the departure place and the destination (step S81). The travel route / time prediction unit 22 then sets the departure point, destination, movement start time, movement end time, history of the route between the matching OD pair or similar OD pair (in the figure, “history pair route”). Is output as the predicted behavior information 104 (step S82).

  Furthermore, the travel route / time prediction unit 22 determines whether or not the processing from step S72 to step S82 has been performed for all the members extracted by the movement schedule extraction unit 21 (step S83). As a result of the determination, when the travel route / time prediction unit 22 determines that the process has been completed for all the members (step S83: Yes), the process of generating the predicted behavior information 104 ends. When it is determined that there is a member that has not been processed (step S83: No), the travel route / time prediction unit 22 reads the schedule of the unprocessed member in step S71 (step S71).

・ Modification 2
FIG. 7 is a diagram for explaining a second modification of the first embodiment. The road traffic demand prediction device 5 of the second modification shown in FIG. 7 includes a travel schedule extraction unit 21, a travel route / time prediction unit 22, a road traffic demand prediction unit 13, and a travel schedule estimation unit 78. The movement schedule estimation unit 78 uses information distributed on the Internet cloud service 79 as a personal context, and estimates a schedule related to movement of personnel from the personal context.

  In the second modification, the movement behavior of people (a plurality of individuals who do not depend on the affiliation of organizations, associations, etc.) is estimated from information distributed on the Internet. The movement schedule estimation unit 78 acquires information available on the Internet such as the SNS message 701, the blog message 702, and the search history 703 from the Internet cloud service 79. Further, the movement schedule estimation unit 78 collects the position / movement information 706 of the target person who estimates the movement schedule from a positioning unit 71 such as GPS (Global Positioning System). The location / movement information 706 includes the current location of the person (information provider) who provided the location information to the road traffic demand prediction device 5 and the movement information in which the location of the current location is recorded in association with the time. .

The movement schedule estimation unit 78 extracts information on an event in which people are interested in movement by data mining from the SNS message 701, the blog message 702, and the search history 703. And the behavior regarding the future movement of people is estimated from the extracted information. From the estimated result, the movement schedule estimation unit 78 creates estimated movement schedule information 705 and outputs it to the travel route / time prediction unit 22.
The travel route / time prediction unit 22 is connected to the general movement history D / B 77 and acquires the movement history information 704 of the individual who has been permitted to provide information. Note that the movement history information 704 includes a departure place, a destination, and a means of moving from the departure place to the destination (transport mode: car, train, airplane, walking, etc.) of the movement that the individual has made in the past. .

Further, the travel route / time prediction unit 22 acquires address information 207, network information 205, and location information 206. Then, the travel route / time prediction unit 22 predicts the individual travel route, the movement start time and the end time, generates personal travel route / time prediction result information, and sends the predicted behavior information 104 to the road traffic demand prediction unit 13. Output.
The road traffic demand prediction unit 13 predicts the road traffic demand using the personal travel route / time prediction result information obtained by estimating the individual travel route, movement start time and end time. For this reason, 1st Embodiment can take in the future behavior of a general vehicle from a personal context, and can improve the prediction precision of a road traffic demand.

[Second Embodiment]
2.1 Configuration of Road Traffic Demand Prediction Device Hereinafter, a second embodiment of the present invention will be described. In the description of the second embodiment, the same components as those described in the first embodiment are denoted by the same reference numerals, and the description thereof is partially omitted. The road traffic demand prediction apparatus according to the second embodiment divides a target area to be subjected to road traffic demand prediction into smaller areas (hereinafter referred to as “travel zones”) and performs prediction for each travel zone. Is.

FIG. 8 is a diagram showing a system including the road traffic demand prediction device 6 of the second embodiment. The road traffic demand prediction device 6 includes a personal context input unit 11, a travel behavior prediction unit 82, and a road traffic demand prediction unit 83 for each travel zone. The travel zone-specific road traffic demand prediction unit 83 includes an antenna 14 that acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103.
The driving behavior prediction unit 82 predicts the driving behavior when the individual drives the vehicle from the personal context. In the second embodiment, the driving behavior prediction unit 82 predicts a road that the member is supposed to travel in the future and a time when the member travels, and indicates a future driving route / time prediction result information 114 (prediction behavior) indicating a prediction result. Information 104).

The road traffic demand prediction unit 83 for each travel zone uses the future travel route / time prediction result information 114 indicating the future travel route / time prediction result predicted by the road traffic demand data, the global context 101 and the travel behavior prediction unit 82, Predict road traffic demand for each travel zone.
FIGS. 9A and 9B are diagrams for explaining that the road traffic demand prediction unit 83 for each travel zone divides the target area into travel zones. As shown in FIG. 9A, the travel zone-specific road traffic demand prediction unit 83 divides the target area into travel zones in a grid pattern. Or the road traffic demand prediction part 83 according to driving | running | working zone divides | segments a target area into a driving | running zone according to the administrative boundary, as shown in FIG.9 (b).

2.2 Road Traffic Demand Prediction Method FIG. 10 is a diagram illustrating a road traffic demand prediction process executed by the road traffic demand prediction unit 83 for each travel zone of the road traffic demand prediction apparatus 6 shown in FIG. . In addition, the symbol or function shown in FIG. 10 shows the following factors.

  The travel zone-specific road traffic demand prediction unit 83 receives the future travel route / time prediction result information 114 output from the travel behavior prediction unit 82. Future travel route / time prediction result information 114 is generated for each travel zone. Then, the road traffic demand prediction unit 83 for each travel zone adds up the individual future travel route / time prediction results for each travel zone, sums up each travel zone, and totals the individual future road traffic demand. (Step S91). Then, the travel zone-specific road traffic demand prediction unit 83 extracts the travel zone i and the time zone T to be processed (step S92).

  The travel zone-specific road traffic demand prediction unit 83 inputs the global context 101, and extracts the average road traffic demand in the global context of the travel zone i and the time zone T extracted from the road traffic demand D · B2 in step S92 (step S92). S93). Subsequently, the road traffic demand prediction unit 83 for each travel zone extracts the average road traffic demand of the members in the global context of the travel zone i and the time zone T extracted in step S92 from the road traffic demand D · B2 (step S94). ).

The travel zone-specific road traffic demand prediction unit 83 obtains the magnitude of the influence of the members of the travel zone by a method to be described later, and determines the magnitude of the influence of the members of the travel zone as predetermined rank A, rank B, rank Allocate to any of C (step S95).
If the magnitude of the members of the impact of traveling zone i corresponds to the rank A, traveling zone by road traffic demand prediction unit 83, to estimate future road traffic demand by using the function f _A (step S96). Further, the road traffic demand prediction unit 83 for each travel zone estimates the future road traffic demand using the function f _B when the influence of the members of the travel zone i corresponds to rank B (step S97). When the magnitude of the influence of the members in the traveling zone i corresponds to the rank C, the future road traffic demand is estimated using the function f _C (step S98). The function f _A , the function f _B , and the function f _C are expressed by equations with different values of the weighting coefficient α in the equation (1).

In the second embodiment, when the influence rank is medium (for example, rank B), the future road traffic demand is estimated using a function (for example, function f _B ) having a relatively small weighting coefficient. When the influence rank is relatively high (for example, rank A), the future road traffic demand is estimated using a function (for example, function f _A ) having a relatively large weighting coefficient. Furthermore, in the second embodiment, when the influence rank is relatively low (for example, rank C), a future road is used by using a function (for example, function f _C ) having a relatively small weighting coefficient (for example, “0”). Estimate traffic demand.

  FIG. 11 is a diagram for explaining a method in which the road traffic demand prediction unit 83 for each travel zone shown in FIG. 8 evaluates the magnitude of the influence of the target person's road traffic demand on the entire road traffic demand for each travel zone. It is a table. The rows of the table shown in FIG. 11 indicate the traveling zones, and the columns indicate the number of organizations corresponding to the traveling zones (number of factories, offices, government offices, etc.), the total number of members of the organizations, and the number of members' homes. The number of POIs (in the second embodiment, including not only unusual POIs such as sightseeing spots and event venues but also daily stops such as commercial facilities, hospitals, government offices, etc.), length of main roads, general The road length of the road, the number of intersections, etc. (road attributes) are counted, and the influence of the road traffic demand in the travel zone on the average road traffic demand is evaluated by ranking.

