WO2019038987A1 - Computer program, travel lane identification device, and travel lane identification system - Google Patents

Abstract

In this invention, a computer is caused to function as: a first acquisition unit for acquiring, for a prescribed interval, speed transition information pertaining to a teaching vehicle having an established travel lane in the prescribed interval; a second acquisition unit for acquiring, for the prescribed interval, speed transition information pertaining to other vehicles having unknown travel lanes in the prescribed interval; a feature value extraction unit for extracting, from the speed transition information of both the teaching vehicle and the other vehicles, a feature value of each of the vehicles; a classification unit for classifying the other vehicles by classifying, into one or more groups, at least the feature values of the other vehicles from among the feature value of the teaching vehicle and the feature values of the other vehicles; and a travel lane identification unit for identifying the travel lanes of the other vehicles included in each of the groups on the basis of the feature value of the teaching vehicle.

Description

Computer program, lane identification device and lane identification system

The present invention relates to a computer program for identifying a travel lane of a vehicle, a travel lane identification device, and a travel lane identification system.
This application claims priority based on Japanese Patent Application No. 2017-161706 filed on Aug. 25, 2017, and incorporates all the contents described in the aforementioned Japanese application.

BACKGROUND ART Conventionally, a traffic information creation device that creates traffic information based on probe information collected from a plurality of vehicles is known (see, for example, Patent Document 1).

The traffic information creation device described in Patent Document 1 creates traffic information such as travel time of each link for each vehicle speed range by classifying probe information for each vehicle speed range and performing statistical processing.

JP, 2009-223551, A

Hiroshi Yarai, "Development of correction technology for surveying by Quasi-Zenith Satellite and efforts to multi-GNSS surveying", 8th Hokkaido Surveying Technical Conference, January 27, 2011

(1) In order to achieve the above object, a computer program according to an embodiment of the present invention acquires speed transition information in a predetermined section of a teacher vehicle which is a vehicle whose traveling lane in the predetermined section is known. An acquisition unit, a second acquisition unit for acquiring speed transition information in the predetermined section of a target vehicle whose traveling lane is unknown in the predetermined section, speed transition information of each of the teacher vehicle and the target vehicle From the feature quantity extraction unit for extracting the feature quantity of each vehicle, and by classifying at least the feature quantity of the target vehicle among the feature quantities of the teacher vehicle and the feature quantity of the target vehicle into one or more groups A classification unit that classifies the target vehicle, and a traveling lane identification unit that specifies a traveling lane of the target vehicle included in each group based on the feature amount of the teacher vehicle; To function.

(10) A travel lane identification device according to another embodiment of the present invention includes: a first acquisition unit that acquires speed transition information in the predetermined section of a teacher vehicle whose traveling lane is a known vehicle in the predetermined section; The second acquisition unit for acquiring speed transition information in the predetermined section of the target vehicle whose traveling lane is unknown in the predetermined section, and the speed transition information of each of the teacher vehicle and the target vehicle The target vehicle is classified by classifying at least the feature amount of the target vehicle among the feature amount of the teacher vehicle and the feature amount of the target vehicle into a group of one or more of a feature amount extraction unit that extracts a feature amount; And a traveling lane identification unit that identifies a traveling lane of the target vehicle included in each group based on the feature amount of the teacher vehicle.

(11) A travel lane identification system according to another embodiment of the present invention includes a travel lane identification device for identifying a travel lane of a target vehicle whose vehicle is an unknown vehicle in a predetermined section, and the travel lane identification device. A vehicle mounted on the target vehicle for transmitting probe information in the predetermined section of the target vehicle, and the travel lane identification device is a teacher vehicle whose traveling lane in the predetermined section is a known vehicle A first acquisition unit for acquiring speed transition information in the predetermined section, and the probe information acquired from the in-vehicle device, and based on the acquired probe information, speed transition information in the predetermined section of the target vehicle A second acquisition unit for acquiring the feature information, and a feature quantity extraction unit for extracting feature quantities of each vehicle from the speed transition information of each vehicle of the teacher vehicle and the target vehicle; A classification unit that classifies the target vehicle by classifying at least the characteristic quantity of the target vehicle among the characteristic quantities of the supervised vehicle and the characteristic quantities of the target vehicle into one or more groups, and the characteristic of the teacher vehicle And a travel lane identification unit that identifies a travel lane of the target vehicle included in each group based on the amount.

It is a figure which shows the structure of the traffic information provision system which concerns on Embodiment 1 of this invention. It is a block diagram which shows the structure of the vehicle-mounted apparatus installed in the object vehicle which concerns on Embodiment 1 of this invention. It is a figure which shows an example of the data structure of the probe information which the probe information generation part produced | generated. It is a block diagram which shows the structure of the vehicle-mounted apparatus installed in the teacher vehicle which concerns on Embodiment 1 of this invention. It is a block diagram which shows the structure of the server which concerns on Embodiment 1 of this invention. It is a figure which shows an example of the information of a road link, a lane link, and an object area contained in a map database. It is a figure which shows the feature-value extracted from the probe information of a target vehicle. It is a figure for demonstrating the clustering process by a clustering part. It is a figure which shows an example of the result of the clustering process by a clustering part. It is a sequence diagram which shows the flow of the process which a traffic information provision system performs. It is a flowchart which shows the detail of a driving lane specific process (S4 of FIG. 10). It is a figure for demonstrating the clustering process by a clustering part. It is a figure for demonstrating the clustering process by a clustering part, and the traveling lane identification process by a traveling lane identification part.

[Problems to be solved by the present disclosure]
According to the conventional traffic information creating device, although it is possible to create traffic information for each vehicle speed range, traffic information can not be created for each lane because a means for specifying a traveling lane of a vehicle is not disclosed.

In the case of congestion at the exit of an expressway, congestion due to falling objects, broken vehicles or accident regulations, the vehicle speed may drop in a specific lane and congestion may occur, but in other lanes no congestion may occur. The traffic situation may be different every time. In such a case, if traffic information can be provided for each lane, the driver (including an automatic driving function) can change lanes in advance, and traffic congestion can be avoided. However, in order to create traffic information for each lane using probe information, it is necessary to specify which lane the vehicle is traveling.

