JP4426253B2 - Representative section travel time prediction device, representative section travel time prediction method and program - Google Patents

Description

  The present invention relates to a travel time prediction apparatus for predicting a travel time of a representative section of a vehicle between two points on an expressway.

Conventionally, when calculating the section travel (movement) time of a vehicle between two points on a road, the vehicle detection device detects the spot speed of the vehicle, and based on the spot speed of the vehicle detection device at a specific time. A representative travel time of a certain section in charge (hereinafter referred to as a representative section travel time) is estimated, and a plurality of representative section travel times included between the two points at the specific time are added up. Here, in the method of calculating the section travel time based on the vehicle spot speed, the vehicle detection device is congested in a certain section of the fixed section, but it is congested in the vicinity where the vehicle detection device is installed. Since the point speed is not constant in all the constant sections, the accuracy of the section travel time is poor. Therefore, there is a problem that the accuracy of the travel time of the representative section that is finally predicted is deteriorated.
In addition, a technique for predicting the time required to travel between toll booths on a highway (consent with representative section travel time) is disclosed (for example, see Non-Patent Document 1). In this technology, vehicles that enter and exit in nearby tollgate sections are used as samples, but there are few people who use high-speed reasoning at short distances, so it is difficult to secure a sufficient number of samples depending on the location. It is not possible to calculate the representative section travel time with high accuracy, and it cannot be separated from the section travel time calculated due to the congestion time at the tollgate exit, and is calculated as the section travel time of the main line Therefore, there is a problem that the required time between two points on the expressway cannot be accurately predicted.
Hideki Ueno and two others, “Comparison of required time prediction methods using toll gate data”, 1st ITS Symposium 2002 Proceedings, ITS-Japan, December 14, 2002, p. 515-520

  Conventionally, there is an expressway information providing service that notifies the expressway users of the representative section travel time between two points on the expressway, but the service is the representative section travel time in the traffic situation at the time of information provision Yes, it does not take into account future changes in traffic conditions. Therefore, there is a demand for an accurate prediction service in which the travel time of the representative section is taken into consideration in the future traffic situation. Therefore, the present invention provides a representative section travel time prediction device capable of accurately calculating the future representative section travel time between any two points on the expressway rather than the current representative section travel time using the vehicle spot speed. And a representative section travel time prediction method and a program thereof.

The present invention has been made to solve the above-described problem, and is a sample that holds a section travel time of a vehicle between two points on a road, the vehicle identification information, and the acquisition time of the vehicle identification information in association with each other. Of the section travel time of the vehicle in a predetermined time zone, the storage section and the section travel time not less than the upper limit value and less than the lower limit value and the section travel time outside the predetermined time zone are determined as invalid samples, and the rest are valid Sample selection means for determining a sample, and processing for calculating the representative section travel time in the predetermined time zone by averaging each section travel time of the effective sample is repeated at predetermined time intervals, and the representative section travel time a representative section travel time calculating means for calculating each representative section travel time between two points at a particular time of calculation of the specific time and the 2 locations at that particular time Downstream information creation that creates downstream information indicating the transition of the correspondence relationship between the accumulated traffic volume and the time at the downstream point based on the accumulated traffic volume obtained from the accumulated traffic volume detection device installed at the downstream point And the cumulative traffic volume and time at the upstream point of the two points based on the means and the cumulative traffic volume at the downstream point at the specific time and the time obtained by subtracting the representative section travel time at the specific time from the specific time. Upstream information creating means for creating upstream information indicating the transition of the correspondence relationship, the transition of the cumulative traffic volume and time indicated by the downstream information, and the transition of the cumulative traffic volume and time indicated by the upstream information When, by using the future trend calculation method, the future of the cumulative traffic volume and time correspondence between the transition and the downstream location of the correspondence between the future cumulative traffic volume and time in the upstream point Based on the transition of two future correspondences between the upstream point and the downstream point, and the time of the upstream point corresponding to the same cumulative traffic and the downstream point A representative section travel time prediction device comprising: representative section travel time prediction means for predicting a difference in time as a representative section travel time from the upstream point to the downstream point at a future time.