  In the second embodiment, the magnitude of influence is set to three levels, and each level is represented by “×”, “◯”, and “「 ”. For example, “x” shown in FIG. 11 indicates the level having the least influence, “◎” indicates the level having the greatest influence, “◯” is greater than “×”, and the influence is greater than “◎”. Indicates a small level. The travel zone-specific road traffic demand prediction unit 83 aggregates “×”, “◯”, “◎” for each travel zone, and determines the influence rank (A, B, C in the second embodiment) of the travel zone. .

  In the second embodiment, all the parameters shown in FIG. 11 (the number of organizational bodies for each traveling zone, the number of members of the organizational bodies, the number of departure points related to the movement of the members, the number of POIs, road attributes) It is not limited to the one that uses to evaluate the degree of influence of individual road traffic demand. The second embodiment only needs to evaluate the degree of the influence of the individual road traffic demand using at least one of such parameters.

[Third Embodiment]
3.1 Configuration of Road Traffic Demand Prediction Device Hereinafter, a third embodiment of the present invention will be described. In the description of the third embodiment, components similar to those described in the first embodiment and second embodiment are denoted by the same reference numerals, and the description thereof is partially omitted. The road traffic demand prediction apparatus according to the third embodiment is based on the average road traffic demand margin for each travel zone extracted from the road traffic demand D · B2 and the future of the individual predicted from the individual average road traffic demand and the personal context. Future road traffic congestion is estimated using road traffic demand.

  FIG. 12 shows a system including the road traffic demand prediction device 7 of the third embodiment. The road traffic demand prediction device 7 includes a personal context input unit 11, a travel behavior prediction unit 122, and a road traffic demand / congestion prediction unit 123 for each travel zone. The road traffic demand / congestion prediction unit 123 by travel zone includes an antenna 14 that acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103.

The driving behavior prediction unit 122 predicts the driving behavior when the individual drives the vehicle from the personal context. In the third embodiment, the driving behavior prediction unit 122 predicts a road that the member is expected to travel in the future and a time when the member travels, and generates future travel route / time prediction result information 114 indicating a prediction result. .
The road traffic demand / congestion prediction unit 123 for each travel zone includes road travel demand data, future travel route / time prediction result information 114 indicating the future travel route / time prediction result predicted by the global context 101 and the travel behavior prediction unit 122. Used to predict road traffic demand and traffic congestion for each driving zone.

2.2 Road Traffic Demand / Congestion Prediction Method FIG. 13 is a flowchart of road traffic demand / congestion prediction processing executed by the road traffic demand / congestion prediction unit 123 for each travel zone of the road traffic demand prediction apparatus 7 shown in FIG. FIG. The symbols or functions shown in FIG. 13 indicate the following factors.

The road traffic demand / congestion prediction unit 123 according to the travel zone uses the global context 101 and the road traffic demand data 103 acquired from the road traffic demand database 2 to calculate the average roads of the members' travel zones i and time zones k in the global context 101. A traffic demand is extracted (step S131). Next, the road traffic demand / congestion prediction unit 123 for each travel zone calculates the average road traffic demand margin for the travel zone i and time zone k in the global context 101 from the average road traffic demand margin calculated using the road traffic demand data 103. Data is extracted (step S132). In the third embodiment, the extracted average road traffic demand margin data is accumulated to create average road traffic demand margin data D / B124.
The average road traffic demand margin represents the level of road traffic demand that can be accepted before the occurrence of traffic jams, such as how much traffic traffic demand increases with respect to average road traffic demands.

  Next, the road traffic demand / congestion prediction unit 123 for each travel zone extracts the road traffic demand of the travel zone i and the time zone k of the member of the next day from the future travel route / time prediction result information 114 of the member ( Step S133). Subsequently, the road traffic demand / congestion prediction unit 123 for each travel zone obtains the average value of the road traffic demand of the past member corresponding to the next day in the travel zone i and the time zone k from the road traffic demand D / B2. Then, the difference (difference) between the road traffic demand of the next day member's travel zone i and time zone k and the average value of the next day's member's travel zone i and time zone k's road traffic demand is calculated (step S134). ). Furthermore, the road traffic demand / congestion prediction unit 123 for each travel zone compares the difference calculated in step S134 with the average road traffic demand margin (step S135).

  As a result of the comparison in step S135, if the difference is smaller than the average road traffic demand margin (step S135: No), it is predicted that no traffic jam will occur. At this time, the travel-zone-specific road traffic demand / congestion prediction unit 123 proceeds to step S137 and determines whether or not the above processing is completed for all global contexts g, all travel zones i, and all time zones k (step S137). ). When it is determined that the processing has been completed for all the global contexts g, all the travel zones i, and all the time zones k (step S137: Yes), the road traffic demand / congestion prediction unit 123 for each travel zone ends all the processing.

  On the other hand, if the difference is larger than the average road traffic demand margin in step S135 (step S135: Yes), it is predicted that a traffic jam will occur. For this reason, the road traffic demand / congestion prediction unit 123 for each travel zone calculates the difference between the difference and the average road traffic demand margin (step S136), and calculates the difference between the difference and the average road traffic demand margin as the travel zone i, time. It is output as a predicted value of the degree of traffic jam in band k (step S136). The travel zone-specific road traffic demand / congestion prediction unit 123 determines whether or not the above processing has been completed for all global contexts g, all travel zones i, and all time zones k (step S137). The road traffic demand / congestion prediction unit 123 for each travel zone terminates all the processes when the processes for all the global contexts g, all the travel zones i, and all the time zones k are completed (step S137: Yes), If the process has not been completed (step S137: No), the process returns to step S131 to execute the next process.

FIG. 14 is a flowchart for explaining a procedure by which the road traffic demand / congestion prediction unit 123 for each travel zone creates the average road traffic demand margin database 124 for each travel zone.
The symbols or functions shown in FIG. 14 indicate the following factors.

  The road traffic demand / congestion prediction unit 123 for each travel zone extracts the road traffic demand for the travel zone i and the time zone k in the global context from the road traffic demand database 2 (step S141). Next, an average value of road traffic demand (hereinafter also referred to as “data group”) in a plurality of travel zones i and a plurality of time zones k is calculated (step S142). Next, the road traffic demand / congestion prediction unit 123 for each travel zone calculates the maximum value of the data group (step S143). The maximum value calculated in step S143 is a saturated road traffic flow, and it is considered that a traffic jam occurs if there is a road traffic demand that exceeds the saturated road traffic flow.

Furthermore, the road traffic demand / congestion prediction unit 123 for each travel zone calculates the difference between the average value and the maximum value of the data group (step S144). The difference between the average value and the maximum value of the data group calculated in step S144 is an average road traffic demand margin in the travel zone i and the time zone k in the global context. The road traffic demand / congestion prediction unit 123 for each travel zone writes the average road traffic demand margin in the average road traffic demand margin D · B 124 (step S145).
The travel zone-specific road traffic demand / congestion prediction unit 123 determines whether or not the processing has been completed for all global contexts g, all travel zones i, and all time zones k (step S146). When the process is finished, all the processes are finished (step S146: Yes), and when the process is not finished (step S146: No), the process returns to step S141 to execute the next process.

[Fourth Embodiment]
4.1 Configuration of Road Traffic Demand Prediction Device Hereinafter, a fourth embodiment of the present invention will be described. In the description of the fourth embodiment, components similar to those described in the first to third embodiments are denoted by the same reference numerals, and a part of the description is omitted. The road traffic demand prediction apparatus according to the fourth embodiment uses a target area for which road traffic demand prediction is performed as a road, divides the road into road links, and performs prediction for each road link. In the fourth embodiment, a road is divided at a branch point, and each of the divided roads (road sections) is referred to as a “road link”.

  FIG. 15 shows a system including the road traffic demand prediction device 8 of the fourth embodiment. The road traffic demand prediction device 8 includes a personal context input unit 11, a travel behavior prediction unit 152, and a road traffic demand prediction unit 153 for each road link. The road traffic demand prediction unit 153 for each road link includes an antenna 14 that acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103.

The driving behavior prediction unit 152 predicts the driving behavior when the individual drives the vehicle from the personal context. In the fourth embodiment, the driving behavior prediction unit 152 predicts a road that the member is expected to travel in the future and a time when the member travels, and generates future travel route / time prediction result information 154 indicating a prediction result. .
The road traffic demand prediction unit 153 by road link uses the future travel route / time prediction result information 154 indicating the future travel route / time prediction result predicted by the road traffic demand data, the global context 101 and the travel behavior prediction unit 152, Predict road traffic demand for each road link.