On the other hand, with the development of GPS (Global Positioning System) augmentation technology or GPS complementation technology by quasi-zenith satellites, it is expected that the position accuracy of GPS will be dramatically improved in the future. By using the GPS augmentation technology or the GPS complementation technology, it is possible to identify the traveling position of the vehicle with high accuracy, and it is considered that it becomes easy to identify the traveling lane. However, it will take some time before all the vehicles will be equipped with a receiver that can receive Quasi-Zenith satellite signals and pinpoint the travel position with high accuracy.

For this reason, a vehicle capable of transmitting probe information including highly accurate position information to be able to specify a traveling lane, and a vehicle transmitting probe information including position information of conventional accuracy where it is difficult to specify a traveling lane Will be mixed for a while.

Under such circumstances, in order to create traffic information for each lane, it is desirable to specify a traveling lane even for a vehicle that can transmit only probe information including position information with conventional accuracy.

The present disclosure has been made in view of such circumstances, and is a computer program capable of specifying a traveling lane in which the vehicle travels with respect to a vehicle that can transmit only probe information including position information with conventional accuracy. It is an object of the present invention to provide a lane identification device and a lane identification system.

[Effect of the present disclosure]
According to the present disclosure, it is possible to specify a traveling lane in which the vehicle travels for a vehicle that can transmit only probe information including position information with conventional accuracy.

[Outline of the embodiment of the present invention]
First, the outline of the embodiment of the present invention will be listed and described.
(1) A computer program according to an embodiment of the present invention includes a computer, a first acquisition unit for acquiring speed transition information in a predetermined section of a teacher vehicle which is a vehicle whose traveling lane in a predetermined section is known, and The second acquisition unit for acquiring speed transition information in the predetermined section of the target vehicle whose traveling lane is unknown in the predetermined section, and the speed transition information of each of the teacher vehicle and the target vehicle The target vehicle is classified by classifying at least the feature amount of the target vehicle among the feature amount of the teacher vehicle and the feature amount of the target vehicle into a group of one or more of a feature amount extraction unit that extracts a feature amount It functions as a classification unit to be classified and a traveling lane identification unit that identifies the traveling lane of the target vehicle included in each group based on the feature amount of the teacher vehicle.

According to this configuration, at least the feature quantities of the target vehicle are classified into groups, and the traveling lane of the target vehicle is specified based on the classification result of the feature quantities of the teacher vehicle into the group. Therefore, a target vehicle having a speed transition similar to that of the teacher vehicle is classified into the same group as that of the teacher vehicle, so that the traveling lane of the target vehicle can be specified.

(2) Preferably, the feature quantity of each vehicle extracted by the feature quantity extraction unit includes at least one of an average speed, a maximum speed, a minimum speed, a deceleration point and an acceleration point of the vehicle.

According to this configuration, it is possible to use a feature that can accurately represent the change in speed as the feature of the vehicle. This makes it possible to accurately identify the travel lane of the target vehicle.

(3) Further preferably, the first acquisition unit further acquires type information of the teacher vehicle, and the second acquisition unit further acquires type information of the target vehicle, and the classification unit Classifies at least the feature amount of the target vehicle into one or more groups for each type of the target vehicle, and the traveling lane identification unit is based on the feature amount of the teacher vehicle for each type of the teacher vehicle To identify the travel lane of the target vehicle included in each group.

According to this configuration, it is possible to specify the traveling lane of the target vehicle for each type of vehicle. Therefore, the travel lane of the target vehicle can be accurately identified.

(4) Further, the classification unit may classify the representative feature quantity of each traveling lane of the teacher vehicle and the feature quantity of the target vehicle into one or more groups.

According to this configuration, the feature quantities of the target vehicle can be classified into groups including the representative feature quantities of the teacher vehicle. Therefore, for each group, the target vehicle included in the group can be identified as having traveled in the same lane as the teacher vehicle included in the group.

(5) Further, the classification unit is characterized in that the feature quantities of the teacher vehicle and the features of the target vehicle are subject to the constraint that the feature quantities of the teacher vehicle traveling on the same lane are classified into the same group. The amounts may be classified into one or more groups.

According to this configuration, the feature quantities of the target vehicle can be classified into groups including the feature quantities of the teacher vehicle without the feature quantities of the teacher vehicle traveling on the same lane being classified into a plurality of groups. As a result, for each group, the target vehicle included in the group can be identified as having traveled in the same lane as the teacher vehicle included in the group.

(6) Further, the traveling lane identifying unit is configured to compare the representative feature of each traveling lane of the teacher vehicle with the center of gravity of the feature of the target vehicle belonging to each group. The traffic lane of the target vehicle included in each group may be specified by applying the amount to any group.

According to this configuration, it is possible to specify, for each group, a representative feature amount of the teacher vehicle closest to the center of gravity of the feature amounts of the target vehicles belonging to the group. Thus, the target vehicle included in the group can be identified as having traveled in the same lane as the teacher vehicle.

(7) Further, the classification unit may further classify at least the feature quantities of the target vehicle into one or more groups under the constraint that the distance between the groups is equal to or more than a predetermined threshold.

According to this configuration, groups including similar feature quantities can be integrated. When the difference in traveling speed between traveling lanes is small, there is no significant difference in travel time etc. regardless of which lane the vehicle travels. For this reason, it is not necessary to forcibly specify the traveling lane of the target vehicle, but this can be realized by performing group integration.

(8) Further, when there is a second group in which the distance to the first group is equal to or less than a predetermined threshold, the traveling lane identification unit selects the traveling lane of the target vehicle included in the first group. The travel lanes of the target vehicles included in two groups may be added.

According to this configuration, for the first group and the second group including similar feature quantities, candidate traveling lanes can be identified without uniquely identifying the traveling lanes of the target vehicle.

(9) Further, the traveling lane identification unit classifies the traveling lanes of the target vehicles included in the plurality of groups into a plurality of groups for a plurality of groups in which the distance between the groups is equal to or less than a predetermined distance. It may be identified as any of the driving lanes of the said teacher vehicle.

According to this configuration, for a plurality of groups including similar feature quantities, candidate traveling lanes can be identified without uniquely identifying the traveling lanes of the target vehicle.

