JP2010287214A - Device, computer program and method for estimating traffic information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately estimate traffic information. <P>SOLUTION: A traffic information estimation device is configured to estimate the traffic information of an estimation object road link on the basis of the traffic information of the other road link, and provided with: an estimation system 11 for estimating the traffic information of the estimation object road link by using the traffic information of the other road link and a parameter for estimation; and a leaning system 12 for storing the combination of the traffic information based on an execution value in a road link which may be the estimation object road link and the traffic information of the other road link to be used for estimating the traffic information of the estimation object road link as data for learning, and for optimizing the parameter for estimation by using the stored data for learning. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a traffic information estimation device for estimating traffic information, a computer program for traffic information estimation, and a traffic information estimation method.

As a technology for providing road traffic information to drivers, VICS (Vehicle Information and Communication System: “VICS” is a registered trademark of the foundation) is widely known.
This VICS tabulates traffic information including traffic congestion and link travel time on each route based on fixed point observation information consisting of the number of vehicles, vehicle speed, etc. collected from various roadside sensors (vehicle detectors, loop coils, etc.). The traffic information is provided to the driver by wide-area communication such as narrow-area communication using beacons or FM broadcasting.

As another technique for providing road traffic information to a driver, a traffic information estimation system using a probe car (hereinafter referred to as a probe system) is also known.
For example, as shown in Patent Documents 1 and 2, this probe system uses a vehicle (probe vehicle) that actually travels on a road as a moving body sensor, and wirelessly communicates probe information such as the current vehicle position and time. Are collected from each probe vehicle to generate road traffic information.

Japanese Patent Laid-Open No. 5-151696 Japanese Patent Laying-Open No. 2005-4467

  However, even if the VICS and the probe system are used, traffic information cannot be obtained for road links for which no data is obtained from either the VICS or the probe system.

  Then, an object of this invention is to provide the new technique for estimating the traffic information of an estimation object road link based on the traffic information of another road link.

(1) The present invention is a traffic information estimation device that estimates traffic information of an estimation target road link based on traffic information of another road link, and uses the traffic information and estimation parameters of the other road link. , An estimation unit for estimating traffic information of the estimation target road link, traffic information based on measured values in the road link that can be the estimation target road link, and a road link used for estimating the traffic information of the estimation target road link And a learning unit that stores the combination with the traffic information as learning data and optimizes the estimation parameter using the stored learning data.
According to the present invention, the estimation parameter is optimized by the learning unit, and traffic information can be estimated appropriately.

(2) Moreover, it is preferable that the said estimation part estimates the traffic information of the said estimation object road link also using the traffic information of the said estimation object road link already acquired.
In the same road link, it is considered that the traffic information is correlated even if the time is slightly different. For this reason, if the estimation part estimates the traffic information of the said estimation object road link also using the traffic information of the estimation object road link already acquired, it will be able to estimate traffic information more appropriately.

(3) Moreover, it is preferable that the learning information used by the learning unit includes traffic information that has already been acquired with respect to the road link that can be the estimation target road link.
In the same road link, it is considered that the traffic information is correlated even if the time is slightly different. For this reason, the traffic information already acquired regarding the road link that can be the estimation target road link is included in the combination of the learning data, and the traffic information of the estimation target road link is estimated using the learning data. If an estimation parameter for learning is learned, a more appropriate estimation parameter can be obtained.

(4) In the traffic information estimation device, when the traffic information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the estimation unit includes the traffic information based on the actual measurement value. The traffic information of the estimation target road link is estimated using the traffic information of the road link and the estimation parameter, and the learning unit includes the traffic information based on the actual measurement value of the predetermined road link, and the predetermined road link. The estimation parameter is optimized by using, as the learning data, a combination of the road link traffic information used for estimating the traffic information and the traffic information estimated by the estimation unit using the actual measurement value. It can be set as the structure to do.
In this case, when traffic information based on measured values is acquired for a predetermined road link that can be an estimation target road link, the estimation unit estimates the traffic information of the estimation target road link, and then uses the estimated traffic information. Thus, the learning unit optimizes the estimation parameters.

(5) On the contrary, in the traffic information estimation device, when the traffic information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the learning unit is based on the actual measurement value. The estimation parameter is optimized using a combination of traffic information and road link traffic information used for estimating the traffic information of the predetermined road link as the learning data, Using the estimation parameters optimized by the learning unit using the traffic information of road links including traffic information based on the actual measurement values and the learning data combined with the traffic information based on the actual measurement values, It can be set as the structure which estimates the traffic information of a road link.
In this case, when traffic information based on measured values is acquired for a predetermined road link that can be an estimation target road link, the learning unit optimizes the estimation parameters using the traffic information based on the measured values, and then The estimation unit estimates the traffic information of the estimation target road link using the estimation parameters.

(6) In the case of (5), the learning unit is based on the traffic information based on the actual measurement value and the road link traffic information used for estimating the traffic information on the predetermined road link. The estimation parameter can be optimized using a combination of the temporarily estimated traffic information of the predetermined road link as learning data.
In this case, when the traffic information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the traffic information of the predetermined road link based on the actual measurement value and the traffic of the predetermined road link estimated temporarily A parameter for estimation is learned using a combination with information as learning data.

(7) The present invention is a computer program for causing a computer to function as the device according to any one of (1) to (6) above.

(8) Moreover, this invention is a traffic information estimation method which estimates the traffic information of an estimation object road link based on the traffic information of another road link, Comprising: About the predetermined | prescribed road link which can become an estimation object road link When the traffic information based on the actual measurement value is acquired, the traffic information of the estimation target road link is estimated using the traffic information of the road link including the traffic information based on the actual measurement value and the estimation parameter, and the predetermined road link A combination of the traffic information based on the actual measurement value in the traffic information and the traffic information of the road link used for estimating the traffic information of the predetermined road link and estimated using the actual measurement value The estimation parameter is optimized by using it as data.
According to the present invention, when traffic information based on measured values is acquired for a predetermined road link that can be an estimation target road link, the traffic information of the estimation target road link is estimated, and then the estimated traffic information is used. Thus, the estimation parameters are optimized, and traffic information can be estimated appropriately.

(9) Moreover, this invention is a traffic information estimation method which estimates the traffic information of an estimation object road link based on the traffic information of another road link, Comprising: About the predetermined | prescribed road link which can become an estimation object road link When the traffic information based on the actual measurement value is acquired, a combination of the traffic information based on the actual measurement value and the road link traffic information used for estimating the traffic information of the predetermined road link is used as the learning data. The estimation parameter is used as an optimization, the road link traffic information including traffic information based on the actual measurement value, and the learning data combined with the traffic information based on the actual measurement value is optimized. The traffic information of the estimation target road link is estimated using the estimation parameter.
According to the present invention, when traffic information based on an actual measurement value is acquired for a predetermined road link that can be an estimation target road link, the estimation parameter is optimized using the traffic information based on the actual measurement value. The traffic information of the estimation target road link is estimated using the estimation parameters. As a result, traffic information can be estimated appropriately.

  The parameters for estimation are optimized by the learning unit, and traffic information can be estimated appropriately.

1 is an overall configuration diagram of a traffic information system. It is a block diagram of a traffic information estimation apparatus. It is a block diagram of an input information processing part. It is a figure which shows conversion information. It is explanatory drawing for calculating | requiring a link travel speed from probe information. It is a flowchart which shows the traffic information estimation process procedure. It is a block diagram of a neural network. It is a figure which shows the example of a road link. It is a figure which shows the initial state of an estimation database. It is a figure which shows the initial state of a weight database. It is a figure which shows the VICS information and probe information which were acquired as input information. It is a figure which shows the estimation database after the 1st update. It is a figure which shows the estimation database after the 2nd update. It is a figure which shows the estimation database after the 3rd update. It is a figure which shows the part extracted by taking a snapshot. It is a figure which shows the database for learning. It is a flowchart which shows a learning process. It is a figure for demonstrating the problem in the case of estimating with speed. It is explanatory drawing explaining the process (1st embodiment) by an estimation engine and a learning engine, (a) is estimation and (b) is learning. It is a figure which shows the other example of a road link. It is explanatory drawing explaining the process (2nd embodiment) by an estimation engine and a learning engine, (a) is learning, (b) is estimation. It is a flowchart which shows a traffic information estimation process procedure (2nd embodiment). It is explanatory drawing of an estimation database and a weight database. It is explanatory drawing of an estimation database and the updated weight database. It is explanatory drawing of the updated estimation database and weight database. It is a flowchart which shows a traffic information estimation process procedure (3rd embodiment). It is explanatory drawing of an estimation database and a weight database. It is explanatory drawing of a temporary estimation database. It is explanatory drawing of an estimation database and the updated weight database. It is explanatory drawing of the updated estimation database and weight database.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.
[1. overall structure]
FIG. 1 shows a traffic information system including a traffic information estimation device (central device) 1 according to an embodiment of the present invention. In addition to the traffic information estimation device 1, the traffic information system includes a probe vehicle 3 equipped with an in-vehicle device 2, a roadside communication device 4 that wirelessly communicates with the in-vehicle device 2, and a roadside sensor 5.

