JP2007249845A - Traffic situation estimation program and traffic situation estimating method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To simply provide a high comprehensiveness traffic situation infer at a low cost. <P>SOLUTION: A detecting part 2402 detects a link, in which the traveling speed in the link is unclear among a plurality of links constituting a road network, and a peripheral link extracting part 2403 extracts a peripheral link, having a prescribed correlation with respect to the detected link among real peripheral links existing around the detected link. A generating part 2404 acquires the present and past actual result information, by associating traveling speeds in the extracted peripheral link in each one and the same date and the hour, and a similarity calculating part 2405 calculates the similarity between the acquired present actual result information and past actual result information in each date and hour. Thereafter, an inferred value calculating part 2406 calculates the estimated value of the traveling speed in the detected link, on the basis of the calculated similarity. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a traffic situation estimation program and a traffic situation estimation method for estimating a traffic situation.

  Conventionally, various attempts have been made to estimate traffic conditions such as the location and scale of traffic jams. For example, based on the counter value of a traffic counter embedded in the road, the vehicle moving speed at a specific point and the time required for moving a link including the specific point are calculated. There is a technology that estimates traffic conditions such as scale.

  Further, in Patent Document 1 below, on the basis of traffic jam correlation data with other links for a predetermined link created based on past road traffic information for each link and current road traffic information for each link, The congestion occurrence probability of each link is calculated based on the congestion pattern data for a given link created based on the past road traffic information for each link and the current road traffic information for each link. A technique is disclosed in which traffic jam prediction data for each link is calculated using the calculated traffic jam persistence probability and traffic jam occurrence probability.

JP 2004-272408 A

  By the way, the traffic counter is embedded only in some roads such as a highway with a high traffic volume. For this reason, as described above, there is a problem that the method of estimating the traffic situation using the traffic counter can be applied only to a part of roads such as a main road having a large traffic volume.

  On the other hand, in order to increase the number of roads where traffic conditions can be estimated using traffic counters, the number of traffic counters embedded in the roads must be increased, and there was a problem that it would be costly to develop new infrastructure . In addition, even if a new traffic counter is embedded, the traffic situation cannot be estimated unless there is a passing vehicle. Therefore, there is a problem that only the traffic situation on a road with a large amount of traffic can be estimated.

  An object of the present invention is to provide a traffic situation estimation program and a traffic situation estimation method capable of easily and cost-effectively estimating a traffic situation with high completeness in order to eliminate the above-described problems caused by the prior art. And

  In order to solve the above-described problems and achieve the object, a traffic situation estimation program, a recording medium recording the program, a traffic situation estimation apparatus, and a traffic situation estimation method according to the present invention include a plurality of links constituting a road network. A link whose current moving speed is unknown is detected from among the actual peripheral links existing around the detected link (hereinafter referred to as “detected link”), and a predetermined correlation with the detected link is detected. By extracting related peripheral links and associating the movement speeds of the extracted peripheral links for each same date and time, current and past performance information is generated, and the generated current performance information and past performance information are generated. Is calculated for each date and time, and an estimated value of the moving speed of the detection link is calculated based on the calculated similarity. To.

  According to the present invention, since it is possible to estimate the unknown moving speed on the detection link without newly providing physical infrastructure facilities, it is possible to easily and inexpensively realize estimation of traffic conditions with high completeness. it can.

  In the above invention, the peripheral link may be extracted in accordance with a link speed of the detected link and a correlation coefficient established between each of the actual peripheral links.

  According to the present invention, the reliability of the estimated moving speed is ensured by using the track information of the peripheral links having a clear correlation with the detected link, and traffic conditions with high completeness can be estimated easily and easily. It can be realized at low cost.

  Further, in the above invention, the current date time and a past date time having substantially the same traffic situation as the current date time are specified, and the moving speed in the peripheral link is specified for each specified date time. It is good also as producing | generating the said present and past performance information by making it link.

  According to the present invention, by adding the date and time to be estimated in addition to the spatial correlation, it is possible to ensure the reliability of the estimated moving speed and to easily estimate the traffic situation with high completeness. Can be realized at a cost.

  Moreover, in the said invention, it is good also as acquiring the said performance information of the same number of months as the number of months to which the specified date time belongs.

  According to this invention, in addition to the spatial correlation, the reliability of the estimated moving speed is ensured by taking into account the season to be estimated, and the estimation of traffic conditions with high completeness can be easily and lowly performed. Can be realized at a cost.

  Moreover, in the said invention, it is good also as acquiring the said performance information of the same week number as the week number to which the specified date time belongs.

  According to the present invention, the reliability of the estimated moving speed is ensured by taking into account the number of weeks to be estimated in addition to the spatial correlation, and the estimation of traffic conditions with high completeness can be easily and lowly performed. Can be realized at a cost.

  Moreover, in the said invention, it is good also as acquiring the said performance information according to whether the specified date time is the day of the week and holiday which belong in a 7th calendar.

  According to this invention, in addition to the spatial correlation, the reliability of the estimated moving speed is ensured by taking into account the day of the week and public holidays to be estimated, and the estimation of traffic conditions with high completeness can be performed easily and easily. It can be realized at low cost.

  Moreover, in the said invention, it is good also as acquiring the said performance information according to the time in the specified date time.

  According to the present invention, the reliability of the estimated moving speed is ensured by taking into account the time to be estimated in addition to the spatial correlation, and the estimation of the traffic situation with high completeness is simple and low-cost. Can be realized.

  Further, in the above invention, when there is a link whose movement speed is unknown among the peripheral links, the past movement speed in the unknown link is set as a random variable in a discrete distribution, and each of the random variables and the random variables Based on the occurrence probability, the expected value of the moving speed in the unknown link is calculated, and the current and past results are obtained by associating the moving speed in the peripheral links including the calculated expected value for each same date and time. Information may be generated.

  According to the present invention, it is possible to estimate the unknown moving speed in the detected link even when there is a surrounding link whose moving speed is unknown, so that it is possible to easily and inexpensively estimate the traffic situation with high completeness. Can do.

  Further, in the above invention, the similarity expressed by the Euclidean distance between the detected link and each of the peripheral links in the Euclidean space having the same number of dimensions as the number of the peripheral links may be calculated for each date and time. Good.

  According to the present invention, by using track record information having a clear similarity, it is possible to secure the reliability of the estimated moving speed and to easily estimate the traffic situation with high completeness at low cost. .

  Moreover, in the said invention, it is good also as calculating the moving speed of the same date time as the performance information with the highest calculated similarity degree as the moving speed of the said detection link.

  According to the present invention, by suppressing an increase in processing load in the estimation of the moving speed, it is possible to easily and inexpensively realize estimation of traffic conditions with high completeness.

  Moreover, in the said invention, it is good also as calculating the moving speed of the said detection link the average value of the moving speed of the same date time as the predetermined number of performance information selected in order with high calculated similarity.

  According to this invention, the reliability of the estimated moving speed can be ensured and the estimation of the traffic situation with high completeness can be realized easily and at low cost based on a plurality of pieces of record information.