(10) A travel lane identification device according to another embodiment of the present invention includes: a first acquisition unit that acquires speed transition information in the predetermined section of a teacher vehicle whose traveling lane in the predetermined section is a known vehicle; The second acquisition unit for acquiring speed transition information in the predetermined section of the target vehicle whose traveling lane is unknown in the predetermined section, and the speed transition information of each of the teacher vehicle and the target vehicle The target vehicle is classified by classifying at least the feature amount of the target vehicle among the feature amount of the teacher vehicle and the feature amount of the target vehicle into a group of one or more of a feature amount extraction unit that extracts a feature amount; And a traveling lane identification unit that identifies a traveling lane of the target vehicle included in each group based on the feature amount of the teacher vehicle.

The traffic lane specifying device includes, as a configuration, a processing unit in which a computer functions according to the above-described computer program. For this reason, the same operation and effect as the above-described computer program can be exhibited.

(11) A travel lane identification system according to another embodiment of the present invention includes a travel lane identification device for identifying a travel lane of a target vehicle whose vehicle is an unknown vehicle in a predetermined section, and the travel lane identification device. A vehicle mounted on the target vehicle for transmitting probe information in the predetermined section of the target vehicle, and the travel lane identification device is a teacher vehicle whose traveling lane in the predetermined section is a known vehicle A first acquisition unit for acquiring speed transition information in the predetermined section, and the probe information acquired from the in-vehicle device, and based on the acquired probe information, speed transition information in the predetermined section of the target vehicle A second acquisition unit for acquiring the feature information, and a feature quantity extraction unit for extracting feature quantities of each vehicle from the speed transition information of each vehicle of the teacher vehicle and the target vehicle; A classification unit that classifies the target vehicle by classifying at least the characteristic quantity of the target vehicle among the characteristic quantities of the supervised vehicle and the characteristic quantities of the target vehicle into one or more groups, and the characteristic of the teacher vehicle And a travel lane identification unit that identifies a travel lane of the target vehicle included in each group based on the amount.

This travel lane specification system includes, as a configuration, a processing unit in which a computer functions by the above-described computer program. For this reason, the same operation and effect as the above-described computer program can be exhibited.

[Details of the Embodiment of the Present Invention]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The embodiments described below all show one preferable specific example of the present invention. Numerical values, shapes, components, arrangement positions and connection forms of components, steps, order of steps, and the like shown in the following embodiments are merely examples, and are not intended to limit the present invention. The invention is specified by the claims. Therefore, among the components in the following embodiments, components that are not described in the independent claim showing the highest concept of the present invention are not necessarily required to achieve the object of the present invention, It is described as constituting a preferred embodiment.

In addition, the same components are denoted by the same reference numerals. Since their functions and names are also the same, their descriptions will be omitted as appropriate.

Embodiment 1
<Structure of Traffic Information Providing System>
FIG. 1 is a diagram showing a configuration of a traffic information provision system according to Embodiment 1 of the present invention. In the first embodiment, a traffic information providing system capable of specifying a traveling lane of a road on which the vehicle travels from the probe information collected from the vehicle and providing traffic information for each traveling lane to the vehicle will be described.

Referring to FIG. 1, traffic information provision system 1 functions as a travel lane identification system, and includes server 2, on-vehicle device 6 installed on target vehicle 5 traveling on road 9, and a teacher vehicle traveling on road 9. And an on-vehicle device 8 installed in the vehicle.

Here, the target vehicle 5 indicates a vehicle whose traveling lane on the road 9 is unknown. That is, a vehicle whose position accuracy of the vehicle detected by the on-vehicle device 6 is not high enough to identify the traveling lane is regarded as the target vehicle 5. Specifically, a vehicle whose traveling position is determined by a conventional GPS receiver is taken as the target vehicle 5.

On the other hand, the teacher vehicle 7 indicates a vehicle whose traveling lane on the road 9 is known. That is, a vehicle whose position accuracy of the vehicle detected by the on-vehicle device 8 is high enough to be able to specify the traveling lane is the teacher vehicle 7. Specifically, a vehicle capable of correcting the traveling position measured by the conventional GPS receiver with the reinforcement signal received from the quasi-zenith satellite is the teacher vehicle 7.

The in-vehicle device 6 generates probe information including at least a traveling position of the target vehicle 5 on which the in-vehicle device 6 is installed and information of a passing time when the traveling position passes, at a predetermined time interval or a predetermined distance interval. The in-vehicle device 6 transmits the generated probe information to the server 2 via the base station 4 and the network 3.

Similar to the on-vehicle device 6, the on-vehicle device 8 has probe information including at least a traveling position of the teacher vehicle 7 on which the on-vehicle device 8 is installed and passing time when passing the traveling position. Generate at distance intervals. In addition, the in-vehicle device 8 transmits the generated probe information to the server 2 via the base station 4 and the network 3.

The in-vehicle devices 6 and 8 may be dedicated devices such as in-vehicle communication devices installed in the target vehicles 5 and 7, respectively, or general-purpose devices such as smartphones carried by passengers of the target vehicles 5 and 7 It may be.

The server 2 is installed, for example, in a traffic control center or the like. The server 2 functions as a travel lane identification device, receives probe information from the in-vehicle devices 6 and 8, and identifies the lane of the road 9 on which the target vehicle 5 travels based on the received probe information.

<Configuration of in-vehicle device installed in target vehicle>
FIG. 2 is a block diagram showing the configuration of the in-vehicle device 6 installed in the target vehicle 5 according to the first embodiment of the present invention.

Referring to FIG. 2, the on-vehicle apparatus 6 includes a GPS receiver 61, a speed sensor 62, an azimuth sensor 63, an acceleration sensor 64, a position detection unit 65, a probe information generation unit 66, and a probe information provision unit. 67, a communication I / F (Interface) unit 68, and a traffic information acquisition unit 69.

The GPS receiver 61 measures the position of the target vehicle 5 based on radio waves received from GPS satellites, and outputs position information. The position information of the target vehicle 5 includes latitude information and longitude information of the target vehicle 5.

The speed sensor 62 measures, for example, the traveling speed of the target vehicle 5 by measuring the number of rotations of the wheels of the target vehicle 5, and outputs the measurement result. However, the method of measuring the traveling speed is not limited to this.

The direction sensor 63 includes, for example, a magnetic sensor or a gyro sensor, measures the direction of the target vehicle 5, and outputs the measurement result.