Further, the present invention is the above-described representative section travel time prediction device, wherein the correspondence transition prediction means is a correspondence relationship between a cumulative traffic volume and a time of a certain day in the past at each of the upstream point and the downstream point. And a method of determining the similarity between the transition of the current day's cumulative traffic volume and the transition of the correspondence relationship of the time as the future transition calculation method, and the future cumulative traffic volume on the current day of the upstream point and the downstream point It is characterized by predicting the transition of time correspondence.

The present invention, in a representative section travel time prediction system described above, the correspondence relation transition prediction means, on the basis of transition of the past correspondence relationship in the downstream point, respectively and the upstream point of the time series data a prediction method the It is characterized by using the future transition calculation method to predict the two future correspondences.

In the representative section travel time prediction apparatus according to the present invention, the correspondence transition prediction unit may determine the future of the two correspondences among the method for determining the similarity or the time series data prediction method. As a prediction method, the method is characterized in that the predicted future traffic volume and time are switched to a method with high accuracy .

According to the present invention, the representative section travel time prediction device described above includes a past information storage unit that stores information on a transition of a correspondence relationship between a cumulative traffic volume and a time in a certain period of the past in the past, and the past information. And updating means for updating information on the transition of the correspondence relationship between the cumulative traffic volume and the time recorded in the storage means for a certain day in the past in the past.

The present invention also provides sample storage means for associating and holding a section travel time of a vehicle between two points on a road, the vehicle identification information, and the acquisition time of the vehicle identification information, and the vehicle in a predetermined time zone. A sample selection means for determining a section travel time that is greater than or equal to an upper limit value and less than a lower limit value and a section travel time that is out of the predetermined time zone as invalid samples and the other as valid samples. In the representative section travel time prediction method in the representative section travel time prediction apparatus, the process of calculating the representative section travel time in the predetermined time zone by averaging each section travel time of the effective sample is a predetermined time. repeated at intervals, a representative section travel time calculating process of the calculating each representative section travel time between two points at a particular time calculated the representative section travel time, before Based on the specific time and the cumulative traffic obtained from the cumulative traffic detector installed at the downstream point of the two points at the specific time, the transition of the correspondence relationship between the cumulative traffic and the time at the downstream point Based on the downstream information creation process of creating downstream information, and the accumulated traffic volume of the downstream point at the specific time and the time obtained by subtracting the representative section travel time at the specific time from the specific time. and upstream information creation process of creating an upstream information showing changes in the correspondence between the accumulated traffic volume and time at a point upstream of the out, transition and correspondence between the accumulated traffic volume and time in which the downstream information indicating indicates said upstream information changes and correspondence between the accumulated traffic volume and time by using the future trend calculation method, the transition and the downstream location of the correspondence between the future cumulative traffic volume and time in the upstream point Based on the correspondence transition prediction process for predicting the future cumulative traffic volume and the transition of the correspondence relationship of the time, and the two future correspondence transitions of the upstream point and the downstream point. A representative section travel time prediction process for predicting a difference in time between the corresponding upstream point and the downstream point as a representative section travel time from the upstream point to the downstream point at a future time. Is the representative section travel time prediction method

The present invention also provides sample storage means for associating and holding a section travel time of a vehicle between two points on a road, the vehicle identification information, and the acquisition time of the vehicle identification information, and the vehicle in a predetermined time zone. A sample selection means for determining a section travel time that is greater than or equal to an upper limit value and less than a lower limit value and a section travel time that is out of the predetermined time zone as invalid samples and the other as valid samples. A program to be executed by the computer of the representative section travel time prediction apparatus , wherein the processing of calculating the representative section travel time in the predetermined time zone by averaging each section travel time of the effective sample is performed at a predetermined time interval. Again, the representative section travel time calculating process for calculating each representative section travel time between two points at a particular time calculated the representative section travel time and at Based on the specific time and the cumulative traffic obtained from the cumulative traffic detection device installed at the downstream point of the two points at the specific time, the correspondence between the cumulative traffic at the downstream point and the time Based on the downstream information creation process for creating downstream information indicating transition, the accumulated traffic volume at the downstream point at the specific time and the time obtained by subtracting the representative section travel time at the specific time from the specific time the transition of the upstream information creation processing, correspondence between the accumulated traffic volume and time in which the downstream information indicating that creates upstream information showing changes in the correspondence between the accumulated traffic volume and time at a point upstream of said upstream information changes and correspondence between the accumulated traffic volume and time indicated, using the future trend calculation method, transition and the downstream locations of a correspondence between a future cumulative traffic volume and time in the upstream point Based on the correspondence transition prediction process for predicting the future cumulative traffic volume and the transition of the correspondence relation in time and the transition of the two future correspondence relations of the upstream point and the downstream point. Causing the computer to execute a representative section travel time prediction process for predicting a difference between the time at the corresponding upstream point and the downstream point as a representative section travel time from the upstream point to the downstream point at a future time. It is a program.