  The traffic information estimation device 1 indicates, for example, one of various functions in the central device, and the central device 1 acquires traffic information (observation information) such as VICS information and probe information and provides it to the vehicle. It has a function of generating traffic information for provision for the purpose. The central device 1 may perform various traffic processing such as traffic signal control and traffic control based on the traffic information.

  The traffic information estimation device (central device) 1 is configured by a computer having a processing device and a storage device. A computer program for causing the computer to function as the traffic information estimation device (central device) is stored in the storage device. It is remembered. This computer program is executed by the processing device, and the processing device performs functions as the traffic information estimation device (central device) 1 by inputting and outputting to the storage device and the like. Note that the functions of the traffic information estimation apparatus (central apparatus) 1 described below are realized by the computer program unless otherwise specified.

  The in-vehicle device 2 generates probe information as observation information of the probe vehicle 3 and wirelessly transmits it to the roadside communication device 4. The probe information is traffic information including the position of the probe vehicle, the passing speed of the position, the vehicle ID of the probe vehicle, and the like. The probe information may include other information such as the speed of the probe vehicle. The position of the probe vehicle is calculated based on the GPS signal received by the GPS receiver included in the in-vehicle device 2.

  The roadside communication device 4 transmits / receives information to / from the in-vehicle device 2 by wireless communication. Specifically, the roadside communication device 4 receives probe information as observation information transmitted by the in-vehicle device 2 and transfers the probe information to the traffic information estimation device (central device) 1. Further, the roadside communication device 4 can acquire traffic information for provision to the vehicle from the traffic information estimation device (central device) 1 and transmit the traffic information to the in-vehicle device 2. The roadside communication device 4 and the traffic information estimation device 1 are connected by a communication line.

  The roadside sensor 5 is for detecting traffic information as observation information, for example, a vehicle detector for ultrasonically detecting a vehicle passing directly below, a loop coil for detecting the vehicle by inductance change, or It consists of an image sensor that measures the traffic volume and vehicle speed by image processing of the camera video, and is installed on expressways and major trunk roads for the purpose of measuring the number of vehicles flowing into the intersection and the vehicle speed. .

  The roadside sensor 5 transmits the detected observation information 6 to the VICS center server via the communication line, and the VICS center server 6 generates VICS information based on the observation information of the roadside sensor 5. This VICS information is transmitted to the traffic information estimation device 1 via a communication line.

The VICS information is traffic information including traffic jams and link travel times on each road link. Since the VICS information is updated by obtaining observation information from the roadside sensor 5 every 5 minutes, high-density information can be obtained in time. However, the roadside sensor 5 is not installed on all roads (20% or less on main roads), and the area coverage rate is low.
On the other hand, since the probe information is acquired from the probe vehicle 3 traveling on the road, the area coverage rate can be increased. However, since the penetration rate of the in-vehicle device 2 to become the probe vehicle 3 is still low, only low-density data can be obtained in time.

  That is, when a road link is set on a road in a predetermined area, traffic information (VICS information) is always obtained for each VICS information update unit time for a road link from which VICS information is obtained. On the other hand, as for road links for which VICS information cannot be obtained, there are road links for which probe information can be obtained as traffic information, and road links for which probe information cannot be obtained for each VICS information update unit time. become.

[2. Details of Traffic Information Estimator]
[2.1 Overall configuration of estimation device]
The traffic information estimation device 1 according to the present embodiment estimates the traffic information of a road link for which neither VICS information nor probe information is obtained, and complements the traffic information of the road link, so that all road links in the predetermined area Generate traffic information. In this way, by supplementing road link traffic information for which neither VICS information nor probe information can be obtained, more accurate traffic information can be provided to the vehicle (navigation system), or traffic control can be performed with high accuracy. Is possible.
In the following, a road link for which VICS information or probe information is not obtained and traffic information needs to be estimated and complemented is referred to as “estimation target road link”.

  As shown in FIG. 2, the traffic information estimation device 1 learns an estimation engine 11 (estimation unit) for estimating traffic information of an estimation target road link and a parameter (estimation parameter) used for estimation. And a learning engine 12. The estimation engine 11 and the learning engine 12 are functions of the traffic information estimation device 1 and are an estimation system and a learning system realized by a computer program stored in the traffic information estimation device 1.

  The traffic information estimation device 1 accumulates data used for learning in the learning engine 12 among the estimation database (traffic information database) 13 for accumulating traffic information estimated by the estimation engine 11 and data in the estimation database 13. A learning database 14 and a weight database (estimated parameter database) 15 for accumulating estimated parameters (in this embodiment, “weight”).

[2.2 Input information processing section]
Furthermore, the traffic information estimation device 1 according to the present embodiment performs input processing on VICS information and probe information (hereinafter, referred to as “input information” when both information are collectively) acquired by the traffic information estimation device 1. A processing unit 16 is provided.

  The input information processing unit 16 performs processing for generating “speed information” of each link from the input information. The generated “speed information” is given to the estimation database 13 to update the estimation database 13. Used.

The input information processing unit 16 includes a VICS information processing unit 17 that generates “speed information” from the acquired VICS information, and a probe information processing unit 18 that generates “speed information” from the acquired probe information. .
The VICS information processing unit 17 calculates a link travel speed calculation unit 17a that calculates the link travel speed of the road link from the link travel time included in the VICS information, and a speed that converts the link travel speed into “speed information” of the link. And an information conversion unit 17b.

  The link travel speed calculation unit 17a of the VICS information processing unit 17 divides the link length of the road link for which the link travel speed is to be calculated by the link travel time of the road link, so that the link travel speed [km / h] Is calculated. The link lengths of all road links in the target area are set in the device 1 in advance.

  The speed information conversion unit 17b of the VICS information processing unit 17 converts the link travel speed obtained by the link travel speed calculation unit 17a into an index value (speed index information) called “speed information”. This conversion is performed by the speed information conversion unit 17 b referring to the “speed-speed information” conversion information 19 preset in the apparatus 1. The “speed-speed information” conversion information 19 is set as shown in FIG. 4 and will be described later in detail.

  The link travel speed calculation unit 18b of the probe information processing unit 18 calculates the link travel speed of the road link for which the link travel speed is to be calculated based on the position and time included in the probe information. A method for calculating the link travel speed [km / h] from the probe information will be described later.

  Similar to the speed information conversion unit 17b of the VICS information processing unit 17, the speed information conversion unit 18b of the probe information processing unit 18 converts the link travel speed into “speed information”. The processing content is the same as that of the speed information conversion unit 17b.

[2.3 Speed information]
The “speed information” is an index value that takes a value according to the size of the link travel speed. In this embodiment, the speed information takes a value from 0 to 1, and “speed information” = 0 is the link travel speed = 100 [km / H] and higher speeds, “speed information” = 1 corresponds to link travel speed = 0 [km / h].

However, the “speed information” in the present embodiment is not simply inversely proportional to the link travel speed. That is, as shown in FIG. 4, the link travel speed-speed information relational expression is different depending on a plurality of speed regions of the link travel speed.
More specifically, the plurality of speed regions are a first speed region (near zero speed region) having a link travel speed of 0 [km / h] to about 10 (more strictly 10.53) [km / h]. ), A second speed region in which the link travel speed is about 10 (more precisely 10.5) [km / h] to about 60 (more precisely 57.9) [km / h], the link travel speed is about The three speed regions are the third speed region (free travel speed region) of 60 (more precisely, 57.9) [km / h].