The acceleration sensor 64 measures the traveling acceleration of the target vehicle 5 according to a capacitance detection method or a piezoresistive method, and outputs the measurement result.

The position detection unit 65 detects the position of the target vehicle 5 based on the outputs of the GPS receiver 61, the speed sensor 62, the direction sensor 63, and the acceleration sensor 64. For example, when the GPS receiver 61 outputs position information, the position detection unit 65 detects the position indicated by the position information as the position of the target vehicle 5. However, there may be a case where the position of the target vehicle 5 can not be determined by the GPS receiver 61 in a place where a radio wave from a GPS satellite is disturbed, such as in a tunnel. In such a case, the position of the target vehicle 5 is interpolated and detected based on the outputs of the speed sensor 62, the direction sensor 63, and the acceleration sensor 64.

The probe information generation unit 66 generates probe information including information on the position of the target vehicle 5 detected by the position detection unit 65.

FIG. 3 is a diagram showing an example of a data structure of probe information generated by the probe information generation unit 66. As shown in FIG. As shown in FIG. 3, the probe information 150 includes in-vehicle apparatus identification information, position information, speed information, time information, and position accuracy information.

The in-vehicle apparatus identification information is information for identifying the in-vehicle apparatus 6 and is information uniquely assigned to the in-vehicle apparatus 6. Vehicle identification information for identifying the target vehicle 5 may be used instead of the in-vehicle apparatus identification information.
The position information is information on the position of the target vehicle 5 detected by the position detection unit 65.
The speed information is information on the traveling speed of the target vehicle 5 measured by the speed sensor 62.

The time information is information on the time when the target vehicle 5 has passed the position of the target vehicle 5 detected by the position detection unit 65.

The position accuracy information is information indicating the accuracy of the position of the target vehicle 5 detected by the position detection unit 65. That is, it is information indicating whether the position information is high enough to identify the traveling lane. Information that is high enough to identify the driving lane is hereinafter referred to as “high accuracy information”, and information that is not high enough to identify the traveling lane is hereinafter referred to as “low accuracy information”. The in-vehicle device 6 measures the position only by the GPS receiver 61. Therefore, the position accuracy information indicates low accuracy information.

The traveling speed of the target vehicle 5 can be calculated from the position and time of the target vehicle 5. Therefore, the velocity information may not be included in the probe information 150.

In addition, when identification of the transmission source of the probe information is unnecessary, the in-vehicle apparatus identification information may not be included.

Further, in the case where the position accuracy information can be specified from the in-vehicle apparatus identification information or the like, the position accuracy information may not necessarily be included in the probe information 150.

In addition, when position accuracy information is not included in the probe information 150, it may be determined that the position accuracy indicated by the probe information 150 is low accuracy that is not high enough to identify the traveling lane.

Referring again to FIG. 2, probe information provision unit 67 transmits the probe information generated by probe information generation unit 66 to server 2 via communication I / F unit 68.

The communication I / F unit 68 is a communication interface for wirelessly transmitting data. For example, the communication I / F unit 68 conforms to a communication standard such as 3G or LTE (Long Term Evolution), and the in-vehicle device 6 and the base Establish a connection for communication with station 4 The communication I / F unit 68 transmits probe information to the server 2 via the base station 4 and the network 3.

The traffic information acquisition unit 69 acquires traffic information for each lane from the server 2 via the communication I / F unit 68. For example, the traffic information acquisition unit 69 acquires travel time for each lane. The traffic information acquisition unit 69 displays the acquired traffic information on a display device or provides the navigation system with the travel time for each lane to the destination.

<Configuration of in-vehicle device installed in teacher vehicle>
FIG. 4 is a block diagram showing the configuration of the in-vehicle apparatus 8 installed in the teacher vehicle 7 according to the first embodiment of the present invention.

Referring to FIG. 4, the on-vehicle device 8 includes a GPS receiver 61, a speed sensor 62, an azimuth sensor 63, an acceleration sensor 64, a reinforcement signal receiver 81, a camera 82, and a position detection unit 83. A probe information generation unit 84, a probe information provision unit 67, a communication I / F unit 68, and a traffic information acquisition unit 69 are provided.

The processing units 61 to 64 and 67 to 69 having the same reference numerals as those of the in-vehicle apparatus 6 shown in FIG. 2 have the same configuration as those of the in-vehicle apparatus 6. However, the difference is that the vehicle to be processed is not the target vehicle 5 but the teacher vehicle 7.

The reinforcement signal receiver 81 receives the reinforcement signal transmitted by the quasi-zenith satellite. A reinforcement signal is a signal for improving the positioning accuracy by GPS, for example, a LEX signal corresponds to it. However, the L1-SAIF signal may be used instead of the LEX signal. In addition, about the detail of a reinforcement signal, it is disclosed by the nonpatent literature 1, for example.

The camera 82 is installed in the teacher vehicle 7 and captures an image of the surroundings of the teacher vehicle 7. For example, the camera 82 shoots the front of the teacher vehicle 7. The image captured by the camera 82 may be a still image or a moving image.

Note that at least one of the reinforcement signal receiver 81 and the camera 82 may be included in the in-vehicle device 8.

The position detection unit 83 detects a position with a positioning accuracy of about several cm to 1 m by correcting the position of the teacher vehicle 7 measured by the GPS receiver 61 using the reinforcement signal received by the reinforcement signal receiver 81. Do. As a result, the position detection unit 83 can detect a position having an accuracy with which the traveling lane of the teacher vehicle 7 can be identified. However, the position detection unit 83 detects the position of the teacher vehicle 7 based on the outputs of the speed sensor 62, the direction sensor 63, and the acceleration sensor 64 in a tunnel or the like where radio waves from GPS satellites or quasi-zenith satellites are interrupted. Interpolate and detect.

The probe information generation unit 84 generates probe information including information on the position of the teacher vehicle 7 detected by the position detection unit 83 and an image around the teacher vehicle 7 photographed by the camera 82.

The data structure of the probe information of the teacher vehicle 7 generated by the probe information generation unit 84 is the data structure of the probe information of the target vehicle 5 generated by the probe information generation unit 66, an example of which is shown in FIG. It is. However, when the in-vehicle apparatus 8 includes the reinforcement signal receiver 81, the position accuracy information indicates that the information is high accuracy information. On the other hand, when the in-vehicle apparatus 8 does not include the reinforcement signal receiver 81, the position accuracy information indicates that the information is low accuracy information.