4.2 Road Traffic Demand Prediction Method FIG. 16 is a diagram showing a flow chart of road traffic demand prediction processing executed by the road link demand prediction unit 153 for each road link of the road traffic demand prediction device 8 shown in FIG. . Note that the symbols or functions shown in FIG. 16 indicate the following factors.

  The road traffic demand prediction unit 153 by road link receives the future travel route / time prediction result information 154 output from the travel behavior prediction unit 152. Future travel route / time prediction result information 154 is generated for each road link. Then, the road traffic demand prediction unit 153 for each road link adds up the individual future travel route / time prediction results for each road link, sums up each road link, and totals the individual future road traffic demand. (Step S161). And the road traffic demand prediction part 153 classified by road link extracts the road link i and the time slot | zone T used as a process target (step S162).

The road link demand prediction unit 153 by road link inputs the global context, and extracts the average road traffic demand in the global context of the road link i and the time zone T extracted from the road traffic demand D · B2 in step S162 (step S163). ). Subsequently, the road traffic demand prediction unit 153 for each road link extracts the average road traffic demand of the members in the global context of the road link i and the time zone T extracted in step S162 from the road traffic demand D · B2 (step S164). ).
The road traffic demand prediction unit 153 for each road link obtains the magnitude of the influence of the members of the road link by a method to be described later, and determines the magnitude of the influence of the members of the road link by a predetermined rank A, rank B, rank Assign to any one of C (step S165).

When the magnitude of the influence of the members of the road link corresponds to rank A, the road link demand prediction unit 153 by road link estimates the future road traffic demand using the function f _A (step S166). The road link by road traffic demand forecast unit 153, to estimate the future traffic demand by using the function f _B in the case where the magnitude of the members of the influence of the road link corresponding to the rank B (step S167), the road When the magnitude of the influence of the members of the link corresponds to rank C, the future road traffic demand is estimated using the function f _C (step S168). The function f _A , the function f _B , and the function f _C are expressed by equations with different values of the weighting coefficient α in the equation (1).

In the fourth embodiment, when the influence rank is medium (for example, rank B), the future road traffic demand is estimated using a function (for example, function f _B ) having a relatively small weighting coefficient. When the influence rank is relatively high (for example, rank A), the future road traffic demand is estimated using a function (for example, function f _A ) having a relatively large weighting coefficient. Furthermore, in the fourth embodiment, when the influence rank is relatively low (eg, rank C), a function (eg, function f _C ) having a weighting coefficient (eg, “0”) smaller than the weighting coefficient of rank B is used. Estimate future road traffic demand.

  FIG. 17 is a table for explaining a method by which the road traffic demand prediction unit 153 for each road link shown in FIG. 15 evaluates the magnitude of the influence of the target person's road traffic demand on the overall demand for each road link. is there. The rows in the table shown in FIG. 17 indicate road links, and the columns indicate the number of organizations corresponding to the road links (factories, offices, government offices, etc.), the total number of members of the organizations, and the number of members' homes. , The number of POIs (in the fourth embodiment, including not only unusual POIs such as sightseeing spots and event venues, but also daily stops such as commercial facilities, hospitals, government offices, etc.), length of main roads, general The road length of roads, the number of intersections, etc. are counted, and the impact of road link road traffic demand on average road traffic demand is evaluated by ranking.

In the fourth embodiment, the magnitude of influence is set to three levels, and each level is represented by “×”, “◯”, and “「 ”. For example, “x” shown in FIG. 17 indicates the level having the least influence, “◎” indicates the level having the greatest influence, “◯” is greater than “×”, and the influence is greater than “◎”. Indicates a small level. The road traffic demand prediction unit 153 for each road link counts “×”, “◯”, “◎” for each road link, and determines the influence rank (A, B, C in the fourth embodiment) of the road link. .
In the fourth embodiment, the parameters shown in FIG. 17 (the number of organizations for each traveling zone, the number of members of the organization, the number of departure points related to the movement of the members, the number of POIs, road attributes) It is not limited to the one that uses all of the above to evaluate the degree of influence of individual road traffic demand. The fourth embodiment only needs to evaluate the degree of the influence of the individual road traffic demand using at least one of such parameters.

[Fifth Embodiment]
5.1 Configuration of Road Traffic Demand Prediction Device Hereinafter, a fifth embodiment of the present invention will be described. In the description of the fifth embodiment, the same components as those described in the first to fourth embodiments are denoted by the same reference numerals, and the description thereof is partially omitted. The road traffic demand prediction apparatus according to the fifth embodiment is based on the average road traffic demand margin for each road link extracted from the road traffic demand D / B2 and the personal average road traffic demand and the future of the individual predicted from the personal context 101. The future road traffic congestion is estimated by using the road traffic demand.

  FIG. 18 shows a system including the road traffic demand prediction device 9 of the fifth embodiment. The road traffic demand prediction device 9 of the fifth embodiment includes a personal context input unit 11, a driving behavior prediction unit 152, and a road traffic demand / congestion prediction unit 183 for each road link. The road traffic demand / congestion prediction unit 183 for each road link includes an antenna 14 that acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103.

The driving behavior prediction unit 152 predicts the driving behavior when the individual drives the vehicle from the personal context. In the fifth embodiment, the driving behavior prediction unit 152 predicts a road that the member is supposed to travel in the future and a time when the member travels, and indicates future prediction route / time prediction result information 154 (prediction behavior). Information 104).
The road traffic demand / congestion prediction unit 183 by road link obtains the future travel route / time prediction result information 154 indicating the future travel route / time prediction result predicted by the road traffic demand data, the global context 101 and the travel behavior prediction unit 152. Used, and predicts road traffic demand and congestion for each road link.

5.2 Road Traffic Demand / Congestion Prediction Method FIG. 19 is a flowchart of the road traffic demand / congestion prediction process executed by the road link demand / congestion prediction unit 183 of the road traffic demand prediction device 9 shown in FIG. FIG. The symbols or functions shown in FIG. 19 indicate the following factors.

The road traffic demand / congestion prediction unit 183 by road link calculates the average road traffic demand of the member's road link i and time zone k in the global context from the global context and the road traffic demand data 103 acquired from the road traffic demand database 2. Is extracted (step S191). Next, the road traffic demand / congestion prediction unit 183 by road link calculates the average road traffic demand margin data for the road link i and the time zone k in the global context from the average road traffic demand margin calculated using the road traffic demand data 103. Is extracted (step S192). In the fifth embodiment, the extracted average road traffic demand margin data is accumulated, and average road traffic demand margin data D / B 184 is created.
The average road traffic demand margin represents the level of road traffic demand that can be accepted before the occurrence of traffic jams, such as how much traffic traffic demand increases with respect to average road traffic demands.

  Next, the road traffic demand / congestion prediction unit 183 by road link extracts the road traffic demand of the member's road link i and time zone k of the next day from the member's future travel route / time prediction result information 154 ( Step S193). Subsequently, the road traffic demand / congestion prediction unit 183 by road link obtains the average value of the road traffic demand of the member corresponding to the day corresponding to the past day from the road traffic demand D / B2 on the road link i and the time zone k. To do. Then, a difference (difference) between the road traffic demand of the next day member road link i and time zone k and the average value of the road link i of the next day member road link i and time zone k is calculated (step S194). ). Furthermore, the road traffic demand / congestion prediction unit 183 by road link compares the difference calculated in step S194 with the average road traffic demand margin (step S195).

  As a result of the comparison in step S195, if the difference is smaller than the average road traffic demand margin (step S195: No), it is predicted that no traffic jam will occur. At this time, the road traffic demand / congestion prediction unit 183 by road link proceeds to step S197 and determines whether or not the above processing is completed for all global contexts g, all road links i, and all time zones k (step S197). ). When it is determined that the processing has been completed for all global contexts g, all road links i, and all time zones k (step S197: Yes), the road link demand / congestion prediction unit 183 for each road link ends all the processing.

On the other hand, if the difference is larger than the average road traffic demand margin in step S195 (step S195: Yes), it is predicted that a traffic jam will occur. Therefore, the road link demand / congestion prediction unit 183 by road link calculates the difference between the difference and the average road traffic demand margin (step S196), and calculates the difference between the difference and the average road traffic demand margin as the road link i, time. It outputs as a predicted value of the degree of traffic jam in band k (step S196). The road traffic demand / congestion prediction unit 183 by road link determines whether or not the above processing has been completed for all road links i and all time zones k (step S197). The road traffic demand / congestion prediction unit 183 for each road link ends all the processing when the processing is completed for all road links i and all time zones k (step S197: Yes), and the processing is completed. If not (step S197: No), the process returns to step S191 to execute the next process.
FIG. 20 is a flowchart for explaining a procedure in which the road traffic demand / congestion prediction unit 183 for each road link creates the average road traffic demand margin database 184 for each road link.
In addition, the symbol or function shown in FIG. 20 shows the following factors.