According to the present invention, the representative section travel time between two points at a plurality of specific times is calculated. Further, the downstream information creating means creates downstream information indicating the transition of the correspondence relationship between the accumulated traffic volume and the time at the downstream point, and the upstream information creating means represents the upstream information indicating the transition of the correspondence relationship between the cumulative traffic volume and the time at the upstream point. Create Based on the downstream information and the upstream information, the correspondence transition prediction means determines the transition between the future cumulative traffic volume and the time at the upstream point and the transition between the future cumulative traffic and the time at the downstream point. In addition, the representative section travel time prediction means calculates the time of the upstream point and the time of the downstream point corresponding to the same cumulative traffic volume based on the transition of the two future correspondences between the upstream point and the downstream point. The difference is predicted as the representative section travel time from the upstream point to the downstream point at a future time.
From the transition of the cumulative relationship between the cumulative traffic volume at the downstream point and the time created based on the representative section travel time between the two points on the road and the transitional relationship between the cumulative traffic volume at the upstream point and the time, Since the representative section travel time at a certain time in the future is predicted, the representative section travel time is calculated based on the conventional vehicle spot speed and then the representative section travel time is predicted. The representative section travel time can be predicted. In addition, if the vehicle identification information receiver is installed at two arbitrary points, not only the prediction of the future representative section travel time for a certain section of the toll booth, but also the prediction of the representative section travel time for the other two points can do.

  Hereinafter, a representative section travel time prediction apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an outline of a representative section travel time prediction system of the embodiment. In this figure, reference numeral 1 denotes a representative section travel time prediction device. Reference numeral 2 denotes a vehicle. Reference numeral 3 denotes a highway, which is represented by a straight line. Reference numeral 4 denotes a unique ID receiving device (vehicle identification information receiving device). Reference numeral 5 denotes a cumulative traffic detection device that detects the cumulative traffic of the vehicle. In this representative section travel time prediction system, the unique ID receiving device 4 is provided at point A (upstream point) and point B (downstream point) on the expressway 3 and point C (target point) between the two points. Is installed, and the accumulated traffic detection device 5 is installed at the point B. When the vehicle 2 passes through the vicinity of the unique ID receiving device 4 at each of the points A, B, and C, the unique ID transmission device provided in the vehicle 2 receives the unique ID (vehicle identification) of the vehicle 2. Information) to the unique ID receiving device 4, and the unique ID receiving device 4 at each of the points A, B, and C predicts the travel time of the representative section via the communication network. Transfer to device 1. Further, the accumulated traffic detection device 5 transmits passage information indicating that the vehicle has passed at the point B to the representative section travel time calculation device 1. Then, the representative section travel time prediction device 1 calculates the section travel time from the point A to the point B of the vehicle 2 based on the received unique ID and the received time, and from the valid samples of the plurality of section travel times Calculate representative section travel time. Further, the representative section travel time calculation device 1 counts the cumulative traffic volume for each time at the B point from the passage information transmitted from the cumulative traffic volume detection device 5. Then, the representative section travel time calculation device 1 obtains information on the transition of the correspondence relationship between the accumulated traffic volume at the point A and the time based on the transition of the correspondence relationship between the calculated representative section travel time and the accumulated traffic amount at the point B and the time. Further, the representative section travel time from the A point to the B point at a future time is predicted from the transition of the correspondence relationship between the past accumulated traffic volume at the A point and the B point and the time.