The image captured by the camera 82 may be transmitted to the server 2 without being included in the probe information.

<Server configuration>
FIG. 5 is a block diagram showing the configuration of the server 2 according to Embodiment 1 of the present invention.
Referring to FIG. 5, server 2 includes communication I / F unit 21, probe information acquisition unit 22, storage device 23, link matching unit 24, traveling lane estimation unit 25, and feature quantity extraction unit 26. , A clustering unit 27, a travel lane identification unit 28, and a traffic information calculation unit 29.

The communication I / F unit 21 is a communication interface for connecting the server 2 to the network 3. The communication I / F unit 21 receives probe information transmitted by the in-vehicle devices 6 and 8 via the network 3.

The probe information acquisition unit 22 functions as a first acquisition unit and a second acquisition unit, and acquires probe information from the in-vehicle devices 6 and 8 via the communication I / F unit 21. The probe information acquisition unit 22 stores the acquired probe information in the storage device 23.

That is, the probe information acquisition unit 22 causes the storage device 23 to store probe information whose position accuracy information is low accuracy information and does not contain an image as probe information of the target vehicle 5.

In addition, the probe information acquisition unit 22 stores the probe information of which the position accuracy information is high accuracy information as the probe information of the teacher vehicle 7 in the storage device 23. Further, the probe information acquisition unit 22 stores the probe information including the image whose position accuracy information is low accuracy information and including an image in the storage device 23 as probe information of the teacher vehicle 7.

The storage unit 23 stores various information. The storage device 23 is configured of, for example, a magnetic storage device such as a hard disk drive (HDD) or a semiconductor storage device such as a flash memory.

The storage device 23 stores, for example, the probe information 23A and the map database 23B acquired by the probe information acquisition unit 22.

The probe information 23A is probe information acquired by the probe information acquisition unit 22 from the in-vehicle devices 6 and 8 as described above.

The map database 23B is map information including information on road links on which the target vehicle 5 and the teacher vehicle 7 can travel, and information on lane links forming each road link. The map database 23B also includes information of a target section (see FIG. 1) to be determined for the lane in which the target vehicle 5 has traveled.

FIG. 6 is a diagram showing an example of road link, lane link, and target segment information included in the map database 23B. The map database 23B includes information for specifying the road links 101-105. As shown in the figure, the road link 101, the road link 102, and the road link 103 are connected in this order from the upstream, and the road link 103 branches to the road link 104 and the road link 105 at the lower end. The target section S is, for example, a section from the middle of the road link 101 to the end of the road link 102, and is a three-lane road including three lane links 106A to 106C.

Referring back to FIG. 5, the link matching unit 24 estimates the road link on which the target vehicle 5 has traveled, based on the probe information 23A and the map database 23B. That is, the link matching unit 24 estimates the road link on which the target vehicle 5 has traveled by estimating the road link having the position information closest to the position information indicated by the probe information of the target vehicle 5.

The traveling lane estimation unit 25 estimates a road link and a traveling lane (lane link) on which the teacher vehicle 7 has traveled, based on the probe information 23A and the map database 23B. Each road link is composed of a plurality of lane links each corresponding to a traveling lane. Therefore, when the position accuracy information included in the probe information of the teacher vehicle 7 indicates the high accuracy information, the traveling lane estimation unit 25 has a road having the position information closest to the position information indicated by the probe information of the teacher vehicle 7. By estimating links and lane links, road links and lane links on which the teacher vehicle 7 travels are estimated.

On the other hand, when the position accuracy information included in the probe information of the teacher vehicle 7 indicates low accuracy information, the traveling lane estimation unit 25 has a road link having position information closest to the position information indicated by the probe information of the teacher vehicle 7. Estimate In addition, the traveling lane estimation unit 25 estimates the lane link on which the teacher vehicle 7 has traveled by inputting the image included in the probe information to the classifier learned in advance. For example, in the case where the classifier is a neural network, a neural network is previously learned by using, as teacher data, a set of a lane link and an image captured by a vehicle traveling the lane link. The traffic lane estimation unit 25 estimates a lane link on which the teacher vehicle 7 has traveled by inputting an image to a learned neural network.

The feature amount extraction unit 26 specifies the probe information of the target vehicle 5 that has traveled the predetermined target section based on the road link estimated by the link matching unit 24, and extracts the feature amount from the specified probe information. Further, the feature quantity extraction unit 26 specifies the probe information of the teacher vehicle 7 traveling on the target section based on the road link estimated by the traveling lane estimation unit 25 and extracts the feature quantity from the specified probe information. The feature amount includes, for example, an average velocity, a maximum velocity, a minimum velocity, and a deceleration point in the target section. The deceleration point indicates the first position at which the velocity changes from the predetermined velocity Vth or more to less than the predetermined velocity Vth by searching from the upper end of the target section. Note that the feature amount may include an acceleration point instead of or in addition to the deceleration point. The acceleration point indicates the first position at which the velocity changes from less than the predetermined velocity Vth to at least the predetermined velocity Vth by searching from the upper end of the target section.
By using these as feature quantities, the transition of speed can be accurately represented.

FIG. 7 is a diagram showing feature quantities extracted from probe information of the target vehicle 5. When the vehicles A to F are the target vehicles 5, (a) to (f) in FIG. 7 indicate speed transition information in the target sections of the respective target vehicles 5 (vehicles A to F). The horizontal axis indicates the distance from the upper end of the target section, and the vertical axis indicates the traveling speed at each point. The feature quantity extracted by the feature quantity extraction unit 26 from the speed transition information of the target vehicle 5 is shown in a rectangular frame. For example, as shown in (a) in FIG. 7, “average speed = 60, maximum speed = 80, minimum speed = 30, deceleration point = 100” is extracted as the feature amount from the speed transition information of the vehicle A ing. As shown in (c), (d) and (e) in FIG. 7, “deceleration point = 999” is set for a vehicle in which no deceleration across the predetermined speed Vth occurs. Similar feature quantities are extracted for the teacher vehicle 7 as well.