  The first speed region is set as a near-zero speed region where the vehicle speed is 0 or a value close thereto, and such a speed region value is detected when the road is congested. The third speed region is set as a speed region that is equal to or higher than the speed limit (60 km / h) of a general road. When the road is relatively smooth and the vehicle can be in a free-running state, such a speed region is set. The value of is detected. The second speed area is set as an intermediate area between the first and third speed areas.

  If the value of the horizontal axis (link travel speed) in FIG. 4 is x and the value of the vertical axis (speed information) is y, the relational expression of link travel speed-speed information in the first speed region is y = 1− (x / 1000), the relational expression of the link travel speed-speed information in the second speed region is set to y = 1.2- (x / 50), and the link travel speed-speed information of the third speed region is The relational expression is set to y = 0.1− (x / 1000).

As described above, the link travel speed-speed information relationship line is focused on the third speed area and the second speed area. In the third speed area (inclination = 1/1000), the second speed area (inclination = 1/1000). 50), the negative slope is smaller.
That is, in the third region, which is a free speed region in which the vehicle can be in a free-running state, the rate of change in the speed information (y) with respect to the change in the link travel speed (x) (as compared to the lower second speed region) (Negative slope) is smaller.

In addition, the link travel speed-speed information relational line is focused on the first speed area and the second speed area. In the first speed area (inclination = 1/1000), the second speed area (inclination = 1/50). The inclination is smaller than that.
That is, in the first region where the vehicle speed is near zero, the rate of change in the speed information (y) with respect to the change in the link travel speed (x) (negative slope) compared to the higher-speed second speed region. Is getting smaller.

  According to the conversion information 19 set as described above, in addition to an element of simply “speed”, the road information is busy and traffic jams, or the road is empty and the vehicle is in good condition. It is also possible to have an element of the degree of traffic congestion (the degree of traffic flow).

For example, if the road travel speed is 100 km / h on a road link with a speed limit of 100 km / and the road travel speed is 60 km / h on a road link with a speed limit of 60 km / h, the speed difference is 40 km / h. Although there are vehicles on all road links, all are considered to be free-running.
On the other hand, when the link travel speed of one road link is 60 km / h and the link travel speed of the other road link is 20 km / h, the difference in speed is 40 km / h as described above. However, although the vehicle is flowing smoothly on the former road link, the latter road link is considered to be congested.

In this way, when the link travel speed reaches the free travel speed range, the relationship between the number of vehicles on the road (congestion) and the link travel speed becomes relatively low, which may be caused by other factors such as the road speed limit. The link speed is likely to change.
For this reason, for example, assuming a case where traffic information (link travel speed) of an estimation target road link is to be estimated from traffic information (link travel speed) of other related road links, the third of other road links The influence of the difference in the link travel speed in the speed region (free travel speed region) on the estimated value of the link travel speed of the estimation target road link is smaller than the difference in the link epidemic speed in the slower second speed region.

In the first speed region, as in the third speed region, when the speed is 0 to 10 [km / h], the road is considered to be considerably congested at any speed. The difference in the link travel speed in the area is less meaningful than the difference in the link trend speed in the high-speed second speed area.
The speed information of the present embodiment takes the above into consideration, and the traffic information of the estimation target road link can be accurately estimated by reducing the slopes of the first and third speed regions.

Further, as a result of reducing the slopes of the first and third speed regions, the slope of the second speed region is increased, and the difference in the link travel speed in the second speed region can be more emphasized in the index of speed information. .
For example, when the link travel speed (x) in the range of 0 to 100 km / h is associated with the speed information with one relational expression y = 1− (x / 100), the speed is in the range of 10 to 60 km / h. Of course, the slope is (1/100), but in the case of FIG. 4, the slope is (1/50). Therefore, in the speed information obtained from FIG. 4, the link travel speed at 10 to 60 km / h can be more accurately indexed. As a result, it is possible to improve the estimation accuracy of the traffic information of the estimation target road link.

Conventionally, when the speed of a vehicle on a road link is handled as traffic information, the traffic information is estimated at the speed [km / h]. In this case, the following problem occurs.
For example, it is assumed that road links are configured as shown in FIG. 18, links A and B are highways with a speed limit of 100 km / h, and link C is a general road with a speed limit of 60 km / h. Further, it is assumed that the road is free in any of links A, B, and C, and the vehicle is in a free-running state.
It is assumed that the speed (link travel speed) is detected as 100 km / h for the links A and B, and the speed (link travel speed) is detected as 60 km / h for the link C. In this case, there is a difference in speed between the links A and B and the link C, but this is only due to the difference in the speed limit of the road, and as the degree of congestion (traffic volume) of the road, The links A, B, and C are considered to be almost the same.

However, when trying to estimate the speed of the road link (estimation target road link) related to the links A, B, C based on the speeds of the links A, B, C, the links A, B, and C Since the speeds are different, there are problems that it is difficult to obtain an appropriate estimated value, and that the processing time for obtaining an appropriate estimated value becomes long.
However, as described above, in the free speed region where the vehicle can be in a free-running state, the rate of change in the speed information with respect to the change in the link travel speed is smaller than that in the lower speed region. Such problems can be solved.

[2.4 Calculation of link travel time from probe information]
FIG. 5 shows a method for calculating the link travel speed [km / h] from the probe information in the link travel speed calculation unit 18a.
As shown in FIG. 5, in the case where there are six road links L1 to L6, the probe vehicle 3 is used to calculate the link travel speed of the road link L4 toward the right side of the figure (direction of arrow A). The first probe information pr1 transmitted while the vehicle is located at the road link L3 (first position) immediately before the traveling direction, and the road link L5 (first The second probe information pr2 transmitted while being located at (2 position) is used.

  The distance D between the first position where the first probe information pr1 is transmitted and the second position where the second probe information pr2 is transmitted is determined by the position indicated by the first probe information pr1 and the position indicated by the second probe information pr2. . In addition, the time T required to move from the first position to the second position is determined by the time indicated by the first probe information pr1 and the time indicated by the second probe information pr2. By dividing the distance D by the time T, the travel speed of the section (normal speed calculation section) between the first position and the second position is obtained. Normally, this travel speed is regarded as the link travel speed on the road link L4 between the first position and the second position.

However, when the distance between the first position and the second position (distance of the normal speed calculation section) is equal to or less than the set minimum distance (for example, 1 km), the speed calculation section becomes long. Another piece of probe information used for calculating the link travel speed of the road link L4 is selected.
Specifically, when the distance of the normal speed calculation section is equal to or less than the minimum distance, the link travel speed calculation unit 18a has the vehicle positioned on the road link L2 (front position) one before in the vehicle traveling direction. The probe information pr0 transmitted at times and the probe information pr2 transmitted when located on the road link LL5 are used. As a result, the travel speed for the extended speed calculation section from the front position to the second position is obtained, and this travel speed is calculated as the link travel speed of the road link L4 between the front position and the second position. I reckon.

  The reason for using the extended speed calculation section when the distance of the normal speed calculation section is short is as follows. That is, when the speed calculation section is short, the time required to move in the speed calculation section is greatly affected by the stop time due to the traffic signal. In other words, even if the time required to pass the speed calculation section is, for example, 2 minutes, the time when the vehicle actually traveled is 1 minute, and the remaining 1 minute is a waiting time due to a signal. obtain. On the other hand, another vehicle passing through the same speed calculation section does not stop by a signal, and the time required to pass through the speed calculation section may be one minute. Thus, if the speed calculation section is short, an error due to the influence of waiting for a signal becomes large, and accurate speed calculation becomes difficult.

  On the other hand, if it is a speed calculation section where the distance is sufficiently long and it is necessary to pass many traffic signals (intersections), the waiting time due to one of the traffic signals is almost all of the vehicles. To occur evenly. The error due to signal waiting is small.

Thus, as in the present embodiment, when the distance of the normal speed calculation section is short, the speed can be calculated with higher accuracy by calculating the speed in the extended speed calculation section.
Even if the distance of the normal speed calculation section is short, if the number of vehicles is large (for example, 10 or more), the time required to pass through the speed calculation section is determined by a probe from a plurality of vehicles. By averaging with information, the error can be reduced. Therefore, it is expanded only when the normal speed calculation section is not more than the set minimum distance and the number of probe vehicles per unit time (the number of probe information) is not more than the set minimum number. Speed calculation in the calculated speed calculation section may be performed.