  In the above invention, the calculation result obtained by multiple regression analysis using the movement speed of the detection link as an explanatory variable and the movement speed of a peripheral link as an explanatory variable may be calculated as the movement speed of the detection link. .

  According to the present invention, since the moving speed can be estimated even when the moving speeds of the detection link and the surrounding links take discrete values, it is possible to easily and inexpensively realize the estimation of the traffic situation with high completeness. it can.

  According to the traffic situation estimation program, the recording medium on which the program is recorded, the traffic situation estimation apparatus, and the traffic situation estimation method according to the present invention, it is possible to easily and inexpensively estimate the traffic situation. There is an effect.

  Exemplary embodiments of a traffic situation estimation program, a recording medium recording the program, a traffic situation estimation apparatus, and a traffic situation estimation method according to the present invention will be explained below in detail with reference to the accompanying drawings.

(System configuration of traffic situation estimation system)
First, a system configuration of a traffic situation estimation system according to an embodiment of the present invention will be described. FIG. 1 is a system configuration diagram of a traffic situation estimation system according to an embodiment of the present invention. In FIG. 1, a traffic situation estimation system 100 collects road traffic information and estimates a traffic situation based on the collected road traffic information.

  Here, the “road traffic information” is information that can specify the time required for the vehicle to move on an arbitrary section on the road and the moving speed of the vehicle at an arbitrary point. Specifically, the “road traffic information” is, for example, a point where traffic is expected to travel at a certain speed or less, such as a traffic congestion point on the road or a traffic restriction point due to work such as construction on or around the road. And information indicating the distance that traffic continues at a speed below a certain level.

  In the traffic situation estimation system 100, a traffic situation estimation apparatus 101 and a client terminal 102 are connected to a network 110 such as a LAN, WAN, or the Internet, and can communicate with each other. The traffic situation estimation apparatus 101 is a computer apparatus that calculates the link speed of each link using data stored in various tables in the database 103.

  Here, the “link” is a line segment that configures a road network that represents a road in map data that represents roads throughout Japan, and that connects two nodes that represent arbitrary points in the map data. The “link speed” is information indicating the moving speed when moving the above-described link.

  The “link speed” can be obtained, for example, by measuring the speed of a vehicle that has actually passed through an arbitrary point on the link using a traffic counter embedded in the road. Further, the “link speed” can be obtained by, for example, dividing the link distance by the time required to pass through the link.

  The time required for passing through the link is the time represented by the difference between the times when an arbitrary vehicle passes through both ends of the link. Here, the time of passing through both ends of the link is not limited to the time of actually passing through both ends of the link. For example, a plurality of current position information of an arbitrary vehicle is acquired at different times, and the time estimated to have passed through both ends of the link based on the acquired current position information and the time at which the current position information was acquired May be obtained by calculation.

  The client terminal 102 is, for example, a computer device installed in a vehicle management organization that collects information such as the current location or movement status of a specific vehicle such as a bus or taxi, and manages the vehicle using the collected information. is there. The client terminal 102 transmits information collected from a specific vehicle to the traffic situation estimation apparatus 101. The client terminal 102 may be installed in a management organization that collects data measured by a traffic counter embedded in the road.

(Hardware configuration of computer device)
Next, the hardware configuration of the computer apparatus (traffic situation estimation apparatus 101 or client terminal 102) shown in FIG. 1 will be described. FIG. 2 is a block diagram showing a hardware configuration of the computer apparatus shown in FIG. In FIG. 2, the computer apparatus includes a CPU 201, a ROM 202, a RAM 203, an HDD (hard disk drive) 204, an HD (hard disk) 205, an FDD (flexible disk drive) 206, and an example of a removable recording medium. An FD (flexible disk) 207, a display 208, an I / F (interface) 209, a keyboard 210, a mouse 211, a scanner 212, and a printer 213 are provided. Each component 201 to 213 is connected by a bus 200.

  Here, the CPU 201 controls the entire computer apparatus. The ROM 202 stores a program such as a boot program. The RAM 203 is used as a work area for the CPU 201. The HDD 204 controls data read / write with respect to the HD 205 according to the control of the CPU 201. The HD 205 stores data written under the control of the HDD 204.

  The FDD 206 controls reading / writing of data with respect to the FD 207 according to the control of the CPU 201. The FD 207 stores data written under the control of the FDD 206 or causes the computer device to read data stored in the FD 207.

  In addition to the FD 207, the removable recording medium may be a CD-ROM (CD-R, CD-RW), MO, DVD (Digital Versatile Disk), memory card, or the like. The display 208 displays data such as a document, an image, and function information as well as a cursor, an icon, or a tool box. As the display 208, for example, a CRT, a TFT liquid crystal display, a plasma display, or the like can be adopted.

  The I / F 209 is connected to a network 110 such as the Internet through a communication line, and is connected to other devices via the network 110. The I / F 209 controls an internal interface with the network 110 and controls input / output of data from an external device. For example, a modem or a LAN adapter may be employed as the I / F 209.

  The keyboard 210 includes keys for inputting characters, numbers, various instructions, and the like, and inputs data. Moreover, a touch panel type input pad or a numeric keypad may be used. The mouse 211 performs cursor movement, range selection, window movement, size change, and the like. A trackball or a joystick may be used as long as they have the same function as a pointing device.

  The scanner 212 optically reads an image and takes in the image data into the computer apparatus. The scanner 212 may have an OCR function. The printer 213 prints image data and document data. As the printer 213, for example, a laser printer or an ink jet printer can be employed.

(Link relationship)
Next, the link relationship will be described. FIG. 3 is an explanatory diagram (part 1) for explaining the relationship between links. In FIG. 3, the relationship of the links constituting an arbitrary area 310 on the map 300 will be described. The road included in the district 310 is configured by links L100 to L112.

  In FIG. 3, a link represented by reference sign L100 represents a link for which a link speed is to be calculated (hereinafter referred to as “target link”). The target link L100 is a link whose link speed is unknown in the database 103 of the traffic situation estimation apparatus 101.

  In FIG. 3, links represented by reference numerals L101 to L112 represent links that exist around the target link L100 on the actual map (hereinafter referred to as “real peripheral links”). The actual peripheral link is a link that is determined to have correlation with the target link L100 based on the database 103 of the traffic situation estimation apparatus 101.

  In FIG. 3, in addition to the links L102, L104, L105, and L108 to L110 connected to the target link L100, the actual peripheral links are links L106 and L107 that are parallel to the target link L100 and links that are connected to the links L106 and L107. L101, L103, L111, and L112.

  The actual peripheral links L101 to L112 are determined in advance by calculating the correlation between the links by a predetermined batch process for all the links constituting the road network. The correlation between links can be calculated based on the correlation coefficient between a certain link and all other links.

  In addition, the actual peripheral links L101 to L112 are correlated with the target link L100 for each link existing around the target link L100 from the positional relationship on the map of the link by the administrator of the traffic situation estimation system 100, for example. It can be arbitrarily set by determining whether or not there is.