Referring again to FIG. 5, the clustering unit 27 functions as a classification unit, and the feature quantity (hereinafter referred to as “feature quantity of the target vehicle 5”) extracted from the probe information of the target vehicle 5 and of the teacher vehicle 7. A clustering process is performed to classify representative feature quantities obtained from feature quantities (hereinafter referred to as "feature quantities of the teacher vehicle 7") indicated by the probe information into a plurality of clusters (groups). In the first embodiment, it is assumed that the number of lanes (for example, three lanes) in the target section is known.
Therefore, the clustering unit 27 classifies the feature amount of the target vehicle 5 and the representative feature amount of the teacher vehicle 7 into three clusters, for example, using the k-means method. The clustering unit 27 classifies the target vehicle 5 into clusters by classifying the feature quantities of the target vehicle 5 into clusters.

FIG. 8 is a diagram for explaining clustering processing by the clustering unit 27.

As shown in FIG. 8, the feature quantity of the target vehicle 5 can be indicated as a point in the four-dimensional vector space. That is, the feature quantity of the target vehicle 5 is a four-dimensional feature quantity vector consisting of an average speed, a maximum speed, a minimum speed, and a deceleration point. In addition, the representative feature quantity of the teacher vehicle 7 is generated from the feature quantity vector of the teacher vehicle 7 for each traveling lane. For example, the clustering unit 27 generates a representative vector 121 as a representative feature amount from the feature amount (feature amount vector) of the teacher vehicle 7 traveling in the first lane. An average vector (centroid vector) of feature amounts may be used as the representative vector 121. Similarly, the clustering unit 27 generates a representative vector 122 of the teacher vehicle 7 traveling in the second lane and a representative vector 123 of the teacher vehicle 7 traveling in the third lane.

The clustering unit 27 clusters these feature quantity vectors into three clusters. Thus, cluster 1 including feature quantity vector 111 and representative vector 121, cluster 2 including feature quantity vector 112 and representative vector 122, and cluster 3 including feature quantity vector 113 and representative vector 123 are generated. Note that, as preprocessing for clustering, the clustering unit 27 is assumed to perform normalization of feature quantities of each dimension.

FIG. 9 is a diagram showing an example of the result of the clustering process by the clustering unit 27. As shown in FIG.

8 and 9, cluster 1 includes feature quantities of vehicles A, B, and F, cluster 2 includes feature quantities of vehicles C and E, and cluster 3 includes vehicle D. Contains the feature quantities of

The travel lane identification unit 28 identifies the travel lane of the target vehicle 5 included in the cluster from the representative vector included in each cluster.

Specifically, the cluster 1 includes a representative vector 121 of the teacher vehicle 7 traveling in the first lane. For this reason, the traveling lane identification unit 28 identifies the traveling lane of the target vehicle 5 corresponding to the feature quantity vector 111 included in the cluster 1 as the first lane. Thus, the traveling lanes of the vehicles A, B and F are identified as the first lane. Similarly, since the representative vector 122 of the teacher vehicle 7 traveling in the second lane belongs to the cluster 2, the traveling lanes of the vehicles C and E are identified as the second lane. Furthermore, since the representative vector 123 of the teacher vehicle 7 traveling in the third lane belongs to the cluster 3, the traveling lane of the vehicle D is identified as the third lane.

Referring again to FIG. 5, the traffic information calculation unit 29 detects the probe information of the teacher vehicle 7 whose travel lane is estimated by the travel lane estimation unit 25 and the target vehicle 5 whose travel lane is identified by the travel lane identification unit 28. The traffic information for each lane is calculated based on the probe information of For example, the traffic information calculation unit 29 calculates, for each lane, the travel time of the predetermined section and the traffic jam end position. The traffic information calculation unit 29 transmits the calculated traffic information to the in-vehicle devices 6 and 8 via the communication I / F unit 21.

<Flow of processing of traffic information provision system>
FIG. 10 is a sequence diagram showing a flow of processing performed by the traffic information providing system 1.

The on-vehicle apparatus 6 and the on-vehicle apparatus 8 detect the positions of the target vehicle 5 and the teacher vehicle 7, respectively (S1).

The in-vehicle devices 6 and 8 generate probe information based on the detected position (S2).

The in-vehicle devices 6 and 8 transmit the generated probe information to the server 2, and the server 2 receives the probe information from the in-vehicle devices 6 and 8 (S3). The server 2 stores the probe information received from the in-vehicle devices 6 and 8 in the storage device 23.

The server 2 specifies the traveling lane of the target vehicle 5 in the target section based on the probe information received from the in-vehicle devices 6 and 8 (S4).

The server 2 generates traffic information for each lane in the target section, and transmits the generated traffic information to the in-vehicle devices 6 and 8 (S5).

FIG. 11 is a flowchart showing details of the travel lane identification process (S4 in FIG. 10).

The server 2 stands by until the predetermined traffic information generation timing T is reached (S11). For example, the traffic information generation timing T periodically arrives at an interval of 30 minutes.

When the traffic information generation timing is reached (YES in S11), probe information of the teacher vehicle 7 as to whether the teacher vehicle 7 that has passed the most downstream of the target section S exists from time (TC) to timing T It judges from (S12). Here, C is a predetermined time, for example, 15 minutes.

When the teacher vehicle 7 is present (YES in S12), the traveling lane estimation unit 25 determines the most downstream of the target section S from time (TC) to timing T from the probe information of the teacher vehicle 7 The traveling lane of the passing teacher vehicle 7 is estimated (S13).

Also, the link matching unit 24 identifies the road link of the target vehicle 5 that has passed the most downstream of the target section S from time (TC) to timing T (S14).

The feature amount extraction unit 26 extracts a feature amount from the probe information of the target vehicle 5 and the teacher vehicle 7 that have passed the most downstream of the target section S from time (TC) to timing T (S15).

The clustering unit 27 clusters the feature quantity vector of the target vehicle 5 and the representative vector of the teacher vehicle 7 (S16).

The traveling lane identification unit 28 identifies the traveling lane of the target vehicle 5 included in the cluster as the traveling lane of the teacher vehicle 7 corresponding to the representative vector included in the cluster for each cluster (S17).