  The road traffic demand / congestion prediction unit 183 by road link extracts the road traffic demand of the road link i and the time zone k in a certain global context from the road traffic demand database 2 (step S201). Next, an average value of road traffic demands (hereinafter also referred to as “data group”) for a plurality of road links i and a plurality of time zones k is calculated (step S202). Next, the road traffic demand / congestion prediction unit 183 by road link calculates the maximum value of the data group (step S203). The maximum value calculated in step S203 is a saturated road traffic flow, and it is considered that traffic congestion occurs if there is a road traffic demand that exceeds the saturated road traffic flow.

  Further, the road traffic demand / congestion prediction unit 183 by road link calculates the difference between the average value and the maximum value of the data group (step S204). The difference between the average value and the maximum value of the data group calculated in step S204 becomes the average road traffic demand margin for the road link i and the time zone k in the extracted global context. The road traffic demand / congestion prediction unit 183 by road link writes the average road traffic demand margin into the average road traffic demand margin D · B124 (step S205).

The road traffic demand / congestion prediction unit 183 by road link determines whether or not the processing has been completed for all global contexts, all road links i, and all time zones k (step S206). When the process is finished, all the processes are finished (step S206: Yes), and when the process is not finished (step S206: No), the process returns to step S201 to execute the next process.
As described above, according to the first to fifth embodiments of the present invention, it is possible to improve the prediction accuracy of road traffic demand by capturing the future behavior of a general vehicle from the personal context.

  Since the information source is information on organizational members, service members, and Internet cloud services, a wide range of personal contexts can be captured. Furthermore, by dividing the target area into travel zones and changing the road traffic demand estimation method according to the influence rank of the individual road traffic demand in the road traffic demand for each travel zone, the estimation process can be performed efficiently. Further, by using the average road traffic demand margin derived from the road traffic demand database, it is possible to efficiently estimate future road traffic congestion.

[Sixth Embodiment]
Next, a sixth embodiment of the present invention will be described. Note that in the sixth embodiment, configurations described in any of the first to fifth embodiments are denoted by the same reference numerals, and description thereof is omitted.
FIG. 21 is a diagram illustrating a system including the road traffic demand prediction apparatus 1 according to the sixth embodiment. The road traffic demand prediction apparatus 1 includes a personal context input unit 11, a travel behavior prediction unit 12, and a road traffic demand prediction unit 13. The road traffic demand prediction unit 13 includes an antenna 14 that acquires the global context 101, business / public vehicle operation plan information 102, and road traffic demand data 103. The road traffic demand prediction apparatus 1 of the sixth embodiment is different from that of the first embodiment in that the road traffic demand prediction unit 13 acquires the average road traffic demand of the members from the member road traffic demand database 211 by the antenna 14. To do. Moreover, in 6th Embodiment, the driving | running | working action prediction part 12 shall estimate the driving | running | working action of the member which comprises an organization.

6.1 Road Traffic Demand Prediction FIG. 22 is a flowchart for explaining processing executed by the road traffic demand prediction unit 13 of the sixth embodiment. Note that the symbols or functions shown in FIG. 22 indicate the following factors.

The road traffic demand prediction unit 13 adds up the future road traffic demand of the members from the future driving route / time prediction results (predicted behavior information 104) of the members predicted by the driving behavior prediction unit 12 (step S221). Here, a member refers to each individual constituting an organization that provides a personal context.
Moreover, the road traffic demand prediction part 13 reads the global context 101 (step S222). Further, the road traffic demand prediction unit 13 divides the target area in the global context 101 and divides the target area into a plurality of travel zones {Z _{ii... I} } (step S223). After dividing the target area, the road traffic demand prediction unit 13 executes the subsequent processing for each travel zone {Z _{ii... I} }.

The road traffic demand prediction unit 13 extracts the average road traffic demand in the global context from the road traffic demand D · B2 (step S224). And the road traffic demand prediction part 13 extracts the average road traffic demand of the member in the global context 101 from the road traffic demand D * B211 of the member shown in FIG. 21 (step S225). Further, the road traffic demand prediction unit 13 extracts the average road traffic demand of the members in the global context g from the future road traffic demand of the members tabulated in step S221 (step S226).
The road traffic demand prediction unit 13 estimates the future road traffic demand in the global context 101 using the information extracted by the above processing (step S227).

6.2 Division of target area (generation of driving zone)
FIG. 23A is a flowchart for explaining a processing procedure for dividing the target area into travel zones according to the sixth embodiment. The process shown in FIG. 23A shows the contents of the subroutine of step S223 shown in FIG. 22, and is executed by the road traffic demand prediction unit 13 shown in FIG.
The road traffic demand prediction unit 13 first generates a travel zone group {Z _i } by dividing the target area into travel zones having a preset initial size (step S231). The road traffic demand prediction unit 13 executes the subsequent processing for all the travel zones included in the travel zone group {Z _i } generated in step S231.

Road traffic demand prediction unit 13, takes in the index of the traveling zone, dealing with the index taken as _{Z I} (step S232). Then, road traffic demand prediction unit 13, the travel zone _{Z I} organization number in _{N Site,} configuration membered _{N member,} home number _N Home, POI number _{N Poi,} road length _{L FC,} global context of each road standard FC It reads a variation ΔQ ^g of road traffic demand in (step S233). Here, the organization refers to a providing body in the case where a personal context is provided by a unit such as a company or a group. The number of homes refers to the number of residences where members live. The road standard indicates the classification of road functions such as road classification according to the road structure ordinance.

Then, road traffic demand prediction unit 13 checks the zone size _{S Zone} travel zone _{Z I,} is compared with a minimum value _S ^{Zone min} of the traveling zone set zone size _{S Zone} advance (step S234). Zone size _{S Zone} is greater than the minimum value _S ^{Zone min} running zone (Step S234: No), the traveling zone _Z organization number in _{I N Site,} configuration membered _{N member,} home number _N Home, POI number _{N Poi,} road length _{L FC} of each road standard FC, a variation Delta] Q ^g of traffic demand in the global context, each compared to a predetermined value that is set in advance for each parameter (step S237, step S238, step S239, step S240 , Step S241, Step S242).

In step S242 from step S237, the travel zone _Z organization number in _{I N Site,} configuration membered _{N member,} home number _N Home, POI number _{N Poi,} road length _{L FC} of each road standard FC, road traffic demand in the global context at least one predetermined value (travel zone _Z organization number in _{I N} ^{Site max,} configuration membered _{N member} ^max, home number _N ^Home max, POI number _N ^{Poi max,} road each road standard FC of variation Delta] Q ^g When it is determined that the length L _FC ^{max is} greater than the fluctuation amount ΔQ ^gmax of road traffic demand in the global context (step S237, step S238, step S239, step S240, step S241, step S242: No), the road traffic demand prediction unit 13 Subdividing the travel zone _{Z I} (step S243). Then, the number of digits of the index of the travel zone is increased (step S244).

Further, when the road traffic demand prediction unit 13 determines that the answer is all yes from step S237 to step S242 (step S237, step S238, step S239, step S240, step S241, step S242: Yes), the next travel zone Z _i. Repeat the same process for.
And the road traffic demand prediction part 13 judges whether the process was complete | finished about all i (all the travel zones) (step S235). If the processing for all of the travel zone _{Z I} in step S235 has not been completed (step S234: No), the road traffic demand prediction unit 13 performs processing for the next travel zone _{Z I} returns to step S232.

On the other hand, in step S235, if the processing for all the travel zone _{Z I} is determined to be finished (step S234: Yes), the road traffic demand prediction unit 13, for all of the travel zone _{Z I,} traveling zone group _{Z i} geographic information of _{({Z ii ... i})} , the travel zone _{Z I} organization number in _{N Site,} configuration membered _{N member,} home number _N Home, POI number _{N Poi,} road length _{L FC} of each road standard FC, global out write variation Delta] Q ^g of traffic demand in the context (step S236).