  In the above example, the extended speed section is from the front position (road link L2) to the second position (road link L5), but instead of the second position, the traveling direction is more than the second position. It is good also as a point ahead (road link L6). Further, the extended speed section may be between the first position (road link L3) and the previous position (road link L6).

[First embodiment]
[2.5 Processing by the traffic information estimation device 1]
[2.5.1 Outline of estimation process]
FIG. 6 shows a traffic information estimation method by the traffic information estimation apparatus 1. First, initial values are set in the estimation database (traffic information database) 13 and the weight database 15 (step S1).

  And if the traffic information estimation apparatus 1 acquires VICS information and probe information, the input information processing part 16 will produce | generate the speed information of each road link from VICS information and probe information (step S2). However, the road links for which speed information can be acquired in step S2 are a part of all road links in the target area, and there are road links for which speed information cannot be obtained in step S2.

  Subsequently, the speed information obtained in step S2 is set in the estimation database 13, and the estimation database DB4 is updated (step S3). Then, the estimation engine 11 estimates speed information about the road link for which speed information is not obtained based on the content of the estimation database 13 updated in step S3, and sets the estimated speed information in the estimation database 13. Then, the estimation database 13 is updated (step S4). By this step S4, speed information (a mixture of measured values and estimated values) for all road links in the target area is obtained.

The speed information for all road links obtained in step S4 is output to the outside of the device (step S5). Specifically, it is displayed on the display of the device 1 or output as information to be provided to the in-vehicle device 2.
A part of the content of the estimation database 13 updated in step S4 is accumulated as a snapshot in the learning database 14 and used as learning data by the learning engine 12 (step S6).

  The processes in steps S2 to S6 are repeatedly executed every time speed information is generated. The input information processing unit 16 acquires the VICS information and the probe information in accordance with the update period (for example, 5 minutes) of the VICS information, and generates the speed information. It is repeatedly executed in accordance with the update cycle (for example, 5 minutes).

[2.5.2 Model for speed information estimation]
FIG. 7 shows a neural network for the estimation engine 11 to estimate speed information of the estimation target road link. The illustrated neural network generates an output value y by multiplying each of N input signals x _i (i: 1 to N) having values of 0 to 1 by weights w _i (w _i : 0 to 1). It is configured as a simple perceptron.
Here, the input signal x _i is speed information of a road link other than the estimation target road link (a road link connected to the estimation target road link), and the output value y is speed information of the estimation target road link. .

Further, the weight w _i is a value indicating how much the speed information of each of the plurality of road links other than the estimation target road link is reflected, and the road link having a high correlation with the estimation target road link (for example, A road link that constitutes the same road as the estimation target road link and is adjacent to the estimation target road link) should be set to a larger value, and a road link having a lower correlation should be set to a smaller value.

However, in FIG. 7, unlike a general simple perceptron, an independent parameter w ₀ that is added to a node without being multiplied by the input signal x _i is provided. This independent parameter w ₀ expresses a factor (for example, a difference in speed limit between road links) that a factor other than speed information in a road link other than the estimation target road link gives to the speed information of the estimation target road link. Speed information can be estimated with high accuracy. Further, when learning the estimation parameter (weight), the estimation parameter is easily converged to the optimum value, and the learning process can be performed easily or at high speed.

  As described above, in order for the estimation engine 11 to obtain the speed information of a certain estimation target road link, the speed information of another road link whose correlation with the estimation target road link is recognized to some extent, and the other road It is only necessary to obtain a weight indicating how much the link speed information is reflected in the speed information of the estimation target road link. The speed information of such other road links is accumulated in the estimation database (traffic information database) 13, the weights are accumulated in the weight database 15, and the estimation engine 11 obtains necessary information from both the databases 13 and 15. get.

  Note that the neural network for estimating the velocity information is not limited to a single-layer perceptron as shown in FIG. 8, and may be a multilayer perceptron having an input layer, an intermediate layer, and an output layer. When configured as a multilayer perceptron, the speed information of the estimation target road link can be appropriately obtained even when the speed information has a non-linear relationship between the estimation target road link and another road.

[2.5.3 Details of estimation processing]
Here, a road link connected as shown in FIG. 8 is assumed. In FIG. 8, Vx indicates road link speed information (value of 0 to 1). The subscript x in Vx indicates the link number of the road link and takes a value of 1 to 24. That is, in FIG. 8, there are 24 road links with link numbers 1 to 24. Moreover, the arrow direction of each link shows the traveling direction of the vehicle.

In FIG. 8, road links (link numbers x = 1 to 10) indicated by solid-line arrows are links from which VICS information can be acquired, for example, road links corresponding to expressways and main trunk roads. Moreover, the road link (link number x = 11-24) shown by the dotted line arrow is a link from which VICS information cannot be acquired, and is, for example, a general road other than a highway or a main trunk road. A road link indicated by a dotted arrow may be an estimation target road link of speed information. That is, the road link indicated by the dotted arrow is not the estimation target road link when the probe information is obtained, and is the estimation target road link when the probe information is not obtained.
Even if the VICS information can be acquired, if the VICS information cannot be acquired for some reason, it is treated as an estimation target road link.

Hereinafter, the speed information estimation procedure will be described in detail with reference to the road link in FIG. 8 and with reference to FIG. 6 again.
First, initial values are input to the estimation database (traffic information database) 13 and the weight database 15 by the device administrator (step S1).

  As shown in FIG. 9, the estimation database 13 has data items of “check” 31, “speed information” 32, “related road link information” 33, “link length” 34, and “learning?” The value of each data item can be stored for each road link (link number).

Among the data items, “check” 31 is an item indicating whether or not the speed information of each road link is updated and the order of update. “Speed information” 32 is an item in which speed information of each road link is set.
The “related road link information” 33 indicates a road link (related road link) correlated with each road link, and here indicates a road link (link number) connected to each road link. .

The “related road link information” 33 is divided into five types of “reverse”, “A order”, “A reverse”, “B order”, and “B reverse”.
“Reverse” indicates a link number of a road link opposite to an arbitrary road link. For example, for a road link with link number 1, the road link with link number 8 becomes a “reverse” road link.
“A order” is a road link connected in the forward direction behind a certain road link. For example, for the road link with link number 1, the road links with link numbers 12, 7, and 15 are “A order”. It becomes the road link.
“A reverse” is a road link connected in the reverse direction behind a certain road link. For example, for the road link with link number 1, the road links with link numbers 11, 15, and 13 are “A reverse”. It becomes the road link.

“B order” is a road link connected in the forward direction in front of a certain road link. For example, for the road link with link number 1, the road links with link numbers 4 and 9 are “B order”. It becomes a link.
“B reverse” is a road link connected in the reverse direction in front of a certain road link. For example, for the road link with link number 1, the road links with link numbers 3 and 6 are “B reverse”. It becomes a link.

  Of the data items, “link length” 34 indicates the link length of each road link. “Learning?” 35 indicates whether or not the speed information of each road link is data (learning data) for learning by the learning unit 20 (learning engine 12). Takes a value of 1. “0” indicates that the speed information of the road link is not the learning data, and “1” indicates that the speed information of the road link is the learning data.

  The initial value input in step S1 is performed when the operation of the apparatus 1 is started or reset. Regarding the estimation database 11, among the above data items, “speed information” 32, “related road link information” 33, “link length”. An initial value is set for 34. As an initial value of the “speed information” 32, for example, 0 may be set for all road links. The initial values of “related road link information” 33 and “link length” 34 are set according to the road configuration of the target area. The initial values of “related road link information” 33 and “link length” 34 are not updated even if traffic information (speed information) including VICS information and probe information is acquired.

On the other hand, the value of “speed information” 34 is updated for all road links every time traffic information (speed information) composed of VICS information and probe information is acquired (steps S3 and S4).
The “check” 31 is initialized to 0 each time the traffic information (speed information) is acquired (step S2). “Learning?” 35 is also set to 0 or 1 every time traffic information (speed information) is acquired (step S2), depending on whether or not the speed information of each road link should be learned data. The

As shown in FIG. 10, in the weight database 15, a weight wi used when an estimated value of speed information for each road link is obtained from speed information of other road links is set as an initial value. The weight wi is set to “number of related road links + 1” for each road link. In FIG. 10, w ₀ is an independent parameter, and w _{1 and} subsequent weights are weights to be multiplied by the related road link speed information.