<Effect of Embodiment 1>
As described above, according to the first embodiment of the present invention, the travel lane of the target vehicle 5 is specified by clustering the feature quantity vector of the target vehicle 5 and the representative vector of the teacher vehicle 7. Therefore, the target vehicle 5 having a speed transition similar to the speed transition of the teacher vehicle 7 is classified into the same cluster as the teacher vehicle 7. Therefore, the target vehicle 5 included in a certain cluster can be identified as having traveled in the same lane as the teacher vehicle 7 included in the cluster. Thus, the travel lane of the target vehicle 5 can be identified.

(Modification 1 of Embodiment 1)
In the first embodiment, the traveling lane identification unit 28 of the server 2 determines a cluster to which the teacher vehicle 7 belongs using the representative vector generated from the feature amount of the teacher vehicle 7, and the traveling lane of the target vehicle 5 for each cluster. It was determined. In this modification, not the representative vector of the teacher vehicle 7 but the feature amount of the teacher vehicle 7 is used to determine the cluster to which the teacher vehicle 7 belongs.

That is, in the clustering unit 27 shown in FIG. 5, the feature quantities of the teacher vehicle 7 traveling in the same lane are the feature quantities of the target vehicle 5 and the features of the teacher vehicle 7 under the constraint that they belong to the same cluster. Cluster without distinction from quantity.

FIG. 12 is a diagram for describing clustering processing by the clustering unit 27.

As shown in FIG. 12, the clustering process is performed such that the feature amount vector 131, the feature amount vector 132 and the feature amount vector 133 of the teacher vehicle 7 each traveling in the same lane are classified into the same cluster. . For example, the feature vector 131 is a feature vector of the teacher vehicle 7 traveling in the first lane, the feature vector 132 is a feature vector of the teacher vehicle 7 traveling in the second lane, and the feature vector 133 Is a feature quantity vector of the teacher vehicle 7 traveling in the third lane.

As a result of the clustering processing, the feature amount vector 111 of the target vehicle 5 and the feature amount vector 131 of the teacher vehicle 7 are classified into the cluster 1. Further, the feature quantity vector 112 of the target vehicle 5 and the feature quantity vector 132 of the teacher vehicle 7 are classified into clusters 2. Furthermore, the feature amount vector 113 of the target vehicle 5 and the feature amount vector 133 of the teacher vehicle 7 are classified into clusters 3.

The feature amount vector 131 is a feature amount vector of the teacher vehicle 7 traveling in the first lane. Therefore, the travel lane identification unit 28 identifies the travel lane of the target vehicle 5 corresponding to the feature quantity vector 111 belonging to the same cluster 1 as the feature quantity vector 131 as the first lane. Similarly, the traveling lane identification unit 28 identifies the traveling lane of the target vehicle 5 belonging to the cluster 2 as the second lane, and identifies the traveling lane of the target vehicle 5 belonging to the cluster 3 as the third lane.

According to this modification, the feature quantities of the target vehicle 5 are classified into clusters including the feature quantities of the teacher vehicle 7 without the feature quantities of the teacher vehicle 7 traveling on the same lane being classified into a plurality of clusters. Can. Thus, for each cluster, the target vehicle 5 included in the cluster can be identified as having traveled in the same lane as the teacher vehicle 7 included in the cluster.

(Modification 2 of Embodiment 1)
In the first embodiment and the first modification of the first embodiment, clustering of feature quantities is performed using the k-means method. Therefore, the feature amount is classified into the same number of clusters as the number of lanes. However, when the traveling speed transition of the vehicle is similar between the lanes, it may be difficult to identify the lanes. For example, in the vicinity of an approach road on a freeway, a vehicle traveling in the first lane closest to the approach road travels at a relatively low speed, but a vehicle traveling in the other second lane and the third lane has a high speed There is a case to travel by. In such a case, the vehicle traveling in the first lane and the vehicle traveling in the second and third lanes can be clearly classified because their speed transition differs. However, when vehicles traveling in the second and third lanes travel at similar speed transitions, it is difficult to classify these vehicles.

So, in this modification, a target vehicle 5 whose speed transition is similar between the lanes like the above-mentioned vehicle traveling in the second and third lanes is a candidate without being uniquely identified the traveling lane. Identify the driving lane.

Specifically, the traveling lane identification unit 28 of the server 2 calculates the distance between the clusters. For example, the traveling lane identification unit 28 calculates the distance between the centers of gravity of the feature quantity vectors of the target vehicle 5 included in the cluster as the distance between the clusters. However, the method of calculating the distance between clusters is not limited to this.

The traffic lane specifying unit 28 specifies a plurality of clusters in which the distance between clusters is equal to or less than a predetermined threshold. For example, in clusters 1 to 3 shown in FIG. 8, the distance between cluster 2 and cluster 3 is equal to or less than a predetermined threshold, and the distance between cluster 1 and cluster 2 and the distance between cluster 1 and cluster 3 are predetermined threshold Suppose that it is larger than.

In this case, the travel lane identification unit 28 does not uniquely identify the travel lane for the target vehicles 5 included in the clusters 2 and 3. However, the cluster 2 includes the representative vector 122 of the teacher vehicle 7 traveling in the second lane, and the cluster 3 includes the representative vector 123 of the teacher vehicle 7 traveling in the third lane. Based on this, the travel lane specifying unit 28 specifies that the target vehicle 5 included in the cluster 2 and the cluster 3 has traveled in one of the second lane and the third lane. That is, the travel lane identification unit 28 identifies the second lane and the third lane as travel lanes of the target vehicle 5 included in the cluster 2 and the cluster 3.

In other words, the traveling lane identification unit 28 is included in the traveling lane of the target vehicle 5 included in the first group in the second group when there is a second group whose distance to the first group is equal to or less than a predetermined threshold. Travel lane of the target vehicle 5 to be In the above example, the first group corresponds to cluster 2 and the second group corresponds to cluster 3.

Second Embodiment
In the first embodiment and its modification, clustering of feature quantities is performed using the k-means method. However, the clustering method is not limited to this. In the second embodiment, an example in which the x-means method is used instead of the k-means method will be described.
The configuration of the traffic information providing system 1 is the same as that of the first embodiment.

The clustering unit 27 illustrated in FIG. 5 classifies the feature quantities of the target vehicle 5 into clusters using the x-means method. Unlike the k-means method, the x-means method changes the number of classified clusters.