In the sixth embodiment, the number of organizations N _Site in the traveling zone Z _I , the number of members N _member , the number of homes N _Home , the number of POIs N _Poi , the road length L _{FC for} each road standard _FC , road traffic in the global context not intended to be limited to determining the subdivision of the travel zone with all variation Delta] Q ^g demand. In the sixth embodiment, the number of organizations N _Site in the traveling zone Z _I , the number of members N _member , the number of homes N _Home , the number of POIs N _Poi , the road length L _{FC for} each road standard _FC , the road traffic demand in the global context as long as it determines the subdivision of the travel zone using at least one of variation Delta] Q ^g.

6.3 Subdivision of driving zones (subdivision of driving zones into a rectangular grid)
FIG. 23-2 is a flowchart for explaining the contents of the subroutine of step S234 shown in FIG. The flowchart shown in FIG. 23-2 illustrates a process for explaining a process of subdividing the travel zone into a rectangular lattice.
Figure 24 is a diagram for explaining a traveling zone Z _I is subdivided by the process of the flowchart shown in Figure 23-2. Traveling zone Z _I shown in FIG. 24, the length in the vertical direction [Delta] Y I in the _figure, the length of the lateral direction in the figure has a rectangular shape of [Delta] X _I. The flowchart shown in FIG. 23-2 illustrates an example in which the travel zone Z _I shown in FIG. 24 is subdivided into four: a travel zone Z _I1 , a travel zone Z _I2 , a travel zone Z _I3 , and a travel zone Z _I4. .

Returning to FIG. 23-2, the road traffic demand prediction unit 13 reads the end point coordinates (X _I , Y _I ) and the size (ΔX _I , ΔY _I ) of the travel zone Z _I (step S251). In the sixth embodiment, the travel zone Z _I, X direction, bisecting the Y directions. For this reason, the road traffic demand prediction unit 13 sets the size of the travel zone Z _I1 to (ΔX _I / 2, ΔY _I / 2). The road traffic demand prediction unit 13, the running figure the lower left point of the zone Z _I X coordinate of the (end point) and an end point X coordinate of the travel zone Z _I1, the travel zone Z travel zone Y coordinate of _I1 Z _I The coordinate is obtained by adding ΔY _I / 2 to the Y _I coordinate of the end point (step S252).

Further, the road traffic demand prediction unit 13 sets the size of the travel zone Z _I2 to (ΔX _I / 2, ΔY _I / 2), and travels the coordinates obtained by adding ΔX _I / 2 to the X coordinate of the end point of the travel zone Z _I. The X coordinate of zone Z _I 2 is set, and the coordinate obtained by adding ΔYI / 2 to the Y coordinate of the end point of travel zone Z _{I is set} as the Y coordinate of travel zone Z _I2 (step S253). Further, the road traffic demand prediction unit 13 sets the size of the travel zone Z _I3 to (ΔX _I / 2, ΔY _I / 2), sets the X coordinate of the end point of the travel zone Z _I as the X coordinate of the travel zone Z _I3 , and travels the Y coordinate of the end point of the zone _{Z I} and Y coordinates of the traveling zone _{Z I3} (step S254). Further, the road traffic demand prediction unit 13 sets the size of the travel zone Z _I4 to (ΔX _I / 2, ΔY _I / 2), and travels the coordinates obtained by adding ΔX _I / 2 to the X coordinate of the end point of the travel zone Z _I. The X coordinate of zone Z _{I4 is used} , and the Y coordinate of the end point of travel zone Z _I is the Y coordinate of travel zone Z _I4 (step S255).
After the above processing, the road traffic demand prediction unit 13 outputs the travel zone Z _I1 , travel zone Z _I2 , travel zone Z _I3 , travel zone Z _I4 size and endpoint coordinates, as shown in FIG. Return to the main routine.

FIG. 25 is a schematic diagram showing a state in which the traveling zone is subdivided into a rectangular lattice shape of the sixth embodiment. FIG. 25 shows the target area as an initial size, and the target area is the number of organizations N _Site , the number of members N _member , the number of homes N _Home , the number of POIs N _Poi , the road length L _FC for each road standard _FC (target the length of each road in the area standard), road traffic demand (road traffic demand D · B2 traffic demand extracted from) results variation ΔQ was subdivided according to ^g of the global context, the target area It shows a state divided into four levels of zone sizes. According to FIG. 25, it can be seen that the zone size is small at places where the starting point, destination, POI, road, etc. of the individual driving behavior are densely arranged. Further, according to FIG. 25, it can be seen that the zone size of an area in which the starting point, destination, POI, road, etc. of the personal driving action are sparse is increased.

6.4 organization number in the traveling zone _{Z I N} ^{Site max,} configuration membered _{N member} ^max, home number _N ^Home max, POI number _N ^{Poi max,} road length _L ^{FC max} of each road standard FC, road traffic in a global context Determination of demand fluctuation amount ΔQ ^{gmax In} the sixth embodiment, the number of organizations N _Site in the traveling zone Z _I , the number of members N _member , the number of homes N _Home , the number of POIs N _Poi , and the road length L _{FC for} each road standard _FC , The number of organizations N _Site ^max , the number of members N _member ^max , the number of homes N _Home ^max , the number of POIs N in the traveling zone Z _I corresponding to at least one of the fluctuation amounts ΔQ ^g of road traffic demand in the global context _Poi ^max, road length _L ^{FC max} of each road standard FC, global Road traffic demand variation Delta] Q ^gmax in the context of the running zone subdivided if it exceeds ^{(hereinafter,} referred to as "each variable maximum" or "Variable Resolution."). Such maximum values differ depending on the target area. The maximum value differs depending on the global context even if the target area is the same.

In the sixth embodiment, each variable maximum value is determined in advance for a representative global context for each target area.
The subdivision of the travel zone can be performed with higher accuracy corresponding to the travel zone as the travel zone is subdivided. However, if the travel zone is subdivided, the number of travel zones increases, thereby increasing the computational resource load. Therefore, the determination of the maximum value of each variable is required to balance the resolution of the variables so that the maximum calculation accuracy can be obtained under the limitation of calculation resources.
In the sixth embodiment, the determination of the maximum value of each variable is considered as an extreme value problem of an objective function in a multivariable space under a given constraint condition for the above reason. In the sixth embodiment, the change in the estimated value of the future road traffic demand, which is the final objective function caused by the minute change in the maximum value of each variable, is minimized within the range of the number of travel zones determined by the limitation of the calculation resource. We are looking for a state to become.

FIG. 26 shows the number of tissue bodies N _Site ^max , the number of members N _member ^max , the number of homes N _Home ^max , the number of POIs N _Poi ^max , in the traveling zone Z _I shown in steps S 237 to S 242 shown in FIG. road length L _{FC max} of each road standard ^FC, is a flowchart for explaining a process for determining a variation Delta] Q ^gmax road traffic demand in the global context. In the sixth embodiment, the process illustrated in FIG. 26 is performed in advance by the road traffic demand prediction unit 13 of the road traffic demand prediction apparatus of the sixth embodiment. However, the processing shown in FIG. 26 may be performed in advance by another personal computer and the result may be stored in the road traffic demand prediction unit 13.
The road traffic demand prediction unit 13 considers the memory capacity, calculation speed, application status, and the like of calculation resources that can be used for the process of calculating the variable maximum value, and calculates the maximum number of travel zones N _ZONE ^max that can be generated by subdivision. Set (step S261).

Next, the road traffic demand prediction unit 13 includes the number of organizations N _Site in the target area, the number of members N _member , the number of homes N _Home , the number of POIs N _Poi , the road length L _{FC for} each road standard FC, and roads in the global context. giving an appropriate initial value to each variable in the variation Delta] Q ^g of traffic demand, performing driving by zone road traffic demand prediction process (step S262).
The road traffic demand prediction unit 13 calculates the total number of zones N _ZONE from the result of the road zone traffic demand prediction process performed in step S264 (step S263). Then, to calculate the estimated value TD ii _{... i} ^g of future road traffic demand of each traveling zone. At this time, road traffic demand prediction unit 13 calculates the estimated value TD ii _{... i} ^g while changing by a relatively small value setting each variable in advance (step S264). The road traffic demand prediction unit 13, the total number of zones in the range of the maximum number of zones (step S265), the change of the sum of estimates TD ii _{... i} ^g future traffic demand to explore the condition that the minimum (Step S266). In step S266, when the estimated value satisfies the condition (step S266: Yes), the road traffic demand prediction unit 13 outputs each variable used for calculating the estimated value.