(Table in database 103)
Next, a database and a table stored in the database 103 shown in FIG. 1 will be described. The database 103 stores a link information performance database (see FIG. 4). Below, a link information performance database is demonstrated with reference to FIG.

  FIG. 4 is an explanatory diagram showing a link information performance database. The link information performance database (hereinafter referred to as “DB”) 400 indicates the link speeds of all the links that are the target of link speed monitoring, including the target link L100 and the actual peripheral links L101 to L112, for each “date time”. It is a database stored in In the link information performance DB 400, the link speeds from the time when acquisition of the “link speed” is started to the present are accumulated and stored.

  As shown in FIG. 4, the link information performance DB 400 stores “link speeds” of all links to be monitored for link speeds for each “date time”. Here, the “date time” is information indicating each time when the link speed of each link is acquired. The “date time” is set for each predetermined unit time. The record information is realized by a set of link speeds associated with each other at the same date and time.

  In this embodiment, the “date time” is set in units of 5 minutes from the hour. For this reason, the link speed is acquired 12 times per hour. From FIG. 4, for example, the link speed at the date time t3 of the link Ln is vn. 3 is understood.

  In FIG. 4, for simplicity of explanation, the link information performance DB 400 in which one “link speed” is stored for one “link” is shown. There may be a plurality of link speeds corresponding to.

  For example, when the link connecting any node A and node B is a link that constitutes a road that can be face-to-face, the link information performance DB 400 includes a link speed when moving from node A to node B, Two types of link speed when moving from node B to node A are stored.

  Further, when the link connecting any node A and node B is a link constituting a road with one-way restrictions, either one of node A to node B or node B to node A is stored in the link information record DB 400. Stores the link speed.

  Further, when the link connecting any node A and node B is a link that constitutes a road that is restricted to one-way only in a specific time zone, the link information result DB 400 includes one or more according to the date and time. Two link speeds are stored.

  The information stored in the link information record DB 400 includes an actual peripheral link record table (see FIG. 6), a peripheral link table (see FIG. 9), a target record table (see FIG. 11), and a similarity calculation result table (see FIG. 15). It is used for the creation of a presently very similar achievement set (see FIG. 16). Hereinafter, in this embodiment, the case where the link L100 is the target link in the links L100 to L112 configuring the model example illustrated in FIG. 3 will be described as an example.

  FIG. 5 is an explanatory diagram (part 2) for explaining the relationship between links. FIG. 5 shows a moving direction that is a target of link speed estimation in the target link L100. As shown in FIG. 5, the moving direction that is the target of link speed estimation in the target link L100 is the direction of moving from the node N1 to the node N2.

  In this case, among the link speeds of the actual peripheral links L102, L104, and L105, the link speed related to the estimation of the link speed in the target link L100 is the link speed in the moving direction flowing into the node N1. Of the link speeds of the actual peripheral links L108, L109, and L110, the link speed related to the estimation of the link speed in the target link L100 is the link speed in the moving direction flowing out from the node N2.

  Of the link speeds of the actual peripheral links L101, L103, L106, L107, L111, and L112, the link speed related to the estimation of the link speed in the target link L100 is the link speed in the same movement direction as the target link L100.

  In the model example shown in FIG. 5, the target link L100 and the actual peripheral links L101 to L112 intersect in parallel or orthogonal directions for the sake of simplification. However, the model is not limited to this on an actual road. Absent. The direction is appropriately selected according to the inflow / outflow directions with respect to the nodes N1 and N2 at both ends of the target link L100, the distance to the target link L100, and the like.

  Further, the actual peripheral links L101 to L112 are not necessarily limited to surrounding the target link L100. For example, there may be no actual peripheral link parallel to the target link L100, or there may be only one actual peripheral link flowing into / out of the target link L100.

  Next, the actual peripheral link record table will be described. FIG. 6 is an explanatory diagram of an actual peripheral link performance table. As shown in FIG. 6, the actual peripheral link performance table 600 is a data table that stores the link speed of the target link L100 and the link speeds of the actual peripheral links L101 to L112 for each date and time.

  Next, the correlation coefficient table will be described. FIG. 7 is an explanatory diagram (part 1) of a correlation coefficient table. The correlation coefficient table 700 is a data table that stores the correlation coefficient between the link speed of the target link L100 and the link speeds of the actual peripheral links L101 to L112 for each of the actual peripheral links L101 to L112. The correlation coefficient table 700 is appropriately created in the database 103.

  Here, the “correlation coefficient” is a coefficient represented by the covariance of both variables and the standard deviation of each variable for two variables. When the correlation coefficient is r and the two variables are x and y, respectively, the correlation coefficient r is expressed by the following equation (1).

  In the above equation (1), X is the average value of the variable x, and Y is the average value of the variable y. In equation (1), x is the link speed of the target link L100 at a certain date and time, and y is the link speed of any of the actual peripheral links L101 to L112 at the same date and time.

  FIG. 8 is an explanatory diagram (part 2) of the correlation coefficient table. The correlation coefficient table 800 is a data table that stores the data in the correlation coefficient table 700 shown in FIG. 7 in a state where the data is rearranged in descending order of the correlation coefficient. The correlation coefficient table 800 is appropriately created in the database 103 based on the correlation coefficient table 700.

  The correlation coefficient calculated by the above equation (1) can determine the strength of the correlation by setting in advance an absolute value as a reference for interpreting the calculated correlation coefficient. it can.

表１には、相関係数の絶対値およびその解釈内容の関係が示されている。表１では、算出された相関係数が、４段階に分類されている。算出された相関係数が「０．０〜０．２」の範囲に含まれる場合、対象リンクＬ１００のリンク速度と実周辺リンクＬ１０１〜Ｌ１１２のリンク速度との間には「ほとんど相関関係がない」と解釈できる。また、算出された相関係数が「０．２〜０．４」の範囲に含まれる場合、対象リンクＬ１００のリンク速度と実周辺リンクＬ１０１〜Ｌ１１２のリンク速度との間には「やや相関関係がある」と解釈できる。

Table 1 shows the relationship between the absolute value of the correlation coefficient and its interpretation. In Table 1, the calculated correlation coefficients are classified into four stages. When the calculated correlation coefficient is included in the range of “0.0 to 0.2”, there is “almost no correlation between the link speed of the target link L100 and the link speeds of the actual peripheral links L101 to L112. Can be interpreted. When the calculated correlation coefficient is included in the range of “0.2 to 0.4”, the link speed between the target link L100 and the link speeds of the actual peripheral links L101 to L112 is “slightly correlated. Can be interpreted.

  Further, when the calculated correlation coefficient is included in the range of “0.4 to 0.7”, the link speed between the target link L100 and the link speeds of the actual peripheral links L101 to L112 is “significantly correlated. Can be interpreted. When the calculated correlation coefficient is included in the range of “0.7 to 1.0”, a “strong correlation” exists between the link speed of the target link L100 and the link speeds of the actual peripheral links L101 to L112. Can be interpreted. The correlation table 800 is used to create a peripheral link table (see FIG. 9).