  FIG. 2 is a diagram showing a functional block configuration of the representative section travel time prediction system of the embodiment. In the representative section travel time prediction device 1 in FIG. 2, reference numeral 11 denotes a receiving unit that receives the unique ID and the reception time from the unique ID receiving device 4 via the communication network. Further, 12 calculates a section travel time for each unique ID based on the received unique ID and its reception time, determines a valid sample and an invalid sample among a plurality of section travel times, and further represents a representative section travel time. It is a representative section travel time calculation part (representative section travel time calculation means) to calculate. Reference numeral 13 denotes a representative section travel time prediction unit (upstream information creation means, correspondence transition prediction means, representative section travel time prediction means) that predicts a future representative section travel time from point A to point B. Further, 14 indicates the maximum value and minimum value of the number of valid samples of the unique ID, each ratio for determining the upper limit value and the lower limit value of the section travel time as the valid sample, the unique ID reception table, the section travel time table, and the like. It is a memory | storage part to memorize | store. Reference numeral 15 denotes a cumulative traffic count unit (downstream) that counts the passing information transmitted from the cumulative traffic detector 5 and associates the time with the cumulative traffic and records it in the B-point cumulative traffic table (downstream information). Information creation means).

  Reference numeral 21 denotes a unique ID transmission device provided in the vehicle 2, a unique ID transmission processing unit 22 that transmits the unique ID to the unique ID reception device 4 by a DSRC (Narrow Area Dedicated Communication) wireless communication method, and a unique ID A storage unit 23 for storing the ID is provided. Further, in the unique ID receiving device 4, reference numeral 41 is a unique ID receiving unit that receives the unique ID transmitted from the unique ID transmitting device 21 of the vehicle 2 by the DSRC wireless communication method, and 42 is a unique ID and its unique ID. It is a unique ID transfer unit that transmits the reception time to the representative section travel time prediction device 1 via the communication network. The DSRC wireless communication method is a wireless communication method used in an automatic toll road payment system. In the cumulative traffic detection device 5, reference numeral 51 denotes a vehicle detection unit that detects a vehicle that has passed on the road by infrared rays, and reference numeral 52 represents passage information indicating that the vehicle detection unit 51 has detected the passage of the vehicle. It is a passage information transmission processing unit that transmits to the section travel time prediction device 1.

Next, the processing of the representative section travel time prediction system will be described with reference to FIG. FIG. 3 is a diagram showing a first processing flow of the representative section travel time prediction system.
First, when the vehicle 2 passes the unique ID receiving device 4 installed at the point A, the unique ID transmission processing unit 22 of the unique ID transmitting device 2 provided in the vehicle 2 is recorded in the storage unit 23. The unique ID is transmitted to the unique ID receiver 4 using the DSRC wireless communication system (step S1). Then, the unique ID receiver 41 of the unique ID receiver 4 receives the unique ID of the vehicle 2 and transmits the unique ID and the reception time of the unique ID to the unique ID transfer unit 42. Next, the unique ID transfer unit 42 transmits the unique ID received from the unique ID receiving unit 41 and the reception time to the representative section travel time prediction device 1 via the communication network (step S2). The representative section travel time prediction device 1 records the unique ID received from the unique ID receiver 4 and the reception time in the unique ID reception table (FIG. 4) in the storage unit 14 in association with each other (step S3). Then, each time the vehicle 2 passes through the points C and B, the unique ID receiving device 4 at the points C and B is unique to the representative section travel time prediction device 1 in the same manner as the above-described steps S1 to S3. The ID and the reception time are transmitted, and the unique ID and the reception time are separately associated with each other and recorded in the unique ID reception table of the storage unit 14 as information obtained by the unique ID receiving device 4 at the points B and C. Is done.

  The section travel time calculation unit 12 of the representative section travel time prediction apparatus 1 records the unique ID reception time at the points A and B for a specific unique ID in the unique ID reception table of the storage unit 14. It is confirmed whether or not the ID is also received by the unique ID receiver 4 at the point C (step S4). Here, in the unique ID reception table, when the reception time at the point C (the target point passage information) is not recorded even though the reception time at the point B is recorded, the vehicle 2 selects the point C. You have arrived from point A to point B without passing. That is, the vehicle 2 stops at the service area without passing through the point C. The representative section travel time prediction device 1 invalidates the unique ID of the vehicle 2 when the vehicle 2 stops at the service area (step S5). Then, the representative section travel time prediction device 1 validates the unique ID of the vehicle 2 passing through the point C on the expressway 3, and sets the unique ID between the distance from the point A to the point B and the points A and B. Based on each received reception time, a section travel time from point A to point B of the vehicle 2 is calculated (step S6). Then, the unique ID of the vehicle 2, the reception time at the point C of the unique ID, and the interval travel time of the vehicle 2 holding the unique ID are recorded in association in the zone travel time table (FIG. 5) of the storage unit 14 ( Step S7).