Further, when the total zone number N _ZONE is not less than or equal to the maximum zone number N _ZONE ^{max in} step S265 (step S265: No), the road traffic demand prediction unit 13 estimates the future road traffic demand TD _{ii... In} step S266 _. If the change in the total sum of _i ^g is not the smallest (step S266: No), the traveling zone _Z organization number in _{I N} ^{Site max,} configuration membered _{N member} ^max, home number _N ^Home max, POI number _N ^{Poi max,} The road length L _FC ^{max for} each road standard FC and the road traffic demand fluctuation amount ΔQ ^gmax in the global context are adjusted (step S268) and set as new variables (step S269).
Furthermore, the road traffic demand prediction unit 13 executes the process of step S262 using the set new variable.

[Seventh Embodiment]
Next, a seventh embodiment of the present invention will be described. Note that, in the seventh embodiment, configurations described in any of the first to sixth embodiments are denoted by the same reference numerals, and description thereof is omitted. The road traffic demand prediction apparatus of the seventh embodiment generates a plurality of rectangular travel zones by dividing a target area at boundaries determined by the government.
The road traffic demand prediction apparatus of 7th Embodiment is comprised similarly to the road traffic demand prediction apparatus of 6th Embodiment. For this reason, in 7th Embodiment, illustration and description of a road traffic demand prediction apparatus are abbreviate | omitted.

7.1 Subdivision of driving zones (subdivision based on administrative boundaries)
FIG. 27 is a flowchart for explaining subdivision of the travel zone according to the seventh embodiment. In the seventh embodiment, a traveling zone (also referred to as a “town chome zone” in the seventh embodiment) is expressed as a set of C zones that are areas of “town” and “chome” levels connected to each other.
The road traffic demand prediction unit 13 reads the GIS information (coordinates, area, adjacent zone) of the town chome zone {CCM ^I } (step S271). Then, the town chome zones are sorted in the order of being adjacent in one direction using the coordinate values and the information of the adjacent zones. As a result, the road traffic demand prediction unit 13 generates a town chome zone in which the C zone is stored in the order of {CCM ^I } = {CCM ₁ ^I ,..., CCM _N ^I } (step S272).

Next, the road traffic demand prediction unit 13 generates the travel zone Z _I1 by sequentially adding the areas of the C zones CCM ₁ ^I ,..., CCM _N ^I one by one (steps S273 and S274). Then, the road traffic demand prediction unit 13 compares the area of the travel zone Z _{I1 with} the area of the travel zone Z _I (step S275), and the area of the travel zone Z _I1 is approximately ½ of the area of the travel zone Z _I. Steps S273, S274, and S275 are repeated until the value is reached. Then, when the area of the travel zone Z _I1 reaches approximately ½ of the area of the travel zone Z _I (step S275: Yes), the formation of the travel zone Z _I1 is finished.

As used herein, “substantially” means that the total area of the C zone CCM ₁ ^I ,..., CCM _N ^I is added to half the area of the traveling zone Z _I. It means that the difference from the value is within a predetermined range of positive predetermined values, or is within the range of predetermined negative values.
Further, the road traffic demand prediction unit 13 sets the traveling zone {CCM _{j + 1} ^I ,..., CCM _N ^I } composed of the remaining C zones not added to the traveling zone Z _I1 as the traveling zone Z _I2 (step S277). Then, the road traffic demand prediction unit 13 writes out information on the travel zone Z _I1 {CCM _N ^I1 } and information on the travel zone Z _I2 {CCM _N ^I2 } (step S278).

FIG. 28 is a schematic diagram showing a target area where the subdivision of the travel zone according to the seventh embodiment is performed. In the example shown in FIG. 28, the traveling zone _ZI is composed of town-chome zones CCM ₁ ^I to CCM ₁₃ ^I. In the seventh embodiment, when the travel zone Z _I1 is generated, the area of the travel zone Z _I1 is approximately ½ of the area of the travel zone Z _I when the town-chome zones CCM ₁ ^I to CCM ₄ ^I are added. Has reached the end of processing.

Therefore, in the example shown in FIG. 28, the travel zone Z _I have been subdivided into a driving zone Z _I1 and travel zone Z _I2.
In addition, 7th Embodiment is not limited to the structure which subdivides a travel zone with the method shown with the flowchart shown in FIG. For example, the travel zones may be divided based on the administrative boundaries in consideration of the density of the existence of facilities such as the starting point, destination, and POI of the members of the organization.

FIG. 29 is a diagram in which positions of a departure point, a destination, a facility that can be a POI, and the like are superimposed on the traveling zone subdivided by the method of the seventh embodiment. According to FIG. 29, the area where the facilities that can be the starting point, the destination and the POI are dense is small in the size of the individual traveling zones after the subdivision, and the area where the facilities which can be the destination and the POI are sparse is the subdivision It can be seen that the size of each traveling zone is large.
Further, in the seventh embodiment, the size of the subdivided travel zone is set to at least one of the length distribution for each road standard in the target area and the road traffic demand extracted from the road traffic demand D · B2. It may change based on one.

[Eighth Embodiment]
Next, an eighth embodiment of the present invention will be described. Note that in the eighth embodiment, identical symbols are assigned to configurations described in any of the first through seventh embodiments and descriptions thereof are omitted. In the eighth embodiment, a travel zone is a path formed by connecting road links between one branch point and another branch point on a road.
The road traffic demand prediction device of the eighth embodiment is configured in the same manner as the road traffic demand prediction device of the sixth embodiment. For this reason, in 8th Embodiment, illustration and description of a road traffic demand prediction apparatus are abbreviate | omitted.

8.1 Subdivision of driving zones (driving zones are defined by road links)
FIG. 30 is a flowchart for explaining subdivision of a target area when a path composed of connected road links according to the eighth embodiment is used as a travel zone. In the eighth embodiment, a road link that is the minimum unit of road information in GIS data (a portion between one branch of a road and another branch) is used, and connected road links are connected. The route is described as a path. The eighth embodiment uses a path as a travel zone.

In the eighth embodiment, the road traffic demand prediction unit 13 reads GIS information (coordinates, lengths, connected links) of the road link {Link ^I } constituting the travel zone Z _I (step S301). In the eighth embodiment, a plurality of road links are selected in advance from the road links in the target area based on the connection state, length, and number of road links, and a path is created using the selected road links. Then, information on a plurality of paths is stored as a traveling zone {Zi}.
Road traffic demand prediction unit 13, a road link {Link ^I} which constitute the traveling zone _{Z I,} based on the coordinates and articulated link information of the road link, sorted in order of concatenation (step S302). Then, the road traffic demand prediction unit 13 sets “j” of the travel zone Z _Ij to 1 (step S303).

As a result of the sorting, the road link {Link ^I } is stored in the order of {Link ₁ ^I ,... Link _N ^I } (step S302). Next, the road traffic demand prediction unit 13 sequentially adds a plurality of Link ₁ ^I ,... Link _N ^I included in the road link {Link ^I } (steps S304, 305, and 306). When the length of the travel zone Z _I1 configured by adding the road links reaches approximately ½ of the length of the travel zone Z _I (step S305: Yes), the road traffic demand prediction unit 13 End the process of adding links. The added road link becomes the travel zone Z _I1 .
Here, the term "generally" is a road link Link ₁ ^I, ... Link _N while going sum of the length of the ^I, of the length of the length of the sum that has been summed up and running zone _{Z I} 1/2 This means that the difference from the value is in the range of a predetermined positive value or in the range of a predetermined negative value.

Next, the road traffic demand prediction part 13 produces _| generates driving _| running _{| working} zone _ZI2 by the remaining road links which are not used for the production _| generation of driving _| running _{| working} zone _ZI1 (step S307). After the above processing, the road traffic demand prediction unit 13, out write information of the road links constituting the travel zone _{Z I1} {Link ₁ ^I} and road links constituting the travel zone _{Z I2} {Link ₂ ^I} (step S308).

FIG. 31 is a schematic diagram for explaining a target area obtained by dividing a path formed by connecting the road links described above as a travel zone. In FIG. 31, a road link Link ₁ ^I ... Link ₁₁ ^I constitutes a travel zone Z _I. In the example shown in FIG. 31, when the road links Link ₁ ^I to Link ₅ ^I are added, the total length reaches approximately ½ of the length of the travel zone Z _I. For this reason, in the example shown in FIG. 31, the road link Link ₁ ^I ... Link ₅ ^I constitutes the traveling zone Z _I1 , and the road link Link ₆ ^I ... Link ₁₁ ^I constitutes the traveling zone Z _I2 .