  Since the weight wi that is an estimation parameter is automatically updated to a more appropriate value by learning by the learning engine 12, an appropriate value may be set as an initial value. Therefore, initial setting is easy.

  Now, after performing the above initialization, it is assumed that VICS information and probe information (traffic information) as shown in FIG. 11 have been obtained at a certain point in time for the road link shown in FIG. In FIG. 11, VICS information (link travel time) for road links with link numbers 1 to 10 is obtained, and probe information for road links with link number 20 is obtained. Neither VICS information nor probe information is obtained for other road links. In FIG. 11, probe information about the road link with the link number 20 is indicated by “link travel time” in the same manner as the VICS information for easy notation.

  Further, in the information of FIG. 11, a value “1” indicating that the road link (link number 20) where the probe information exists is set as a learning target is set, and the road link (link number) where the VICS information exists. As for 1 to 10), a value “0” indicating that it is not a learning target is set.

  When the input information as shown in FIG. 11 is acquired, the “travel time” in this information is converted into “speed information” by the input information processing unit 16 (step S2), and the correspondence of the estimation database 13 is determined by the speed information. The “speed information” 32 of the road link to be updated is updated (first update; step S3). For the updated road link, “1” indicating that the speed information has been updated in the first update is set in “check” 31. For the road link whose “learning target” of the input information in FIG. 11 is “1” (here, the road link of link number 20), “learning?” 35 is set to “1”.

  FIG. 12 shows the contents of the estimation database 13 after the first update (set of input information) is performed as described above.

Subsequently, the speed information about the road links (link numbers 11 to 19 and 21 to 24) whose speed information is not updated is estimated, and the estimation database 13 is updated with the estimated value (step S4).
Specifically, first, among the related road links (adjacent road links) of road links (link numbers 1 to 10, 20) in which “1” indicating the previous update is set in the “check” 31 item, A road link (check = 0) for which speed information has not yet been updated based on the current input information (FIG. 11) is extracted. Here, 11 road links of link numbers 12, 15, 11, 13, 19, 18, 17, 21, 21, 24, 14 and 16 are extracted (see FIG. 13). These 11 road links serve as estimation target road links here.

  Then, the estimation engine 11 reads the related road links for each of the 11 estimation target road links from the estimation database 13 and uses the weights (estimation parameters) set for each of the 11 estimation target road links as the weights. Reading from the database 15 and using these, the estimated value of the speed information of each estimation target road link is calculated. The calculated estimated value of the speed information is set in “speed information” of the estimation database 13, and the second update of the speed information is performed. For the updated road link, “2” indicating that the speed information has been updated in the second update is set in “check” 31.

Among the estimation target road links, for example, an arithmetic expression for obtaining an estimated value (V ₁₂ , V ₁₅ , V ₁₁ , V ₁₃ ) of speed information of road links of link numbers ₁₂ , ₁₅ , ₁₁ , and ₁₃ is as follows. It is as follows.
V ₁₂ = w ₀ + w ₁ V ₁₁ + w ₂ V ₁ + w ₃ V ₅ + w ₄ V ₁₃ + w ₅ V ₈ + w ₆ V ₇ + w ₇ V ₁₅
V ₁₅ = w ₀ + w ₁ V ₁₃ + w ₂ V ₂₀ + w ₃ V ₁₆ + w ₄ V ₁₇ + w ₅ V ₁₄ + w ₆ V ₁ + w ₇ V ₁₁
+ W ₈ V ₅ + w ₉ V ₈ + w ₁₀ V ₁₂ + w ₁₁ V ₅
V ₁₁ = w ₀ + w ₁ V ₁₂ + w ₂ V ₈ + w ₃ V ₁₅ + w ₄ V ₇ + w ₅ V ₁ + w ₆ V ₁₁ + w ₇ V _{5
V 13 = w 0 + w 1} V 15 + w 2 V 8 + w 3 V 12 + w 4 V 7 + w 5 V 1 + w 6 V 11 + w 7 V 5
+ W ₈ V ₁₄ + w ₉ V ₁₇ + w ₁₀ V ₁₆ + w ₁₁ V ₂₀

Note that the estimated value (value on the left side) obtained by the arithmetic expression is speed information at time t (time slot including time t) at which VICS information and probe information as shown in FIG. 11 are obtained. On the other hand, the speed information and the weight (value on the right side) used for obtaining the estimated value by the arithmetic expression are the time t-1 (the previous time) before the time t (the time slot including the time t). Slot). That is, when the calculation formula for obtaining the estimated value of the speed information is described including the time, the estimated value of the speed information V ₁₂ of the road link of link number 12 is represented by the following calculation formula.
_{V12 (t)} = _{w0 (t-1,12)} + _{w1 (t-1,12) V11 (t-1)} + _{w2 (t-1,12)} V1 _(t-1)
+ W _{3 (t-1,12)} V _{5 (t-1)} + w _{4 (t-1,12)} V _{13 (t-1)} + w _{5 (t-1,12)} V _{8 (t-1)} + w ₆ (T-1,12) V7 _(t-1) + _{w7 (t-1,12) V15 (t-1)}
Note that the subscript _{(t−1, 12)} “12” attached to the weight in this arithmetic expression means a link number. The time slot may be the update unit time of VICS information.

  And among the related road links (adjacent road links) of the road links (link numbers 11 to 19, 21, 24) in which “2” indicating the previous update is set in the “check” 31 item, this input A road link (check = 0) for which speed information has not yet been updated based on the information (FIG. 11) is extracted. Here, two road links with link numbers 22 and 23 are extracted (see FIG. 14). These two road links become the next estimation target road links.

  Then, the estimation engine 11 reads the related road links for each of the two estimation target road links from the estimation database 13 and uses the weights (estimation parameters) set for each of the two estimation target road links as the weights. Reading from the database 15 and using these, the estimated value of the speed information of each estimation target road link is calculated. The calculated estimated value of the speed information is set in “speed information” of the estimation database 13, and the third update of the speed information is performed. As for the updated road link, “3” indicating that the speed information has been updated in the third update is set in “check” 31.

  The above process is repeated until the speed information for all road links is estimated (step S4). Here, since all speed information has been complemented by three updates, the estimation process is terminated.

Then, the contents of the estimation database are output (step S5).
Furthermore, among the road link data of the estimation database 13 for which the estimation process has been completed based on the current input information (FIG. 11), the road link of the link number 20 with “1” set to “Learning?” 35. A snapshot (see FIG. 15) is added to the learning database 14. Here, the snapshot stores “speed information” of a road link having a link number for which “1” is set in “learn?” 35 and “speed information” of a related road link of the road link. It is. Here, the speed information of the road link having the link number for which “1” is set in “Learning?” 35 is a measured value because it is a value obtained from the probe information. In addition, in the “speed information” of the related road link of the road link, an actual measurement value and an estimated value obtained from VICS information are mixed.

A snapshot occurs every time input information is acquired and all road links in the estimation database are updated (steps S3 and S4). This snapshot is generated for a road link for which probe information has been acquired.
Therefore, in the above example, the snapshot is generated only for the road link with the link number 20, but if all the road links in the estimation database are updated many times (steps S3 and S4), the other road links are also snapped. Shots are accumulated. When sufficient time has passed for the input information to be returned to the city, a plurality of snapshots are accumulated for one road link.

FIG. 16 shows the contents of the learning database 14 in which a plurality of snapshots are accumulated for a plurality of road links.
Based on the learning database 14 in which a large number of snapshots are accumulated in this way, the learning engine 12 learns (optimizes) the weights (estimation parameters) stored in the weight database 15 to obtain the weights. The contents of the database 15 are updated.

FIG. 17 shows the procedure of the learning process by the learning engine 12. First, the learning engine 12 sets parameters for learning, such as an allowable error for error determination, the number of intermediate layers constituting the neural network (FIG. 7), and a learning count (step S11).
The learning engine 12 is configured as a neuro engine that optimizes (learns) a neural network as shown in FIG. That is, the learning engine 12 uses the speed information that is an actual measurement value among the snapshots stored in the learning database 14 as a teacher signal, and the speed information about the related road link of the road link having the speed information that becomes the teacher signal. Then, an appropriate neural network for outputting the teacher signal is reconstructed. That is, the learning engine 12 calculates the optimum value of the weight from the learning data including the speed information that is the actually measured value and the speed information about the related road link having the speed information. The calculation of the optimum value of the weight can be performed by an operation for minimizing the residual sum of squares, that is, a method of least squares, in addition to a method using a neural network.