As a result of the processing from step S1 to step S7 described above, the information held in the section travel time table in which the unique ID, the reception time of the unique ID, and the section travel time are associated with each other increases with time. Here, the representative section travel time calculation unit 12 reads from the storage unit 14 each ratio for determining the upper limit value and the lower limit value of the section travel time as effective samples, and the maximum value and minimum value of the number of effective samples. Then, the representative section travel time calculation unit 12 determines the section travel time as an effective sample among the section travel times recorded in the section travel time table of the storage unit 14 under the following conditions.
(1) In order to keep the number of valid samples in a predetermined time zone below the maximum value, invalid samples are determined in order from the section travel time stored in the section travel time table corresponding to the old reception time (2) predetermined Among the section travel times associated with the acquisition time older than the oldest time of the time zone and held in the section travel time table, the number of valid samples is set as the valid samples in order from the latest section travel time. Maintain above the minimum value (3) Set the section travel time above the upper limit and the section travel time below the lower limit as invalid samples.
Here, the ratio for determining the upper limit value and the lower limit value of the segment travel time is a value obtained from the representative segment travel time in the immediately preceding predetermined time zone. For example, the upper limit value is the representative segment travel time in the immediately preceding predetermined time zone. The 150% value and the lower limit value are 75% of the representative section travel time in the immediately preceding predetermined time zone.

  For example, it is assumed that the predetermined time zone is a time zone from the current time to 5 minutes before. Then, the representative section travel time calculation unit 12 first extracts the section travel time stored in the section travel time table in association with the reception time of the time zone from the current time to five minutes ago (a), If the number is larger than the maximum value n1, among the extracted section travel times, the maximum value n1 is set as an invalid sample in order from the section travel times associated with the old reception time and recorded in the section travel time table. Reduce until (b). In addition, the representative section travel time calculation unit 12 is stored in the section travel time table in association with the time before 5 minutes from the current time when the number of section travel times extracted is less than the minimum value n2. The section travel time is sequentially extracted, and the effective sample is set to the minimum value n2 or more (c). And the representative section travel time calculation part 12 is more than the value (upper limit value) of 150% of the last representative section travel time among each section travel time obtained by the processing of (a), (b), (c). A long section travel time and a section travel time that is less than 75% of the last representative section travel time (lower limit) are used as invalid samples. (D). The representative section travel time calculation unit 12 determines n1 ≧ the number of effective samples ≧ n2 effective samples through the processes (a), (b), (c), and (d) (step S8).

  Next, the representative section travel time calculation unit 12 calculates an average value of the section travel times that are valid samples, and sets the calculated average value as the representative section travel time (step S9). Thereby, the representative section travel time prediction device 1 outputs the representative section travel time. Then, the representative section travel time calculation unit 12 performs the processes in steps S8 and S9 described above at a specific time every 5 minutes, for example. Thereby, the representative section travel time from point A to point B at a specific time every five minutes can be calculated. In step S1 to step S8, the section travel time is calculated using the unique ID of the vehicle 2 acquired using the DSRC wireless communication method. However, the present invention is not limited to this method. The travel time of the representative section may be calculated from the spot speed of the vehicle 2 measured by the sensor. Then, the representative section travel time calculation unit 12 associates the calculated representative section travel time with the time when the representative section travel time is calculated, and records it in the representative section travel time table (FIG. 6) held in the storage unit 13. Then, a section travel time table is created (step S10).

  In this embodiment, the process when the point C is on the three highways is described. However, the point C is between the entrance lamp and the exit lamp of the service area, and the unique ID receiving device is located at the point C. 4 may be installed. In this case, when the unique ID is received by the unique ID receiving device 4, it indicates that the vehicle 2 has entered the service area. Therefore, in step S5 described above, the reception time at point C is recorded in the unique ID reception table. If it is, the unique ID corresponding to the reception time is set as an invalid sample.