  Next, the peripheral link table will be described. FIG. 9 is an explanatory diagram showing a peripheral link table. As shown in FIG. 9, the peripheral link table 900 is a data table in which “peripheral links” for the target link L100 are stored. Here, “peripheral link” means that “there is a strong correlation” with the link speed of the target link L100 based on the correlation table 800 shown in FIG. 8 among the actual peripheral links L101 to L112. The real peripheral links L102, L106, and L109 to be interpreted.

  FIG. 10 is an explanatory diagram showing the relationship between the target link L100 and the peripheral links L102, L106, and L109. As shown in FIG. 10, peripheral links L102, L106, and L109 that are interpreted as “having a strong correlation” with respect to the target link L100 are extracted from the actual peripheral links L101 to L112 having correlation with the target link L100. By doing so, it is possible to reduce the burden on calculation in estimating the link speed in the target link L100 and to improve the accuracy of the estimated link speed. The peripheral link table 900 is used to create a target performance table (see FIG. 11).

  Next, the target performance table will be described. FIG. 11 is an explanatory diagram of a target performance table. The target performance table 1100 is a data table that stores the link speed at the “target date and time” of the target link L100 and the peripheral links L102, L106, and L109 for each “result No.”. The target achievement table 1100 is appropriately created in the database 103.

  Here, “result No.” is information for specifying “target date and time” in the target result table 1100. The “target date / time” is a date / time at which the traffic situation is determined to be similar to the date / time (current time) from which the link speed of the “target link” is estimated among the past date / time. It is.

  The “target date / time” is narrowed down based on, for example, “current date”, “week type”, “WD / WE type”, “current time”, and the like. Here, the “week type” is information indicating the week number in a month. The “WD / WE type” is information indicating whether it is WeekDay (weekdays) or WeekEnd (weekends). On “WD / WE type”, holidays are included in the WeekEnd.

  Next, a method for determining the target date / time will be described. FIG. 12 is an explanatory diagram illustrating a method for determining the target date / time. As shown in FIG. 12, for example, when the date time (current time) for which the link speed is to be estimated is “August 3, 2005 (Wednesday), 9:20, WeekDay”, the target date time is as follows: It is determined as follows.

  In determining the target date and time, first, the week type of “August 3, 2005” is determined. Since “August 3, 2005” corresponds to the first week in August 2005, the month in “target date and time” is August, and the date is the date corresponding to the first week. It is determined.

  In determining the target date and time, the WD / WE type of “August 3, 2005” is determined. Since “August 3, 2005” is WeekDay, the WD / WE type at “Target date / time” is determined to be WeekDay.

  Furthermore, when determining the target date and time, “9:20” is the target time zone in which the time for one day divided in units of five minutes is represented by the time zone for each hour. Determine if it applies. Since “9:20” corresponds to the target time zone of “9: 0 to 9:55”, the target time zone of “target date and time” is determined to be “9:00 to 9:55”. The

  Thus, when the date time (current time) for which the link speed is to be estimated is “August 3, 2005 (Wednesday), 9:20, WeekDay”, the “target date time” is “August, No. 1st week, WeekDay, 9: 00 to 9: 55 ".

  In this case, the target performance table 1100 corresponds to “August, 1st week, WeekDay, 9:00:00 to 9:55” from the link speeds of the peripheral links L102, L106, and L109 in the link information performance DB400. The link speeds of the peripheral links L102, L106, L109 are stored.

  Next, the relationship among the link information record DB 400, the peripheral link table 900, and the target record table 1100 will be described. FIG. 13 is an explanatory diagram showing the relationship among the link information record DB 400, the peripheral link table 900, and the target record table 1100. Reference numeral 1301 in FIG. 13 represents a frame line that spatially narrows the data in the link information performance DB 400 by the peripheral links L102 and L109. Although not shown in FIG. 13, the data in the link information performance DB 400 is also spatially narrowed down by the peripheral link L106.

  Spatial narrowing is performed based on data stored in the peripheral link table 900. Moreover, the code | symbol 1302 in FIG. 13 represents the frame line which narrows down the data in link information performance DB400 temporally according to object date time.

  In FIG. 13, the hatched portion represents data remaining as a result of spatial narrowing and temporal narrowing in the link information performance DB 400. The target achievement table 1100 is created using data narrowed down by the frame lines 1301 and 1302 shown in FIG. 13 among the data in the link information achievement DB 400.

  Next, the current link speed table will be described. FIG. 14 is an explanatory diagram showing a current link speed table. The current link speed table 1400 is a data table that stores the link speeds of the peripheral links L102, L106, and L109 at the date and time (current time) for which the link speed is to be estimated. As shown in FIG. 14, the current link speed table 1400 stores the link speeds of the peripheral links L102, L106, and L109 at the current time. The current link speed table 1400 is appropriately created in the database 103.

  Next, the similarity calculation result table will be described. FIG. 15 is an explanatory diagram of a similarity calculation result table. The similarity calculation result table 1500 includes the results No. of the peripheral links L102, L106, and L109 in the target result table 1100. The degree of similarity between the relationship between each link and the relationship between the peripheral links L102, L106, and L109 at the current time is shown in the record No. It is a data file that stores the degree of similarity in association with each other. In the similarity calculation result table 1500, the similarity q and the similarity ranking are recorded in the record No. Stored for each.

  Here, the “similarity” means the state of the peripheral links L102, L106, L109 at the date time (current time) to be estimated for the link speed, the state of the peripheral links L102, L106, L109 at the target date time, Is information indicating the degree of similarity. The similarity q in the similarity calculation result table 1500 is calculated by calculation using the following equation (2) based on information stored in the target performance table 1100.

  According to the above-described equation (2), the similarity q is the Euclidean distance in the space having the number of dimensions corresponding to the number of peripheral links, with the link speed of each peripheral link L102, L106, L109 in the target performance table 1100 as an axis. Defined by In this embodiment, since the number of the peripheral links L102, L106, L109 in the target performance table 1100 is “3”, the similarity q is defined by the Euclidean distance in the three-dimensional space.

  According to the equation (2), the similarity q can be evaluated as “similarity is higher” as the value is closer to “0”.

  As a specific example, the status of the peripheral links L102, L106, and L109 at the date and time (current time) for which the link speed is to be estimated, and the record No. in the target record table 1100. The evaluation of the similarity between the states of the peripheral links L102, L106, and L109 at the target date and time represented by “J3” will be described.

  In this case, by using the above-described equation (2), the similarity q is calculated as follows.

  Next, the present situation very similar results set is explained. FIG. 16 is an explanatory diagram showing a current situation record set. Based on the value of the similarity q in the above-described similarity calculation result table 1500, the current very similar performance set 1600 includes the similarity q and the actual No. Is a data table that stores a number of items up to a predetermined rank in descending order of similarity rank. The “predetermined order” is a value that is arbitrarily set by an administrator of the traffic situation estimation system 100, for example. The “predetermined order” is the result No. in the similarity calculation result table 1500. The ranking below the number of is set.