  The optimum weight value is calculated until the speed information calculated from the speed information about the related road link approaches the teacher signal and the error from the teacher signal is less than the set allowable error (step S13). ).

  When the weights converge and the learning process ends, the obtained weights are reflected in the weight database 15 and the weight database 15 is updated.

  When input information composed of VICS information and probe information is generated after the weight database 15 is updated, the estimation of the speed information by the estimation engine 11 is performed with higher accuracy using the new weight. Thus, in the traffic information estimation apparatus 1 of this embodiment, the estimation accuracy of traffic information (speed information) naturally improves by continuing operation.

  As described above, the speed information estimation method executed by the traffic information estimation apparatus 1 according to the embodiment described above is speed information based on actual measurement values for a predetermined road link that the traffic information estimation apparatus 1 can be an estimation target road link. (Steps S2 and S3 in FIG. 6), the estimation engine 11 estimates the speed information of the estimation target road link for which speed information based on the actual measurement values could not be acquired (Step S4), and then the estimated speed. The learning engine 12 optimizes the parameter for estimation using the speed information based on the information and the actually measured value (step S6 in FIG. 6, FIG. 17). FIG. 19 is an explanatory diagram collectively explaining the processes performed by the estimation engine 11 and the learning engine 12.

That is, as described with reference to FIG. 11, speed information V ₂₀ based on actual measurement values is acquired for a predetermined road link (link number 20) that can be an estimation target road link at time t in FIG. When the estimated engine 11, as described in FIG. 14 from FIG. 12, at time t, the traffic information (V ₁₁ ~V ₁₉ estimation target road link (link number 11 to 19 and link ID 21 to 24), V _{21 to} V ₂₄ ) using the speed information (V _{1 to} V ₂₄ ) and the estimation parameters (weight wi) of the road links (link numbers 1 to 24) including the speed information V ₂₀ based on the actual measurement values one after another. Estimated.
Of the road link speed information (V _{1 to} V ₂₄ ) used in this estimation process, the road link speed information of link numbers 1 to 19 and link numbers 21 to 24 is the time shown in FIG. This is information at t-1. That is, the speed information based on the actual measurement value is acquired at time t, but the estimation process is executed based on the speed information at time t and the speed information at time t-1 before that. .

Then, after this estimation, the learning engine 12 estimates the speed information V ₂₀ based on the actual measurement value in the predetermined road link (link number 20) and the speed information of the road link (link number 20). Combination of the speed information (V ₁₇ , V ₂₁ , V ₂₄ , V ₁₄ , V ₁₅ , V ₁₆ , V ₁₃ ) estimated using the actual measurement value V ₂₀ , which is the speed information of the related road link used ( 16) is accumulated as learning data, and the process of optimizing the estimation parameters (V _{0 to} V ₇ ) is executed using the learning data as described in FIG. As shown in FIG. 19B, this optimization process (learning process) is performed based on speed information at the same time t.

  According to the first embodiment, when speed information based on actual measurement values is acquired for a predetermined road link that can be an estimation target road link, the speed information of the estimation target road link is estimated, and then the estimated speed The estimation parameter (weight) is optimized using the information. Thereafter, for the speed information at time t + 1 thereafter, the estimation engine 11 uses the new estimation parameter (weight) to determine the accuracy. The estimation process is often performed.

[Second Embodiment]
Further, according to the traffic information estimation apparatus 1 having the configuration shown in FIG. 2, when speed information based on actual measurement values is acquired for a predetermined road link that can be an estimation target road link, it is opposite to the above embodiment. Further, after the learning engine 12 optimizes the estimation parameter using the speed information based on the actual measurement value, the estimation engine 11 cannot acquire the speed information based on the actual measurement value using the estimation parameter. The speed information of the estimation target road link can be estimated. This case will be described as a second embodiment.

[2.6 Processing contents by the traffic information estimation device 1]
FIG. 20 shows a simplified road link model to explain the second embodiment. This road link structure is composed of six road links “V ₁ ”, “V ₂ ”, “V ₃ ”, “V ₄ ”, “V ₅ ”, “V ₆ ”, and the subscript x of Vx indicates the link number. ing. The speed information of the six road links is also expressed by “V ₁ ”, “V ₂ ”, “V ₃ ”, “V ₄ ”, “V ₅ ”, and “V ₆ ”, respectively.

The neural network used for estimating the speed information of the estimation target road link by the estimation engine 11 included in the traffic information estimation apparatus 1 of the present embodiment is the same as that shown in FIG. The learning engine 12 performs a process of optimizing (learning) the neural network using learning data.
Calculation formulas for obtaining estimated values “V ₁ ”, “V ₂ ”, “V ₃ ”, “V ₄ ”, “V ₅ ”, and “V ₆ ” of the speed information of the estimation target road links of link numbers 1 to 6 in FIG. Is as follows.

V1 _(t) = w0 _(t-1,1) + w1 _(t-1,1) V1 _(t-1) + w2 _(t-1,1) V2 _(t-1)
V2 _(t) = w0 _(t-1,2) + w1 _(t-1,2) V1 _(t-1) + w2 _(t-1,2) V2 _(t-1)
+ W _{3 (t-1,2)} V _{3 (t-1)}
V3 _(t) = w0 _(t-1,3) + w1 _(t-1,3) V2 _(t-1) + w2 _(t-1,3) V3 _(t-1)
+ W _{3 (t-1,3)} V _{4 (t-1)}
V4 _(t) = w0 _(t-1,4) + w1 _(t-1,4) V3 _(t-1) + w2 _(t-1,4) V4 _(t-1)
+ W _{3 (t-1,4)} V _{5 (t-1)}
V5 _(t) = w0 _(t-1,5) + w1 _{(t-1,5) V4 (t-1)} + w2 _{(t-1,5) V5 (t-1)}
+ W _{3 (t-1,5)} V _{6 (t-1)}
V6 _(t) = w0 _(t-1,6) + _{w1 (t-1,6) V5 (t-1)} + _{w2 (t-1,6) V6 (t-1)}

Note that the estimated value (value on the left side) obtained by the arithmetic expression is a value at time t (time slot including time t) at which speed information based on the actually measured value is obtained. On the other hand, the speed information and the weight (value on the right side) used for obtaining the estimated value by the arithmetic expression are the time t-1 (the previous time) before the time t (the time slot including the time t). Slot).
Note that “x” in the subscript _{(t−1, x)} attached to the weight in the arithmetic expression means a link number.

For example, when the estimation target road link is the road link with the link number 2, according to the calculation formula, the road link used to estimate the speed information V ₂ of the road link at time t (that is, the related road) speed information of links), in addition to the speed information V ₃ of the road link having the link number 3 in the velocity information V ₁ and time t-1 of the road link having the link number 1 at time t-1, link number at time t-1 2 road link speed information V ₂ is included.
That is, the estimation target road link is the road link with the link number 2, and the road link with the link number 2 is also included in the related road link. For this reason, the speed information V ₂ of the road link included in the related road link is the speed information at the time t−1 that has already been acquired and is accumulated in the estimation database.

As described above, the estimation engine 11 also uses the speed information (V _{2 (t-1)} ) of the estimation target road link (the road link with the link number 2) acquired at the time t−1, at the time t−1. The speed information of the estimation target road link (the road link with the link number 2) is estimated.
In the same road link (for example, the road link with the link number 2), it is considered that a correlation is recognized in the speed information even if the times are slightly different. Therefore, the estimation engine 11 performs the estimation at the time t-1 that has already been acquired. If the speed information (V _{2 (t-1)} ) of the target road link (road link of link number 2) is also used to estimate the speed information of the estimation target road link (road link of link number 2) at time t, Speed information can be estimated more appropriately.

  Thus, the related road link may include its own road link that becomes the estimation target road link. In the embodiment (in the case of FIGS. 8 and 9), the related road link may include its own road link.

  And in this 2nd embodiment, when speed information based on an actual measurement value about a predetermined road link which can become an estimation object road link is acquired, the learning engine 12 first, contrary to the first embodiment, In this case, the learning engine 12 determines the speed information based on the actual measurement value of the predetermined road link and the predetermined road link at a time prior to the acquisition of the actual measurement value. Learning data that combines speed information and speed information of other road links used for estimating speed information of the predetermined road link (speed information at a time prior to the acquisition of the actual measurement value) Is used to optimize the estimation parameters. This specific example will be described later with reference to FIG.