Next, a process for predicting a future representative section travel time will be described with reference to FIG. FIG. 7 is a diagram showing a second processing flow of the representative section travel time prediction system.
First, in parallel with the processing from step S1 to step S10 described above, the vehicle detection unit 51 of the cumulative traffic detection device 5 detects a vehicle that has passed through the point B. The passing information is transmitted to the representative section travel time prediction device 1 via the communication network (step S11). Then, the cumulative traffic counting unit 15 of the representative section travel time prediction device 1 counts the number of passing information received from the cumulative traffic detection device 5 and, for example, the total count of the day at each specific time every five minutes. The number (hereinafter referred to as cumulative traffic volume) is recorded in the B-point cumulative traffic volume table held in the storage unit 14, and the B-point cumulative traffic volume table (FIG. 8) up to the present is created (step S12).

  Then, the representative section travel time prediction unit 13 uses the information recorded in the B point cumulative traffic volume table and the information recorded in the representative section travel time table to perform the following process on the A point cumulative traffic volume table (upstream information). ) Is created (step S13). First, the representative section travel time prediction unit 13 reads a representative section travel time at a certain time (hereinafter referred to as time A) recorded in the representative section travel time table. Then, the storage unit 14 holds the time B obtained by subtracting the read representative section travel time from the time A and the accumulated traffic volume associated with the time A and recorded in the B-point accumulated traffic volume table. Record in the A-point cumulative traffic table. This process is repeated. Thereby, a point A cumulative traffic volume table is created. FIG. 9 is a diagram showing an image of the process for creating the A-point cumulative traffic volume table. The solid line 1 represents the transition of the accumulated traffic volume at the point B (created from the B point accumulated traffic volume table). Then, at one point on the solid line 1, the point moved in parallel up to the previous time by the representative section travel time of the time indicated by the point is plotted. By repeating this, the solid line 2 connecting the plots with lines is the transition of the accumulated traffic volume at the point A. In the solid line 2 representing the transition of the accumulated traffic volume at the point A, a table in which the time at each plotted point is associated with the accumulated traffic volume is the A point accumulated traffic volume table. This figure also shows that the accumulated traffic volume at point A is equal to the accumulated traffic volume at point A at time A at time B, which is the time when the representative section travel time at time A is subtracted from time A. Yes.

  With the above processing, the B point accumulated traffic volume table and the A point accumulated traffic volume table up to the current time are created. Here, since the representative section travel time prediction device 1 holds the past information in the B point accumulated traffic volume table and the A point accumulated traffic volume table, the past certain day of each of the A point and the B point is constant. A graph of the transition of the correspondence relationship between the accumulated traffic volume and the time in the time zone can be created every day. Then, the representative section travel time prediction unit 13 for each of the points A and B, a plurality of graphs of the transition of the correspondence relationship between the accumulated traffic volume and the time in a certain day in the past, the accumulated traffic volume on the current day, A graph of the transition of time correspondence is created (step S14). Then, the representative section travel time prediction unit 13 is similar to the graph of the transition of the correspondence relationship between the accumulated traffic volume on the day and the time for each of the points A and B, and is accumulated in a certain time period of a certain past day. A graph of the transition of correspondence between traffic volume and time is detected using a wavelet analysis technique (step S15), and the transition of the correspondence between cumulative traffic volume and time during a certain period of time in the past is detected. Is predicted as a graph of the transition of the correspondence relationship between the future accumulated traffic volume on the day and the time at each of the points A and B. In other words, the correspondence relationship in the certain time zone of the past in the past and the correspondence relationship of the day are pattern-matched to detect the correspondence relationship in the past in the past that is similar to the correspondence relationship of the day. This process is performed for each of the points A and B. Since the wavelet analysis technique is a known technique, the details are omitted. FIG. 10 is a diagram showing an outline of pattern matching processing.

  Here, instead of using the graph of the transition of the correspondence relationship between the accumulated traffic volume and the time during a certain day in the past, the autoregressive model Using a method (time series data prediction method), a graph of the transition of the correspondence relationship between the future accumulated traffic volume and the time on that day may be created. Then, the representative section travel time prediction unit 13 corresponds to the same cumulative traffic volume in the correspondence relationship between the cumulative traffic volume and the time at the future point A and the correspondence relationship between the cumulative traffic volume and the time at the future point B. It is predicted that the difference between the times is the representative section travel time at the future time (step S16).

  Then, the representative section travel time prediction unit 13 predicts the correspondence between the future accumulated traffic volume and the time based on the pattern matching described above, and predicts the correspondence relationship between the future accumulated traffic volume and the time based on the autoregressive model method. In such a case, the prediction of the correspondence relationship between the future accumulated traffic volume and the time with high accuracy may be selected. In addition, the representative section travel time prediction device may correct the estimated future traffic volume estimated in the past and the actual section travel time if there is a large error.