  The current very similar result set 1600 is used for calculation of the estimated value of the link speed of the target link L100. Below, the calculation method of the estimated value of the link speed of the object link L100 is demonstrated. FIG. 17 is an explanatory diagram (part 1) of a data table used for link speed estimation. In the data table 1700 shown in FIG. The link speed of the target link L100 in accordance with is stored.

  As the estimated value of the link speed of the target link L100, the link speed corresponding to the record No. having the highest similarity q in the data table 1700 is calculated as the estimated value of the link speed of the target link L100. Specifically, in the case of FIG. 17, the link speed “v22” corresponding to the record No. “J22” having the highest similarity q is calculated as the estimated value of the link speed of the target link L100.

  Further, the estimated value of the link speed of the target link L100 is calculated as an estimated value of the link speed of the target link L100, which is the average value of the link speeds corresponding to the upper predetermined number of achievement numbers having a high similarity q in the data table 1700 It is good to do. In this case, the estimated value “v (estimated value)” of the link speed of the target link L100 can be expressed by the following equation (3).

  In the above-described equation (3), m is the number of link speeds used for calculating v (estimated value). As described above, the unknown link speed in the target link L100 can be estimated while improving the reliability of estimation of the unknown link speed in the target link L100 based on the plurality of pieces of record information.

  Here, a specific example in which the link speed is estimated using the average value will be described. FIG. 18 is an explanatory diagram of a data table (part 2) used for estimating the link speed. FIG. 18 shows a data table 1800 used for link speed estimation when “m = 3”. The data table 1800 stores the link speeds “v22”, “v59”, “v5” corresponding to the top three results Nos. “J22”, “J59”, “J5” with the highest similarity q.

  By performing the calculation of the above-described equation (3) using the data table 1800 of FIG. 18, v (estimated value) is calculated as v (estimated value) = (v22 + v59 + v5) / 3.

  In addition, the estimated value of the link speed of the target link L100 is a calculation result obtained by multiple regression analysis using the link speed of the target link L100 as an explained variable and the link speeds of the peripheral links L102, L106, and L109 as explanatory variables. It may be calculated as an estimated value of the link speed of the target link L100.

  In this case, first, a regression equation for estimating the link speed of the target link L100 is created. The created regression equation is a regression equation that expresses the speed phenomenon of the target link L100 as the speed phenomenon of the peripheral links L102, L106, and L109, using the link speeds of the target link L100 and the peripheral links L102, L106, and L109 as variables. It is expressed by the equation (4) shown below.

v (L100) = a.v (L102) + b.v (L106) + c.v (L109) + d
... (4)

  In the equation (4), a, b, c, and d are coefficients. Of these, a, b, and c are matrix operations using the following formulas (5), (6), and (7) based on the link speeds of the target link L100 and the peripheral links L102, L106, and L109. Is a coefficient calculated by.

  In the expressions (5), (6), and (7) described above, x is the link speed of the peripheral link L102, y is the link speed of the peripheral link L106, and z is the link speed of the peripheral link L109. , W is the link speed of the target link L100. Sxw is a covariance value between the link speed of the peripheral link L102 and the link speed of the peripheral link L109, and Sx2 is a dispersion value of the link speed of the peripheral link L102. Sxw and Sx2 are legends, and the other numerical values are also the same.

  FIG. 19 is an explanatory diagram showing a data table that is the basis of the multiple regression analysis. In the data table 1900 that is the basis of the multiple regression analysis, the link speeds of the target link L100 and the peripheral links L102, L106, and L109 are stored for each performance number. The data table 1900 is created every time a current situation record set 1600 is created.

  In this way, the unknown link speed of the target link L100, which is another variable, can be estimated based on the link speeds of the peripheral links L102, L106, L109, which are variables.

  The regression equation represented by the equation (4) can be created every time the target link and the peripheral link are newly determined. However, when the correlation between the links is known in advance, a link correlation table (see FIG. 20) or a regression table (see FIG. 21) can be created in advance and stored in the database 103.

  Next, the link correlation table will be described. FIG. 20 is an explanatory diagram of a link correlation table. The link correlation table 2000 is a data table that stores correlated links in association with each other. As shown in FIG. 20, in the link correlation table 2000, each of the n links is stored in association with a plurality of links having a correlation with each link.

  Next, the regression equation table will be described. FIG. 21 is an explanatory diagram showing a regression equation table. The regression equation table 2100 is a data table that stores the above-described regression equations in association with each link. As shown in FIG. 21, the regression equation table 2100 stores a regression equation in which the link speed of a correlated link is a variable in association with each other based on the link correlation table 2000.

  Next, a coping method when the link speed does not exist in any one of the peripheral links L102, L106, and L109 in the target performance table 1100 described above will be described. As described above, there are various methods for acquiring the link speed. However, it is assumed that the link speed cannot be acquired using any of the methods.

  For example, in the method of acquiring the link speed using the traffic counter, the link speed of the link cannot be acquired unless there is a vehicle passing through the link in which the traffic counter is embedded. Further, for example, in a method of acquiring a link speed based on position information of a specific vehicle such as a bus or a taxi, the link speed of a link through which a specific vehicle does not pass cannot be acquired.

  For this reason, in the embodiment, the expected value of the link speed is supplemented by the method described below for a portion where the link speed does not exist in the target performance table 1100. The expected link speed is calculated based on data stored in the target performance table 1100.

  Here, the preconditions for supplementing the link speed will be described. When replenishing the link speed, as a precondition, there is no link speed at the date and time (current time) of the link speed estimation target of the peripheral link L102 on the inflow side (node N1 side) of the target link L100. Indicates that the link speed at the relevant location is “unknown”, and the link speed “unknown” corresponds to the actual record No. Are excluded from data targets used for estimating the link speed of the target link L100.

  Further, as a precondition, if there is at least one link speed at the date and time (current time) of the link speed estimation target of the peripheral link L102 on the inflow side (node N1 side) of the target link L100, the target link The link speed of L100 is set as an estimation processing target.

  FIG. 22 is an explanatory diagram for explaining preconditions for supplementing the link speed. FIG. 22 shows the presence / absence of link speeds of the peripheral links L102, L106, and L109 at the date and time (current time) for which the link speed is to be estimated.

  As shown in FIG. 22, the link speed of the peripheral link L102 exists on the inflow side (node N1 side) of the target link L100, and the link speed of the peripheral link L109 exists on the outflow side (node N2 side) of the target link L100. If not, the link speed of the target link L100 is set as the estimation processing target according to the above-described preconditions.

  The expected value of the link speed of the peripheral link L109 can be calculated using the following equation (8).