FIG. 32 is a diagram showing the location of a facility that can be a starting point, a destination, and a POI of an individual movement superimposed on a target area obtained by dividing a path formed by connecting road links as a travel zone. According to FIG. 32, it can be seen that the traveling zone Z _I1 is relatively short in the vicinity of the facility that can be the departure point, the destination, and the POI. In addition, it can be seen that the travel zone Z _I1 tends to become relatively longer as the distance from the facility that can be the departure point, the destination, and the POI becomes longer.

  As described above, according to the sixth to eighth embodiments of the present invention, the zone that is a unit for performing the prediction process is set for each global context, the starting point (O) of the individual driving action, By subdividing according to destination (D), POI distribution, length for each road standard, change in average road traffic demand, etc., factors of road network shape, starting point, destination, POI It is possible to subdivide the zone in consideration of the spatial distribution and the factor of temporal change due to the global context, to secure an appropriate resolution according to the location of the target area, and to improve the prediction accuracy and calculation efficiency It becomes possible.

  The first to eighth embodiments described above are examples of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and even if it is a form other than this embodiment, as long as it does not depart from the technical idea of the present invention, it depends on the design and the like. It goes without saying that various modifications are possible.

(Effect of embodiment)
If it is the road traffic demand prediction apparatus and road traffic demand prediction method of embodiment of this invention, it will become possible to show the effect described below.
(1) The road traffic demand prediction apparatus 1 acquires road traffic demand data related to road traffic demand from the road traffic demand database 2, and obtains a global context 101 that is information related to a common environment or a specific area for a plurality of personnel. Input the predicted behavior data related to the driving behavior when the individual drives the vehicle predicted by the antenna 14 to be acquired and the personal context that is information related to the personal environment or a specific organization, road traffic demand data, global context And a road traffic demand prediction unit 13 that performs road traffic demand prediction using the predicted behavior data.
For this reason, the road traffic demand prediction apparatus can capture the future behavior of a general vehicle that occupies a major part of the road traffic, and can predict the road traffic demand with high accuracy.

(2) The road traffic demand prediction apparatus 1 includes a personal context acquisition unit 111 that acquires a personal context, and a traveling behavior prediction unit 12 that generates predicted behavior data from the personal context.
(3) The road traffic demand prediction unit 13 divides a target area for which road traffic demand prediction is performed into travel zones smaller than the target area, and performs prediction for each travel zone.

  For this reason, the road traffic demand prediction apparatus can reflect the characteristic for every driving | running | working zone in an individual's driving | running | working action prediction process.

(4) The personal context input unit 111 acquires, as a personal context, at least one of the information on the organization that has accepted the information provision of the individual who has accepted the information provision and the public information that is available without the consent.
For this reason, a wide range of personal contexts can be captured.
(5) The personal context input unit 111 acquires at least one of the schedule of the organization and the schedule of the members belonging to the organization as a personal context.
For this reason, a lot of individual schedules can be acquired.

(6) The personal context input unit 111 acquires at least one of the information provision service member schedule and the history of information exchanged by the member on the Internet as a personal context, and further relates to movement of the member from the acquired personal context. A movement schedule estimation unit 58 that estimates a schedule is provided.
For this reason, it is possible to acquire an individual schedule having an attribute (member of information providing service) different from the members of the organization.

(7) The personal context input unit 111 acquires information distributed on the Internet cloud service 79 as a personal context, and further estimates a schedule related to personal movement from the acquired personal context. Is provided.
For this reason, it is possible to acquire information and schedules related to a wide range of moving behaviors.
(8) The antenna 14 acquires the acquired average road traffic demand in the global context 101 and the average road traffic demand of the individual in the global context 101 as road traffic demand data, and the driving behavior prediction unit 12 predicts from the personal context The road traffic demand forecasting unit 13 estimates the future road traffic demand from the average road traffic demand, the individual average road traffic demand, and the individual future road traffic demand. To do.
For this reason, it is possible to estimate the future road traffic demand of personnel.

(9) The road traffic demand prediction unit 13 divides the target area for which road traffic demand prediction is to be performed into a plurality of travel zones, and the degree of the influence of individual road traffic demand on the road traffic demand for each travel zone To change the estimation method of future road traffic demand.
For this reason, it is possible to efficiently perform future road traffic demand estimation processing.
(10) The road traffic demand prediction unit 13 divides the target area into travel zones in units of road sections.
Therefore, it is possible to predict road traffic demand on a detailed scale.
(11) The road traffic demand prediction unit 13 determines the degree of the influence of the individual road traffic demand based on the number of organizations in each travel zone, the number of members of this organization, and the departure place related to the movement of this member. Evaluation is performed using at least one of the number, the number of POIs, and the road attribute.
For this reason, it is possible to rank the influence of individual road traffic demand with high accuracy, to efficiently perform future road traffic demand estimation processing, and to obtain high prediction accuracy.

(12) The road traffic demand prediction unit 13 calculates the future road traffic of the individual predicted from the average road traffic demand margin for each travel zone extracted from the road traffic demand database 2 and the personal average road traffic demand and personal context. Estimate future road traffic congestion using demand.
For this reason, it is possible to efficiently estimate future road traffic congestion using only the data and personal context accumulated in the road traffic demand database.
(13) The road traffic demand prediction unit 13 calculates an average road traffic demand margin from the road traffic demand for each global context, travel zone, and time zone extracted from the road traffic demand database 2.
For this reason, the average road traffic demand margin can be derived using only the data acquired from the road traffic demand database.

(14) The road traffic demand prediction unit 13 changes the size of the travel zone according to at least one of the global context and the personal context, and the size of the travel zone includes the departure point of the individual movement, the destination, the POI distribution, It changes corresponding to at least one of the length distribution for each standard of roads in the target area and the road traffic demand extracted from the road traffic demand database.
For this reason, it is possible to select a zone size of an appropriate resolution according to the geographical situation of the region, global context information, and personal context for each location of the target region, and to improve prediction accuracy and efficiency of the prediction process. Can do. In setting the zone size for each location in the target area, the zone size changes corresponding to at least one of the starting point (O), destination (D), and POI distribution of the individual driving behavior. Furthermore, the zone size changes corresponding to any one of the length distribution for each road standard in the target area and the road traffic demand extracted from the road traffic demand database. For this reason, the size of the travel zone suitable for the prediction of the road traffic demand for each location can be selected, and the prediction accuracy can be improved and the prediction process can be made more efficient.

(15) The road traffic demand prediction unit 13 divides the target area into a grid and generates a plurality of rectangular travel zones.
For this reason, the calculation efficiency of the prediction process can be increased and the processing load related to the generation of the travel zone can be reduced.
(16) The road traffic demand prediction unit 13 generates a plurality of travel zones by dividing the target area based on the boundaries determined by the government.
For this reason, it becomes possible to generate | occur | produce a travel zone efficiently according to the condition of an area.
(17) The road traffic demand prediction unit 13 sets a path formed by connecting road links between one branch point and another branch point on the road as a travel zone.
For this reason, the road traffic demand for every part of the road can be predicted directly.

(18) The road traffic demand prediction unit 13 includes the departure point of the individual's movement, the destination, the POI distribution, the length distribution for each road standard in the target area, and the road traffic demand extracted from the road traffic demand database. The size of the travel zone is set so that the value of at least one of the variables for each travel zone is equal to or less than the variable maximum value that is a preset value.
For this reason, the resolution suitable for the variable can be obtained, and the accuracy and calculation efficiency of road traffic demand prediction can be improved.
(19) The variable maximum value is set so that the total number of travel zones is equal to or less than the preset total number of travel zones, and the change in the predicted value of the future road traffic demand due to the change in the variable maximum value is minimized. The
For this reason, zone division with the highest prediction accuracy of future road traffic demand, which is an objective variable, can be performed within the range of processing calculation resources.

(20) A road traffic demand data acquisition process for acquiring road traffic demand data related to road traffic demand from a road traffic demand database, and a global for acquiring a global context that is information related to an environment common to a plurality of personnel or a specific region Input predicted behavior data related to driving behavior when an individual drives a vehicle, which is generated from the context acquisition step and personal context that is information related to the personal environment or a specific organization, road traffic demand data, global A road traffic demand prediction step for performing road traffic demand prediction using the context and predicted behavior data.
For this reason, the road traffic demand prediction method can predict the road traffic demand with high accuracy by taking in the future behavior of ordinary vehicles that occupy the main part of the road traffic.