  Each of the above devices has a computer system inside. The process described above is stored in a computer-readable recording medium in the form of a program, and the above process is performed by the computer reading and executing this program. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

It is a figure which shows the outline of the representative section travel time prediction system by embodiment of this invention. It is a figure which shows the structure of the functional block of the representative section travel time prediction system by the embodiment. It is a figure which shows the 1st processing flow of the representative area travel time prediction system by the embodiment. It is a figure which shows the data structure of the unique ID reception table by the embodiment. It is a figure which shows the data structure of the section travel time table by the embodiment. It is a figure which shows the data structure of the representative area travel time table by the embodiment. It is a figure which shows the 2nd processing flow of the representative area travel time prediction system by the embodiment. It is a figure which shows the data structure of the B point cumulative traffic volume table by the embodiment. It is a figure which shows the image of the preparation process of A point accumulation traffic volume table by the embodiment. It is a figure which shows the process outline | summary of the pattern matching by the same embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Representative section travel time prediction apparatus 2 ... Vehicle 3 ... Highway 4 ... Unique ID receiver 5 ... Cumulative traffic detector 11 ... Receiver 12 ... Representative section Travel time calculation unit 13 ... representative section travel time prediction unit 14, 23 ... storage unit 15 ... cumulative traffic count unit 21 ... unique ID transmission device 22 ... unique ID transmission processing unit 41 .. Unique ID receiver 42 ... Unique ID transfer unit 51 ... Vehicle detection unit 52 ... Passing information transmission processing unit

Claims (7)