  As described above, the past link speed in the peripheral link L109 whose link speed is unknown is set as a random variable in the discrete distribution, and the link in the peripheral link L109 whose link speed is unknown based on each random variable and the occurrence probability of each random variable. By calculating the expected value of the speed, the unknown link speed in the target link L100 can be estimated even when there is a surrounding link L109 whose moving speed is unknown, so that it is easy to estimate traffic conditions with high completeness. It can be realized at low cost.

  FIG. 23 is an explanatory diagram showing a current link speed table before and after calculating the expected value of the link speed of the peripheral link L109. As indicated by reference numeral 2310 in FIG. 23, even when the link speed of the peripheral link L109 cannot be acquired, the current link is obtained as indicated by reference numeral 2320 in FIG. 23 by performing the calculation using the equation (8). The expected value of the link speed of the peripheral link L109 can be supplemented in the speed table.

  Then, by using the current link speed table 2320 supplemented with the expected value of the link speed, the degree of similarity q is calculated in the same manner as described above, thereby calculating the above-described present state very similar performance set 1600.

(Functional configuration of the traffic situation estimation apparatus 101)
Next, a functional configuration of the traffic situation estimation apparatus 101 according to the embodiment of the present invention will be described. FIG. 24 is a block diagram showing a functional configuration of the traffic situation estimation apparatus 101 according to the embodiment of the present invention. 24, the traffic situation estimation apparatus 101 includes a storage unit 2401, a detection unit 2402, an extraction unit 2403, a generation unit 2404, a similarity calculation unit 2405, an estimated value calculation unit 2406, a specification unit 2407, An expected value calculation unit 2408.

  First, the storage unit 2401 stores the above-described link information record database 400, actual peripheral link record table 600, correlation coefficient table, peripheral link table 900, target record table 1100, current link speed table 1400, similarity calculation result table 1500, The current very similar achievement set 1600 and the like are stored. The storage unit 2401 stores information transmitted from the client terminal 102 via the network 110. Specifically, the storage unit 2401 realizes its function by a recording medium such as the ROM 202, the RAM 203, and the HD 205 shown in FIG.

  The detection unit 2402 detects, as the target link L100, a link whose moving speed (hereinafter referred to as “link speed”) in the link is unknown among a plurality of links constituting the road network.

  The extraction unit 2403 selects peripheral links L102, L106, and L109 having a predetermined correlation with the target link L100 from the real peripheral links L101 to L112 that exist around the target link L100 detected by the detection unit 2402. Extract.

  Specifically, the extraction unit 2403 extracts the peripheral links L102, L106, and L109 according to the link speed of the target link L100 and the correlation coefficient established between the actual peripheral links L101 to L112. Thereby, the peripheral links L102, L106, and L109 having a clear correlation with the target link L100 can be extracted.

  The generation unit 2404 generates current and past performance information by associating link speeds in the peripheral links L102, L106, and L109 extracted by the extraction unit 2403 for each same date and time. Thereby, the performance information which has a spatial correlation with respect to the object link L100 among the performance information of the peripheral links L102, L106, and L109 can be generated.

  The generation unit 2404 generates, for example, performance information having the same number of months as the number of months to which the date and time specified by the specifying unit 2407 belongs based on the current date and time. Thereby, in addition to a spatial correlation, the present and past performance information in consideration of the season to be estimated can be generated.

  In addition, the generation unit 2404 may generate performance information having the same number of weeks as the number of weeks to which the date and time specified by the specifying unit 2407 belongs based on the week type. Thereby, in addition to a spatial correlation, the present and past performance information which considered the date used as an estimation object can be generated.

  The generation unit 2404 may generate performance information according to whether or not the date and time specified by the specifying unit 2407 is a day of the week and a holiday that belong to the seven-day calendar based on the WD / WE type. Thereby, in addition to the spatial correlation, current and past performance information can be generated in consideration of a calendar condition such as a day of the week or a holiday to be estimated.

  The generation unit 2404 generates time record information related to the date and time specified by the specifying unit 2407. Here, the corresponding time is, for example, a time included between “9: 0 to 9:55” if the current date time is “9:20” as described above. Thereby, in addition to a spatial correlation, the present and past performance information which considered the time used as an estimation object can be generated.

  The similarity calculation unit 2405 calculates the similarity between the current performance information generated by the generation unit 2404 and the past performance information for each date and time. Then, the estimated value calculation unit 2406 calculates an estimated value of the moving speed of the target link L100 based on the similarity calculated by the similarity calculation unit 2405.

  As a result, the past record information on the peripheral links L102, L106, and L109 having a spatial correlation with the target link L100, and a record having a high degree of similarity to the current record information on the peripheral links L102, L106, and L109. By referring to the information, it is possible to estimate an unknown moving speed in the target link L100, so that it is possible to expand the range in which traffic conditions can be estimated without incurring costs for new infrastructure development.

  In addition, the similarity calculation unit 2405 determines whether the target link L100 and each of the peripheral links L102, L106, and L109 in the Euclidean space having the same number of dimensions (three dimensions in this embodiment) as the number of the peripheral links L102, L106, and L109. The similarity q expressed by the Euclidean distance is calculated for each date and time.

  As a result, the reliability of the estimated travel speed is ensured by using the past link information that has a clear correlation with the detected link, and the estimation of traffic conditions with high completeness can be performed easily and at low cost. Can be realized.

  The specifying unit 2407 specifies the current date and time and the past date and time whose traffic conditions are substantially the same as the current date and time. Here, in this case, the traffic conditions are substantially the same. In this case, the generation unit 2404 described above associates the moving speeds in the peripheral links L102, L106, and L109 for each date and time specified by the specifying unit 2407, thereby And the past performance information is generated. Thereby, in addition to a spatial correlation, the present and past performance information which considered the date and time to be estimated can be generated.

  When there is a link whose link speed is unknown among the peripheral links L102, L106, L109, the expected value calculation unit 2408 uses the past moving speed of the unknown link as a random variable in a discrete distribution, Based on the occurrence probability of each random variable, an expected value of the moving speed in the unknown link is calculated. In this case, the generation unit 2404 described above associates the moving speeds in the peripheral links L102, L106, and L109, including the expected value calculated by the expected value calculation unit 2408, for each same date and time, so that the current and past results are recorded. Generate information.

  Thereby, even when there is a surrounding link L109 whose moving speed is unknown, it is possible to estimate the unknown moving speed in the target link L100, so that it is possible to easily and inexpensively estimate the traffic situation with high coverage. it can.

  For example, the estimated value calculation unit 2406 calculates the link speed at the same date and time as the performance information with the highest similarity calculated by the similarity calculation unit 2405 as the link speed of the target link L100. Thereby, it is possible to suppress an increase in the processing burden in the estimation of the moving speed in the estimated value calculation unit 2406, and it is possible to realize traffic condition estimation with high completeness easily and at low cost.

  In addition, the estimated value calculation unit 2406 calculates, for example, the average value of the moving speeds at the same date and time as the predetermined number of pieces of performance information selected in descending order of the similarity calculated by the similarity calculation unit 2405, as the link of the target link L100. It may be calculated as a speed. Thereby, based on a plurality of performance information, the reliability of the estimated moving speed can be ensured, and the estimation of the traffic situation with high completeness can be realized easily and at low cost.