  For the same road link, it is considered that there is a correlation in the speed information even if the time is slightly different. By including “the speed information of the predetermined road link at the time before the acquisition of the actual measurement value” and learning the estimation parameter using this learning data, a more appropriate estimation parameter can be obtained. It is done.

FIG. 21 is an explanatory diagram illustrating processing by the estimation engine 11 and the learning engine 12 in the second embodiment. The speed information estimation method executed by the traffic information estimation apparatus 1 according to the second embodiment is similar to the first embodiment in that the speed information of the estimation target road link is obtained based on the speed information of other road links. Although it is executed by estimation, in the present embodiment, based on the actual measurement value (probe information) with respect to a predetermined road link (link number 4) that can be the estimation target road link at time t in FIG. When the speed information V ₄ is acquired, the estimation process is executed after the learning process is executed first.

That is, when the speed information V ₄ based on the actual measurement value is acquired for the predetermined road link (link number 4) at time t in FIG. 21A, the learning engine 12 determines the speed information V ₄ based on the actual measurement value. And the speed information (V ₃ , V ₅ ) of the related road link (link numbers 3 and 5) at time t−1, which is used to estimate the speed information of the predetermined road link (link number 4). The combination is used as learning data to optimize the estimation parameters.
Furthermore, in this embodiment, the speed information V ₄ of the road link (link number 4) at the time t−1 is also included in the combination. The speed information V _{4 at the} time t−1 is speed information already accumulated in the estimation database 13 (see FIG. 2).
In the second embodiment, as shown in FIG. 21A, speed information based on the actual measurement value is acquired at time t, and the learning process includes speed information based on the actual measurement value at time t, It is executed based on the speed information at time t-1 before that.

Thereafter, as shown in FIG. 21 (b), using the road link speed information (V _{1 to} V ₆ ) including the speed information V ₄ based on the actually measured value and the optimized estimation parameters, The speed information (V _{1 to} V ₃ , V ₅ , V ₆ ) of the estimation target road link at time t is estimated. This “optimized estimation parameter” is optimized using the learning data combined with the speed information V ₄ based on the actually measured values in FIG.
Of the road link speed information (V _{1 to} V ₆ ), the road link speed information of link numbers 1 to 3 and link numbers 5 and 6 is information at time t-1. In this embodiment, as described in the calculation formula, the speed information of the estimation target road link is estimated including the speed information V ₄ of the own road link at time t−1.
This estimation process is executed based on the speed information based on the actually measured value at time t and the speed information at time t-1 prior to that.

Specific examples will be described below.
FIG. 22 is a flowchart for explaining the second embodiment.
With respect to the road link shown in FIG. 20, it is assumed that probe information is obtained for the road links of link number 1 and link number 5 at a certain time (time t) (step S21). These pieces of information are converted into “speed information” by the input information processing unit 16 (see FIG. 2) (step S22), and by this speed information, as shown in FIG. 23, the corresponding road link in the estimation database is displayed. The data item of “update speed” is updated (step S23). A data item of “update speed” is set in the estimation database, and speed information (V ₁ , V ₅ ) based on the probe information is stored in this data item. Each value (for example, an initial value) is accumulated in the data item of “speed information”.
Further, in FIG. 23, for ease of explanation, the estimation database and the weight database are shown side by side. Each value (for example, initial value) is stored in the weight database.

In the data item of “learning” in the estimation database, a value “1” indicating that the road link (link number 1 and link number 5) where the probe information exists is set as a learning target is set. When the speed information (V ₁ , V ₅ ) based on the actually measured values is acquired, as a part of the contents of the estimation database, a snapshot relating to the road link with the link number 1 and the road link with the link number 5 Are stored in the learning database 14 (step S24). This snapshot becomes learning data, and the learning engine 12 learns weights (estimation parameters) (step S25).

  The learning process by the learning engine 12 is the same as the algorithm described in the first embodiment, and the learning data is used to optimize the weights for the road links with the link numbers 1 and 5 using the learning data. Is obtained, the weight database is updated as shown in FIG. 24 (step S26).

Then, the estimation engine 11 uses the content of the weight database updated in step S26, and traffic information (V ₂ , V ₃ , V) of the road link (estimation target road link) for which speed information based on the actual measurement value was not obtained. V _4, V ₆₎ to estimate (step S27).
As shown in FIG. 25, the speed information (V ₂ , V ₃ , V ₄ , V ₆ ) obtained by this estimation process and the speed information (V ₁ , V ₅ ) based on the actual measurement values are used. The estimation database is updated (step S28).

  As described above, when the speed information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the estimation parameter is optimized using the speed information based on the actual measurement value, and then the estimation The speed information of the estimation target road link is estimated using the parameter. Thereafter, the estimation of the speed information by the estimation engine 11 is performed with higher accuracy by using the new weight obtained by the learning engine 12. Therefore, in the traffic information estimation device 1 of the present embodiment, the accuracy of speed information estimation is naturally improved by continuing operation.

[Third embodiment]
[2.7 Processing Contents by Traffic Information Estimation Device 1]
Still another embodiment (third embodiment) of weight learning and speed information estimation by the traffic information estimation device 1 will be described.
Also in the third embodiment, as in the second embodiment, when speed information based on measured values is acquired for a predetermined road link that can be an estimation target road link, the learning process is performed before the estimation process. .

That is, the learning engine 12 uses a combination of the speed information based on the actual measurement value of the predetermined road link and the speed information of the related road link used for estimating the speed information of the predetermined road link as the learning data. To optimize the estimation parameters.
After this processing, the estimation engine 11 uses the learning data in which the road link speed information including the speed information based on the actual measurement value and the learning data in which the speed information based on the actual measurement value is combined. The speed information of the estimation target road link is estimated using the optimized estimation parameter.

  In the third embodiment, after performing the learning process, the estimation process is performed. This learning process includes a “provisional estimation” process executed based on the neural network or the like used in the second embodiment. included. Hereinafter, the learning process including the temporary estimation and the estimation process will be specifically described.

FIG. 26 is a flowchart for explaining the second embodiment.
With respect to the road link shown in FIG. 20, it is assumed that probe information is obtained as input information for the road link of link number 2 at a certain time (time t) (step S31). Furthermore, it is assumed that VICS information is obtained as input information for the road link of link number 5.
These pieces of information are converted into “speed information” by the input information processing unit 16 (see FIG. 2) (step S32), and the “update speed” of the corresponding road link in the estimation database shown in FIG. "Is updated (step S33). Each value (for example, an initial value) is stored in the data item of “speed information”.
Further, in FIG. 27, for easy explanation, an estimation database (main estimation database) and a weight database are shown side by side. Each value (for example, initial value) is stored in the weight database.

In the estimation database, a temporary estimation database (see FIG. 28) is set in a region different from the main estimation database. Both have the same format, and as shown in FIG. 28 (a), a copy of a part of this estimation database is accumulated in the temporary estimation database (step S34).
In the “learning” data item of the temporary estimation database, a value “1” indicating that the road link (link number 2) in which the probe information exists is a learning target is set. In the temporary estimation database, the data item of “check” indicating the order of the estimation process (temporary estimation) is once reset (set to “0”), and the road link of link number 5 for which VICS information is obtained “1” is set.

  A road link related to a road link (link number 5 in the present embodiment) in which “1” is set in the “check” data item, for which speed information is not provisionally estimated yet, is displayed. Extracted (link numbers 4 and 6 in this embodiment), and “2” is added to the “check” data item. Then, “2” is added to the “check” data item, and the speed information of the road links (link numbers 4 and 6) is temporarily estimated (step S35).

  In this way, the speed information is provisionally estimated one after another while the order of the provisional estimation processing is set. As shown in FIG. 28B, the speed information based on the actual measurement value is obtained for the road link with the link number 5. For other road links, the speed information obtained by provisional estimation is accumulated in the provisional estimation database (step S36).

Since the speed information (V ₂ ) based on the temporary estimation is acquired for the road link (link number 2) for which the probe information is acquired as input information, as part of the contents of the temporary estimation database accumulated in step S36. The snapshot relating to the road link (link number 2) is accumulated in the learning database 14 (step S37). This snapshot becomes learning data, and the learning engine 12 learns weights (estimation parameters) (step S38).