Sample storage means for holding a section travel time of a vehicle between two points on the road, the vehicle identification information, and the acquisition time of the vehicle identification information in association with each other;
Of the section travel times of the vehicle in a predetermined time zone, a section travel time not less than the upper limit value and less than the lower limit value and a section travel time outside the predetermined time zone are determined as invalid samples, and the rest are determined as valid samples. Sample sorting means;
The process of calculating the representative section travel time in the predetermined time zone by averaging each section travel time of the valid sample is repeated at a predetermined time interval, and between the two points at the specific time at which the representative section travel time is calculated Representative section travel time calculating means for calculating each representative section travel time,
Based on the specific time and the cumulative traffic obtained from the cumulative traffic detection device installed at the downstream point of the two points at the specific time, the correspondence between the cumulative traffic at the downstream point and the time Downstream information creating means for creating downstream information indicating the transition;
Based on the cumulative traffic volume at the downstream point at the specific time and the time obtained by subtracting the representative section travel time at the specific time from the specific time, the correspondence between the cumulative traffic volume at the upstream point of the two points and the time Upstream information creation means for creating upstream information indicating the transition of
Using the transition of the cumulative traffic volume and time indicated by the downstream information, the transition of the cumulative traffic volume and time indicated by the upstream information, and the future trend calculation method, the future cumulative at the upstream point Correspondence transition prediction means for predicting transition of correspondence between traffic volume and time and transition of correspondence relation between future cumulative traffic volume and time at the downstream point;
Based on the transition of two future correspondences between the upstream point and the downstream point, the difference between the time at the upstream point and the time at the downstream point corresponding to the same cumulative traffic volume is calculated as the upstream point at a future time. Representative section travel time predicting means for predicting the representative section travel time from to the downstream point,
A representative section travel time prediction apparatus comprising: The correspondence relationship transition predicting means includes a transition of a correspondence relationship between a cumulative traffic volume and a time of a certain day in the past at each of the upstream point and the downstream point, and a transition relationship between a cumulative traffic amount and a time of the day. The method of determining the similarity between the upstream point and the downstream point is used as the future transition calculation method to predict the transition of the correspondence relationship between the future accumulated traffic volume and the time on the day. Item 2. The representative section travel time prediction device according to Item 1. The correspondence relationship transition prediction means uses the time-series data prediction method as the future transition calculation method based on the transition of past correspondence relationships at the upstream point and the downstream point, respectively, to determine the two future correspondence relationships. The representative section travel time prediction device according to claim 1, wherein the prediction is performed. The correspondence relationship transition prediction means includes the predicted future traffic volume as a method for determining the similarity or a method for predicting the future of the two correspondence relationships among the method for predicting the time series data. 4. The representative section travel time prediction device according to claim 2, wherein the method is switched to a method with high time accuracy. Past information storage means for storing information on the transition of the correspondence relationship between the cumulative traffic volume and the time during a certain period of time in the past;
Updating means for updating the information on the transition of the correspondence relationship between the accumulated traffic volume and the time of a certain day in the past recorded in the past information storage means;
Representative section travel time prediction device according to claim 2 to claim 4, characterized in that it comprises a. Sample storage means for holding a section travel time of a vehicle between two points on the road, the vehicle identification information, and the acquisition time of the vehicle identification information in association with each other;
Of the section travel times of the vehicle in a predetermined time zone, a section travel time not less than the upper limit value and less than the lower limit value and a section travel time outside the predetermined time zone are determined as invalid samples, and the rest are determined as valid samples. A representative section travel time prediction method in a representative section travel time prediction device comprising a sample selection means,
The process of calculating the representative section travel time in the predetermined time zone by averaging each section travel time of the valid sample is repeated at a predetermined time interval, and between the two points at the specific time at which the representative section travel time is calculated Representative section travel time calculation process to calculate each representative section travel time,
Based on the specific time and the cumulative traffic obtained from the cumulative traffic detection device installed at the downstream point of the two points at the specific time, the correspondence between the cumulative traffic at the downstream point and the time Downstream information creation process to create downstream information indicating the transition,
Based on the cumulative traffic volume at the downstream point at the specific time and the time obtained by subtracting the representative section travel time at the specific time from the specific time, the correspondence between the cumulative traffic volume at the upstream point of the two points and the time Upstream information creation process to create upstream information showing the transition of
Using the transition of the cumulative traffic volume and time indicated by the downstream information, the transition of the cumulative traffic volume and time indicated by the upstream information, and the future trend calculation method, the future cumulative at the upstream point Correspondence transition prediction process for predicting transition of correspondence between traffic volume and time and transition of correspondence relation between future cumulative traffic volume and time at the downstream point;
Based on the transition of two future correspondences between the upstream point and the downstream point, the difference between the time at the upstream point and the time at the downstream point corresponding to the same cumulative traffic volume is calculated as the upstream point at a future time. A representative section travel time prediction process for predicting a representative section travel time from to the downstream point,
A representative section travel time prediction method characterized by comprising: Sample storage means for holding a section travel time of a vehicle between two points on the road, the vehicle identification information, and the acquisition time of the vehicle identification information in association with each other;
Of the section travel times of the vehicle in a predetermined time zone, a section travel time not less than the upper limit value and less than the lower limit value and a section travel time outside the predetermined time zone are determined as invalid samples, and the rest are determined as valid samples. A program to be executed by a computer of a representative section travel time prediction device comprising a sample selection means,
The process of calculating the representative section travel time in the predetermined time zone by averaging each section travel time of the valid sample is repeated at a predetermined time interval, and between the two points at the specific time at which the representative section travel time is calculated Representative section travel time calculation processing for calculating the representative section travel time of each,
Based on the specific time and the cumulative traffic obtained from the cumulative traffic detection device installed at the downstream point of the two points at the specific time, the correspondence between the cumulative traffic at the downstream point and the time Downstream information creation processing for creating downstream information indicating transition;
Based on the cumulative traffic volume at the downstream point at the specific time and the time obtained by subtracting the representative section travel time at the specific time from the specific time, the correspondence between the cumulative traffic volume at the upstream point of the two points and the time Upstream information creation processing to create upstream information indicating the transition of
Using the transition of the cumulative traffic volume and time indicated by the downstream information, the transition of the cumulative traffic volume and time indicated by the upstream information, and the future trend calculation method, the future cumulative at the upstream point Correspondence transition prediction processing for predicting transition of correspondence between traffic volume and time and transition of correspondence relation between future accumulated traffic volume and time at the downstream point;
Based on the transition of two future correspondences between the upstream point and the downstream point, the difference between the time at the upstream point and the time at the downstream point corresponding to the same cumulative traffic volume is calculated as the upstream point at a future time. Representative section travel time prediction processing for predicting the representative section travel time from to the downstream point,
A program that causes a computer to execute.

Images