  In addition, the estimated value calculation unit 2406 uses, for example, a calculation result obtained by multiple regression analysis using the link speed of the target link L100 as an explained variable and the link speeds of the peripheral links L102, L106, and L109 as explanatory variables. It may be calculated as the link speed of L100. As a result, the unknown moving speed of the target link L100 can be estimated even when the moving speeds of the target link L100 and the peripheral links L102, L106, and L109 take discrete values. It can be realized easily and at low cost.

  Note that the detection unit 2402, the extraction unit 2403, the generation unit 2404, the similarity calculation unit 2405, the estimated value calculation unit 2406, the specifying unit 2407, and the expected value calculation unit 2408 described above are specifically illustrated in FIG. The functions are realized by the CPU 201 executing the program recorded on the recording medium such as the ROM 202, RAM 203, and HD 205 shown, or by the I / F 209.

(Traffic situation estimation processing procedure)
Next, the traffic situation estimation processing procedure of the traffic situation estimation apparatus according to the embodiment of the present invention will be described. FIG. 25 is a flowchart showing the traffic situation estimation processing procedure of the traffic situation estimation apparatus 101 according to the embodiment of the present invention. In the flowchart of FIG. 25, first, the process waits until the target link L100 is detected (step S2501: No), and when detected (step S2501: Yes), the actual peripheral links L101 to L101 of the detected target link L100. Based on the link speed of L112, the actual peripheral link performance table 600 shown in FIG. 6 is created (step S2502).

  Subsequently, the space is narrowed down (step S2503), the peripheral links L102, L106, and L109 are narrowed down from the actual peripheral links L101 to L112, and the peripheral link table 900 shown in FIG. 9 is created (step S2504).

  Subsequently, the time is narrowed down based on the current time (step S2505), and based on the link speed of the target date and time according to the current time from the past link speeds of the peripheral links L102, L106, and L109, The target performance table 1100 shown in FIG. 11 is created (step S2506).

  Then, in the target result table 1100 created in step S2506, it is determined whether or not the link speeds of the peripheral links L102, L106, and L109 are missing (step S2507), and if there is no missing (step S2507: No). Shifts to step S2510.

  In step S2507, if there is an omission (step S2507: Yes), the expected link speed is calculated as described above (step S2508), and the calculated expected link speed is created in step S2506. The target result table 1100 is supplemented (step S2509).

  Subsequently, the similarity calculation result table 1500 shown in FIG. 15 is created based on the target result table 1100 (step S2510), and the similarity calculation result table 1500 shown in FIG. A current actual result set 1600 is created (step S2511).

  Lastly, as described above, the estimated value of the link speed of the target link L100 is calculated (step S2512) based on the current situation record set 1600 created in step S2511, and the series of processing ends.

  In step S2512, the estimated value of the link speed of the target link L100 can be calculated using the various methods described above. For example, when the estimated value of the link speed of the target link L100 is calculated based on the record information having the highest similarity q, among the past link speeds of the target link L100, the record No. Is calculated as an estimated value of the link speed of the target link L100.

  Further, in step S2512, as described above, an estimated value of the link speed of the target link L100 may be calculated based on the upper predetermined number of pieces of performance information having a high degree of similarity, and the link speed of the target link L100 is calculated. A calculation result obtained by multiple regression analysis using the link speeds of the peripheral links L102, L106, and L109 as explanatory variables may be calculated as the link speed of the target link L100.

  The estimated value calculated in this way can be used to estimate traffic conditions such as the occurrence of traffic jams and the size of traffic jams. For example, if the link speed of the target link L100 is equal to or lower than a predetermined threshold, it can be estimated that traffic congestion has occurred in the target link L100. In addition to the link speed of the target link L100, if the link speeds of the peripheral link L102 and the peripheral link L109 are below a predetermined threshold, the peripheral link L109, the target link L100, and the peripheral link L102 are continuously congested. Can be estimated.

  As described above, according to the traffic situation estimation program, the recording medium storing the program, the traffic situation estimation method, and the traffic situation estimation apparatus according to the present invention, it is easy and low-cost to estimate the traffic situation with high completeness. Can be realized in

  The traffic situation estimation method described in the present embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. The program may be a transmission medium that can be distributed via a network such as the Internet.

(Supplementary note 1) a detection step of detecting a link whose current moving speed is unknown from a plurality of links constituting the road network;
An extracting step of extracting a peripheral link having a predetermined correlation with the detected link from actual peripheral links existing around the link detected by the detecting step (hereinafter referred to as “detected link”);
A generation step of generating current and past performance information by associating the moving speeds in the peripheral links extracted by the extraction step for each same date and time,
A similarity calculation step for calculating the similarity between the current performance information generated in the generation step and the past performance information for each date and time;
An estimated value calculating step for calculating an estimated value of the moving speed of the detection link based on the similarity calculated by the similarity calculating step;
A traffic situation estimation program characterized in that a computer is executed.

(Supplementary note 2)
The traffic situation estimation program according to appendix 1, wherein the peripheral link is extracted in accordance with a moving speed of the detection link and a correlation coefficient established between each of the actual peripheral links.

(Additional remark 3) Let the said computer perform the specific process which specifies the present date time, and the said present date time and the past date time in which traffic conditions are substantially the same,
The generation step includes
The traffic situation estimation program according to appendix 1 or 2, wherein the current and past performance information is generated by associating a moving speed in the peripheral link for each date and time specified by the specifying step.

(Supplementary Note 4) The acquisition step includes:
The traffic condition estimation program according to appendix 3, wherein the track record information having the same number of months as the number of months to which the date and time specified by the specifying step belongs is acquired.

(Supplementary Note 5) The acquisition step includes:
The traffic situation estimation program according to appendix 4, wherein the track record information having the same number of weeks as the number of weeks to which the date and time specified by the specifying step belongs is acquired.

(Appendix 6)
The traffic condition estimation according to Supplementary Note 3 or 4, wherein the result information is acquired depending on whether the date and time specified by the specifying step is a day of the week or a holiday according to a day of the week belonging to the seventh calendar. program.

(Appendix 7)
The traffic situation estimation program according to any one of appendices 3 to 6, wherein the performance information corresponding to the time at the date and time specified by the specifying step is acquired.

(Supplementary Note 8) When there is a link whose movement speed is unknown among the peripheral links, the past movement speed of the unknown link is set as a random variable in a discrete distribution, and the occurrence probability of each random variable and each random variable And causing the computer to execute an expected value calculation step of calculating an expected value of the moving speed in the unknown link,
The generation step includes
Any one of appendices 1 to 7, wherein current and past performance information is generated by associating the moving speeds in the peripheral links including the expected value calculated by the expected value calculating step for each same date and time. The traffic situation estimation program according to any one of the above.