That is, the snapshot of this embodiment includes speed information (V ₂ ) based on an actual measurement value of a road link (link number 2) from which probe information has been acquired, and speed information of the road link (link number 2) based on provisional estimation. It consists of a combination with (V ₂ ). The speed information (V ₂ ) of the road link (link number 2) based on the temporary estimation is the related road link (link number) used for estimating the speed information of the road link (link number 2) in step S35. ₁ , ₃ ) is a provisionally estimated value based on the speed information (V ₁ , V ₃ ).
Then, the learning engine 12 uses the learning data based on this combination to optimize the estimation parameters.

The learning process by the learning engine 12 includes speed information (V ₂ ) based on an actual measurement value of a road link (link number 2) from which probe information has been acquired, and speed information (V ₂ ) of the road link (link number 2) based on provisional estimation. It is executed so that the square error with ₂ ) is minimized. When the weight for the road link with the link number 2 is optimized by this learning process and the result is obtained, the weight database is updated as shown in FIG. 29 (step S39).
Note that the “speed information” data item of the estimation database (main estimation database) in FIG. 29 is not the provisionally estimated value (FIG. 28B), but the estimation database (main estimation database) shown in FIG. The value accumulated in is stored.

Then, the estimation engine 11 uses the content of the weight database and the value of “speed information” in the estimation database in FIG. 29 updated in step S39, and the road link (for which speed information based on the actual measurement value was not obtained) ( The speed information (V ₁ , V ₃ , V ₄ , V ₆ ) of the estimation target road link) is estimated (step S40). As shown in FIG. 30, the speed information (V ₁ , V ₃ , V ₄ , V ₆ ) obtained by this estimation process and the speed information (V ₂ , V ₅ ) based on the actual measurement values are estimated. The database is updated (step S41).

  As described above, when the speed information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the estimation parameter is optimized using the speed information based on the actual measurement value, and then the estimation information is obtained. The speed information of the estimation target road link is estimated using the parameter. That is, the estimation of the speed information by the estimation engine 11 is performed with higher accuracy using the new weight obtained by the learning engine 12.

The traffic information estimation device 1 according to each of the above embodiments includes the learning engine 12 and the estimation engine 11, and estimates the speed information of the estimation target road link based on the speed information of other road links. Can do. The estimation engine 11 is configured to estimate the traffic information of the estimation target road link using the speed information of other road links and the estimation parameters. Then, the learning engine 12 uses the speed information based on the actual measurement value in the road link that can be the estimation target road link and the value based on the speed information of the related road link used for estimating the speed information of the estimation target road link. Is stored as learning data, and the estimation parameter is optimized using the stored learning data.
According to such a traffic information estimation device 1, estimation parameters are optimized by the learning engine 12, and speed information can be appropriately estimated by the estimation engine 11. And the estimation accuracy of speed information improves naturally by the process by such a learning engine 12 and the estimation engine 11 being repeatedly performed.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the meanings described above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  For example, the road link (related road link) used to estimate the speed information of the estimation target road link is not limited to that connected to the estimation target road link, but is separated from the estimation target road link. It may be a road link (for example, a road link parallel to the estimation target road link) whose speed information changes in time.

  If you want to reflect the time component in the speed information estimation, use road links that change characteristic in time (for example, road links that are congested only in the morning, road links that are mixed only in the night) It can also be included in the link. In this case, the speed information of the road link that changes temporally is commonly used for all the estimation target road links.

  Furthermore, in this embodiment, only the speed information in the same time zone is used as the speed information of other road links used for estimating the speed information of the estimation target road link. The speed information may be used. For example, in the case of a road link far away from the estimation target road link, there is a time delay until the traffic situation on the road link is reflected on the estimation target road link. Therefore, in the case of a road link far away, for example, more appropriate estimation can be performed by using the speed information of one hour ago for estimation.

  Furthermore, the traffic information to be estimated by the estimation engine 11 of the traffic information estimation apparatus 1 according to the present embodiment is not limited to speed information, but may be other traffic information such as traffic jam information in addition to / in place of the speed information. You can also. In this case, the degree of traffic congestion can be calculated by assigning 1 in the speed information to a busy binary value and 0 as a smooth binary value. Further, when the congestion degree is handled as the traffic information to be estimated, the neural network also requires a threshold for determining one of the two values as an estimation parameter. In this case, the threshold value is also a learning target by the learning unit.

  Furthermore, the present specification discloses a plurality of inventions, and each invention is not limited to the description method of the claims, and should be interpreted in the broadest sense according to the technical significance of each invention. It is.

1: Traffic information estimation device, 2: On-vehicle device, 3: Probe vehicle, 4: Roadside communication device, 5: Roadside sensor, 6: VICS center computer, 11: Estimation engine (estimator), 12: Learning engine, 13: Estimation database, 14: learning database, 15: weight database, 16: input information processing unit, 17: VICS information processing unit, 17a: link travel speed calculation unit, 17b: speed information conversion unit, 18: probe information processing unit, 18a: Link travel speed calculation unit, 18b: Speed information conversion unit, 19: Conversion information, 20: Learning unit

Claims (9)

A traffic information estimation device that estimates traffic information of an estimation target road link based on traffic information of another road link,
An estimation unit that estimates traffic information of an estimation target road link using traffic information of other road links and estimation parameters;
A combination of traffic information based on actual measurement values of road links that can become estimation target road links and traffic information of road links used for estimating traffic information of the estimation target road links is stored as learning data, and stored. A learning unit that optimizes the estimation parameter using the learned data that has been determined;
A traffic information estimation device comprising:   The traffic information estimation apparatus according to claim 1, wherein the estimation unit estimates traffic information of the estimation target road link using the traffic information of the estimation target road link already acquired.   The traffic information estimation apparatus according to claim 1, wherein the learning data used by the learning unit includes traffic information already acquired for the road link that can be the estimation target road link. . When the traffic information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the estimation unit uses the traffic information of the road link including the traffic information based on the actual measurement value and the estimation parameter. , Estimate traffic information for the estimated road link,
The learning unit is traffic information based on the actual measurement value of the predetermined road link and road link traffic information used for estimating the traffic information of the predetermined road link. The traffic information estimation apparatus according to any one of claims 1 to 3, wherein the estimation parameter is optimized by using a combination with traffic information estimated by using as the learning data. When the traffic information based on the actual measurement value is acquired for the predetermined road link that can be the estimation target road link, the learning unit estimates the traffic information based on the actual measurement value and the traffic information of the predetermined road link. Using the combination of road link traffic information used for the learning data as the learning data,
The estimation unit is an estimation parameter optimized by the learning unit using traffic information of road links including traffic information based on the actual measurement values and the learning data in which the traffic information based on the actual measurement values is combined. The traffic information estimation device according to any one of claims 1 to 3, wherein the traffic information of the estimation target road link is estimated by using.   The learning unit includes traffic information based on the measured values, traffic information on the predetermined road link temporarily estimated based on traffic information on the road link used to estimate traffic information on the predetermined road link, The traffic information estimation apparatus according to claim 5, wherein the estimation parameter is optimized using a combination of the two as learning data.   A computer program for causing a computer to function as the traffic information estimation device according to any one of claims 1 to 6. A traffic information estimation method for estimating traffic information of an estimation target road link based on traffic information of another road link,
When traffic information based on measured values is acquired for a given road link that can be an estimation target road link,
Estimate the traffic information of the estimation target road link using the traffic information of the road link including the traffic information based on the actual measurement value and the estimation parameter,
Traffic information based on the actual measurement value in the predetermined road link and traffic information of the road link used for estimating the traffic information of the predetermined road link and estimated using the actual measurement value A traffic information estimation method, wherein a combination is used as the learning data to optimize the estimation parameter. A traffic information estimation method for estimating traffic information of an estimation target road link based on traffic information of another road link,
When traffic information based on measured values is acquired for a given road link that can be an estimation target road link,
A combination of traffic information based on the actual measurement value and road link traffic information used for estimating traffic information of the predetermined road link is used as the learning data to optimize the estimation parameter;
The estimation target road link using the traffic information of the road link including the traffic information based on the actual measurement value and the estimation parameter optimized using the learning data in which the traffic information based on the actual measurement value is combined. A traffic information estimation method characterized by estimating traffic information.

Images