The above-described embodiment of the present invention obtains the personal context, estimates the future road traffic demand of general vehicles, and predicts the overall road traffic demand together with the road traffic demand database. There is an effect.
The personal context includes schedules of organizations such as companies, groups, and local governments, schedules of members of the organizations, schedules of information providing service members, members email, search, text history of social network system (SNS), etc. Can be used. The personal context can use information exchanged on the Internet by members of the organization or information providing service members, search for an unspecified number of users on the Internet cloud service, SNS history, and the like.

  Moreover, the embodiment of the present invention can divide the target area into travel zones when predicting the overall road traffic demand. Then, based on the daily movement behavior of the target person who collected the personal context, the magnitude of the influence of the target person's road traffic demand on the overall demand is evaluated by the distribution of the starting point, destination, road, etc. To do. The embodiment of the present invention can perform ranking according to the result of evaluation of the magnitude of influence, and can change the method for predicting road traffic demand according to the rank. Thus, the embodiment of the present invention can efficiently predict road traffic demand. The travel zone can be generated not only by dividing the target area in a plane but also by dividing the target area as a road link unit.

  Furthermore, in the embodiment of the present invention, a road traffic demand margin from the average road traffic demand to the saturated road traffic demand may be obtained in advance for each traveling zone. It is also possible to determine whether or not the road will be saturated or oversaturated in the future based on the road traffic demand margin and the difference between the average value of the target person's road traffic demand and the future demand. .

  The present invention can be used for all apparatuses for predicting road traffic demand. Moreover, the result of road traffic demand prediction can be used for a service provided using television, radio and the Internet. Furthermore, the results of road traffic demand prediction can also be used for surveys related to traffic, transportation and tourism.

1, 3, 4, 5, 6, 7, 8, 9 Road traffic demand prediction device 2 Road traffic demand database 11 Personal context input unit 12, 82, 122, 152 Traveling behavior prediction unit 13 Road traffic demand prediction unit 14 Antenna 21 Travel schedule extraction unit 22 Travel route / time prediction unit 55 Personal scheduler 58, 78 Travel schedule estimation unit 71 Positioning unit 79 Internet cloud service 83 Road traffic demand prediction unit by travel zone 101 Global context 102 Public vehicle operation plan information 103 Road Traffic demand data 104 Predictive behavior information 111 Personal context input unit 114 Time prediction result information 123 Road traffic demand / congestion prediction unit by driving zone 124 Average road traffic demand margin database 153 Road traffic demand prediction unit by road link 154 Time prediction result information 1 3 Road Traffic Demand / Congestion Prediction Unit 184 Average Road Traffic Demand Margin Database 201 Outlook 202 Movement Schedule Information 203 Member Movement Schedule Information 204 Movement History Information 205 Network Information 206 Location Information 207 Address Information 211 Road Traffic Demand Database 501 Movement Schedule information 502 History information 503, 705 Estimated movement schedule information 504, 704 Movement history information 701 SNS message 702 Blog message 703 Search history 706 Location / movement information

Claims (20)

A road traffic demand data acquisition unit for acquiring road traffic demand data related to road traffic demand from the road traffic demand database;
A global context acquisition unit that acquires a global context that is information related to a common environment or a specific region for a plurality of personnel;
Predicted behavior data relating to a driving behavior when an individual predicted from a personal context, which is information related to an individual environment or a specific organization, drives a vehicle, and the road traffic demand data, the global context, and the predicted behavior are input. Road traffic demand forecasting section that uses data to forecast road traffic demand,
A road traffic demand prediction device comprising: A personal context acquisition unit for acquiring the personal context;
A running behavior prediction unit that generates the predicted behavior data from the personal context;
The road traffic demand prediction apparatus according to claim 1, comprising:   The road traffic demand prediction unit according to claim 1 or 2, wherein the road traffic demand prediction unit divides a target area to be a target for road traffic demand prediction into small areas and performs prediction for each small area. Prediction device.   The personal context acquisition unit acquires, as the personal context, at least one of information of an individual who has agreed to provide information, information on an organization that has accepted the provision of information, and public information that is available without requiring consent. The road traffic demand prediction apparatus according to claim 2 or 3.   4. The road traffic demand prediction device according to claim 2, wherein the personal context acquisition unit acquires at least one of a schedule of the organization and a schedule of members belonging to the organization as the personal context. 5. .   The personal context acquisition unit acquires at least one of a schedule of information providing service members and a history of information exchanged on the Internet by the member as the personal context, and further relates to movement of the member from the acquired personal context. The road traffic demand prediction apparatus according to any one of claims 2 to 5, further comprising a movement schedule estimation unit that estimates a schedule.   The personal context acquisition unit includes information that is distributed on an Internet cloud service as the personal context, and further includes a movement schedule estimation unit that estimates a schedule related to personal movement from the acquired personal context. The road traffic demand prediction apparatus according to any one of claims 2 to 5, wherein The road traffic demand data acquisition unit acquires the average road traffic demand in the global context acquired by the global context acquisition unit and the individual average road traffic demand in the global context as the road traffic demand data.
The driving behavior prediction unit generates a future road traffic demand of the individual predicted from the personal context as the predicted behavior data,
The road traffic demand prediction unit estimates the future road traffic demand from the average road traffic demand, the individual average road traffic demand, and the individual future road traffic demand. Road traffic demand forecasting device.   The road traffic demand prediction unit divides a target area to be subjected to road traffic demand prediction into small areas, and determines the future road traffic according to the degree of influence of individual road traffic demand on the road traffic demand for each small area. The road traffic demand prediction apparatus according to claim 8, wherein a demand estimation method is changed.   The road traffic demand prediction apparatus according to claim 9, wherein the road traffic demand prediction unit divides the target area into the small areas in units of road sections.   The road traffic demand prediction unit determines the degree of the influence of the individual road traffic demand on the number of the organizations for each small area, the number of members of the organizations, and the number of departure points related to the movement of the members. The road traffic demand prediction apparatus according to claim 9, wherein the evaluation is performed using at least one of the number of POIs (Point Of Interest) and road attributes.   The road traffic demand prediction unit is configured to extract the average road traffic demand margin for each small area and the individual average road traffic demand extracted from the road traffic demand database, and the individual for each small area predicted from the personal context. 10. The road traffic demand prediction apparatus according to claim 9, wherein the future road traffic congestion is estimated using the future road traffic demand.   The road traffic demand prediction unit calculates the average road traffic demand margin from the road traffic demand for each of the global context, the small area, and the time zone extracted from the road traffic demand database. The road traffic demand prediction apparatus described in 1.   The road traffic demand forecasting unit changes the size of the small area according to at least one of the global context and the personal context, and the size of the small area is determined based on a starting point, a destination, a POI (Point Of Interest), the length distribution for each standard of roads in the target area, and the road traffic demand extracted from the road traffic demand database. The road traffic demand prediction apparatus according to claim 9.   The road traffic demand prediction device according to claim 3, wherein the road traffic demand prediction unit divides the target area into a lattice shape to generate the small regions having a plurality of rectangular shapes.   The road traffic demand prediction device according to claim 3, wherein the road traffic demand prediction unit generates the plurality of small areas by dividing the target area based on a boundary defined by a government.   4. The road according to claim 3, wherein the road traffic demand prediction unit sets a path formed by connecting road links between one branch point and another branch point on the road as the small region. Traffic demand forecasting device.   The road traffic demand forecasting unit is extracted from the starting point, destination of personal movement, POI (Point Of Interest) distribution, length distribution for each road standard in the target area, and the road traffic demand database. The size of the small area is set such that a value for each small area of a variable related to at least one of the road traffic demands is equal to or less than a variable maximum value that is a preset value. The road traffic demand prediction apparatus according to claim 14.   The maximum value of the variable is set so that the total number of the small areas is equal to or less than a preset total number of small areas, and the change in the predicted value of the future road traffic demand due to the change in the maximum variable value is minimized. The road traffic demand prediction apparatus according to claim 18, wherein Road traffic demand data acquisition process for acquiring road traffic demand data related to road traffic demand from the road traffic demand database;
A global context acquisition step of acquiring a global context which is information related to an environment or a specific area common to a plurality of personnel;
Generated from personal context that is information related to the personal environment or a specific organization, and input predicted behavior data related to the driving behavior when the individual drives the vehicle, the road traffic demand data, the global context, and the A road traffic demand forecasting process for predicting road traffic demand using forecast behavior data;
A road traffic demand prediction method characterized by including:

Images