(Supplementary Note 9) The similarity calculation step includes:
Appendices 1 to 8, wherein the similarity expressed by the Euclidean distance between the detected link and each of the peripheral links in the Euclidean space having the same number of dimensions as the number of the peripheral links is calculated for each date and time. The traffic situation estimation program according to any one of the above.

(Supplementary Note 10) The estimated value calculating step includes:
The moving speed at the same date and time as the record information having the highest degree of similarity calculated in the similarity calculating step is calculated as the moving speed of the detection link. Traffic situation estimation program.

(Supplementary Note 11) The estimated value calculating step includes:
Supplementary note 1 wherein the average value of the moving speeds at the same date and time as a predetermined number of pieces of performance information selected in descending order of the degree of similarity calculated in the similarity calculating step is calculated as the moving speed of the detection link. The traffic situation estimation program according to any one of? 9.

(Supplementary Note 12) The estimated value calculation step includes:
Supplementary notes 1 to 9, wherein a calculation result obtained by multiple regression analysis using the movement speed of the detection link as an explanatory variable and the movement speed of a peripheral link as an explanatory variable is calculated as the movement speed of the detection link. The traffic situation estimation program according to any one of the above.

(Supplementary note 13) A computer-readable recording medium in which the traffic situation estimation program according to any one of supplementary notes 1 to 12 is recorded.

(Additional remark 14) The detection process which detects the link in which the present moving speed in the said link is unknown from the some links which comprise a road network,
An extraction step of extracting a peripheral link having a predetermined correlation with the detection link from real peripheral links existing around the link detected by the detection step (hereinafter referred to as “detection link”);
An acquisition step of acquiring current and past performance information that associates the movement speed in the peripheral links extracted by the extraction step for each same date and time;
A similarity calculation step of calculating the similarity between the current performance information acquired in the acquisition step and the past performance information for each date and time;
An estimated value calculating step of calculating an estimated value of the moving speed of the detection link based on the similarity calculated by the similarity calculating step;
The traffic condition estimation method characterized by including.

(Supplementary Note 15) Detection means for detecting a link whose current moving speed in the link is unknown from a plurality of links constituting the road network;
Extracting means for extracting a peripheral link having a predetermined correlation with the detected link from real peripheral links existing around the link detected by the detecting means (hereinafter referred to as “detected link”);
Acquisition means for acquiring current and past performance information that associates the movement speed in the peripheral links extracted by the extraction means for each same date and time;
Similarity calculation means for calculating the degree of similarity between the current performance information acquired by the acquisition means and past performance information for each date and time;
An estimated value calculating means for calculating an estimated value of the moving speed of the detection link based on the similarity calculated by the similarity calculating means;
A traffic situation estimation apparatus comprising:

  As described above, the traffic situation estimation program, the recording medium on which the program is recorded, the traffic situation estimation apparatus, and the traffic situation estimation method according to the present invention are useful for estimating the travel speed in a link whose current travel speed is unknown. is there.

1 is a system configuration diagram of a traffic situation estimation system according to an embodiment of the present invention. It is a block diagram which shows the hardware constitutions of the computer apparatus shown in FIG. It is explanatory drawing (the 1) explaining the relationship of a link. It is explanatory drawing which shows a link information performance database. It is explanatory drawing (the 2) explaining the relationship of a link. It is explanatory drawing which shows a real periphery link performance table. It is explanatory drawing (the 1) which shows a correlation coefficient table. It is explanatory drawing (the 2) which shows a correlation coefficient table. It is explanatory drawing which shows a periphery link table. It is explanatory drawing which shows the relationship between an object link and a periphery link. It is explanatory drawing which shows a subject performance table. It is explanatory drawing explaining the determination method of object date time. It is explanatory drawing which shows the relationship between link information performance DB, a periphery link table, and a target performance table. It is explanatory drawing which shows the present link speed table. It is explanatory drawing which shows a similarity calculation result table. It is explanatory drawing which shows the present condition very similar performance set. It is explanatory drawing which shows the data table (the 1) used for estimation of a link speed. It is explanatory drawing which shows the data table (the 2) used for estimation of a link speed. It is explanatory drawing which shows the data table used as the origin of multiple regression analysis. It is explanatory drawing which shows a link correlation table. It is explanatory drawing which shows a regression equation table. It is explanatory drawing explaining the precondition at the time of replenishment of a link speed. It is explanatory drawing which shows the present link speed table before and after calculating the expected value of the link speed of a peripheral link. It is a block diagram which shows the functional structure of the traffic condition estimation apparatus concerning embodiment of this invention. It is a flowchart which shows the traffic condition estimation procedure of the traffic condition estimation apparatus concerning embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Traffic condition estimation apparatus 2402 Detection part 2403 Extraction part 2404 Generation part 2405 Similarity calculation part 2406 Estimated value calculation part 2407 Specification part 2408 Expected value calculation part

Claims (5)

A detection step for detecting a link whose current moving speed is unknown from a plurality of links constituting the road network;
An extracting step of extracting a peripheral link having a predetermined correlation with the detected link from actual peripheral links existing around the link detected by the detecting step (hereinafter referred to as “detected link”);
A generation step of generating current and past performance information by associating the moving speeds in the peripheral links extracted by the extraction step for each same date and time,
A similarity calculation step for calculating the similarity between the current performance information generated in the generation step and the past performance information for each date and time;
An estimated value calculating step for calculating an estimated value of the moving speed of the detection link based on the similarity calculated by the similarity calculating step;
A traffic situation estimation program characterized in that a computer is executed. The extraction step includes
The traffic situation estimation program according to claim 1, wherein the peripheral link is extracted according to a moving speed of the detection link and a correlation coefficient established between each of the actual peripheral links. Causing the computer to execute a specific step of identifying a current date time and a past date time that is substantially the same as the current date time and traffic conditions;
The generation step includes
The traffic situation estimation program according to claim 1 or 2, wherein the current and past performance information is generated by associating a moving speed in the peripheral link for each date and time specified by the specifying step. When there is a link whose movement speed is unknown among the peripheral links, the past movement speed in the unknown link is set as a random variable in a discrete distribution, and based on each probability variable and the occurrence probability of each random variable. , Causing the computer to execute an expected value calculation step of calculating an expected value of the moving speed in the unknown link,
The generation step includes
The current and past performance information is generated by associating the moving speed in the peripheral link including the expected value calculated by the expected value calculating step for each same date and time. The traffic situation estimation program according to any one of the above. A detection step of detecting a link whose current moving speed is unknown from a plurality of links constituting the road network;
An extraction step of extracting a peripheral link having a predetermined correlation with the detection link from real peripheral links existing around the link detected by the detection step (hereinafter referred to as “detection link”);
A generation step of generating current and past performance information by associating the moving speed in the peripheral links extracted by the extraction step for each same date and time,
A similarity calculation step for calculating the similarity between the current performance information generated by the generation step and the past performance information for each date and time;
An estimated value calculating step of calculating an estimated value of the moving speed of the detection link based on the similarity calculated by the similarity calculating step;
The traffic condition estimation method characterized by including.

Images