For example, under the constraint that the distance between clusters (for example, the distance between center-of-gravity vectors representing the centers of gravity of feature amounts included in clusters) is less than a predetermined threshold, the clustering unit 27 features the features of the target vehicle 5 Categorize into quantities. Thereby, clusters including similar feature quantities can be integrated. When the difference in traveling speed between traveling lanes is small, there is no significant difference in travel time etc. regardless of which lane the vehicle travels. For this reason, it is not necessary to forcibly specify the traveling lane of the target vehicle 5, but this can be realized by performing the above-described cluster integration.

FIG. 13 is a diagram for describing clustering processing by the clustering unit 27 and traveling lane identification processing by the traveling lane identification unit 28.

The clustering unit 27 clusters the feature quantities of the target vehicle 5 using the x-means method, whereby the feature quantity vectors 111 are classified into clusters 1 and the feature quantity vectors 112 into clusters 2 as shown in FIG. 13. It is assumed that the feature amount vector 113 is classified into cluster 3 after classification.

The traveling lane identification unit 28 obtains the center of gravity vector 141 by calculating the center of gravity of the feature amount vector 111 belonging to the cluster 1. Similarly, the traveling lane identification unit 28 obtains the gravity center vector 142 by calculating the gravity center of the feature amount vector 112 belonging to the cluster 2 and calculates the gravity center vector 143 by calculating the gravity center of the feature amount vector 113 belonging to the cluster 3 Ask.

Further, as in the first embodiment, the traveling lane identification unit 28 calculates the representative vector 121 from the feature amount vector of the teacher vehicle 7 traveling the first lane. Further, the traveling lane identification unit 28 calculates a representative vector 122 from the feature amount vector of the teacher vehicle 7 traveling in the second lane, and calculates the representative vector 123 from the feature amount vector of the teacher vehicle 7 traveling in the third lane .

The traveling lane identification unit 28 determines, for each of the representative vectors 121 to 123, a gravity center vector having the closest Euclidean distance from the representative vector from among the gravity center vectors 141 to 143. Thus, the cluster to which the representative vectors 121 to 123 belong is determined. For example, the centroid vector closest to the representative vector 121 is the centroid vector 141. Therefore, the traveling lane identification unit 28 determines that the cluster to which the representative vector 121 belongs is the cluster 1 to which the feature amount vector 111 which is the source of the calculation of the gravity center vector 141 belongs. Thus, the travel lane identification unit 28 identifies that the target vehicle 5 included in the cluster 1 has traveled the same first lane as the teacher vehicle 7 corresponding to the representative vector 121. Similarly, the travel lane identification unit 28 identifies the travel lane of the target vehicle 5 included in the cluster 2 as the second lane, and identifies the travel lane of the target vehicle 5 included in the cluster 3 as the third lane.

As described above, according to the second embodiment of the present invention, it is possible to identify, for each cluster, the representative feature of the teacher vehicle 7 closest to the center of gravity of the feature of the target vehicle 5 belonging to the cluster. Accordingly, the target vehicle 5 included in the cluster can be identified as having traveled in the same lane as the teacher vehicle 7.

(Modification)
In the above-described first and second embodiments, the travel lane of the target vehicle 5 is specified without distinguishing the type of vehicle. However, even for a vehicle traveling on the same lane, the traveling speed is different between a large vehicle such as a truck or a bus and a small vehicle such as a passenger car. In addition, motorcycles such as motorcycles run differently from four-wheeled vehicles by traveling on the side roads. For this reason, target vehicles 5 which originally travel in the same lane are classified into different clusters, and it is determined that they travel in different lanes, or target vehicles 5 which travel in different lanes originally It may be determined that the vehicles are classified into the same cluster and travel in the same lane.

In this modification, the travel lane of the target vehicle 5 is specified by performing clustering processing for each type of vehicle.

In this modification, it is assumed that the type information of the vehicle is included in the probe information that the target vehicle 5 and the teacher vehicle 7 transmit to the server 2. The type information includes, for example, information on large vehicles, small vehicles and motorcycles. The type information may be further subdivided.

The clustering unit 27 performs clustering processing of probe information for each type of vehicle. The clustering process is the same as that shown in the first and second embodiments except that it is performed for each type of vehicle.

Further, the traveling lane identification unit 28 identifies the traveling lane of the target vehicle 5 from the result of the clustering for each type of vehicle. The lane identification processing is similar to that shown in the first and second embodiments except that it is performed for each type of vehicle.

According to this modification, the travel lane of the target vehicle 5 can be specified for each type of vehicle. Therefore, the travel lane of the target vehicle 5 can be identified accurately.

[Supplementary note]
In the above-described embodiment, the server 2 estimates the road link on which the target vehicle 5 travels, and estimates the road link and lane link on which the teacher vehicle 7 travels. The on-vehicle apparatus 6 of 5 or the on-vehicle apparatus 8 of the teacher vehicle 7 may perform.

For example, when the in-vehicle device 6 includes the map database 23B shown in FIG. 5 and the link matching unit 24, the link matching unit 24 of the in-vehicle device 6 links the probe information and the map database 23B. A matching process is performed to estimate a road link on which the target vehicle 5 has traveled. The in-vehicle device 6 transmits the information of the estimated road link to the server 2.

When the on-vehicle apparatus 8 includes the map database 23B shown in FIG. 5 and the traveling lane estimation unit 25, the traveling lane estimation unit 25 of the on-vehicle apparatus 8 is based on the probe information and the map database 23B. The road link and the lane link on which the teacher vehicle 7 travels are estimated. The in-vehicle device 8 transmits the information of the estimated road link and lane link to the server 2. When the teacher vehicle 7 has the camera 82, the traveling lane estimation unit 25 of the teacher vehicle 7 may estimate the lane link using a classifier such as the above-described neural network.

Specifically, the above-described server and in-vehicle apparatus are configured as a computer system including a microprocessor, a read only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), a display unit, and the like. It is also good. A computer program is stored in the RAM or the HDD. Each device achieves its function by the microprocessor operating according to the computer program. Here, the computer program is configured by combining a plurality of instruction codes indicating instructions to the computer in order to achieve a predetermined function.

Further, some or all of the components constituting each of the above-described devices may be configured as one system LSI. The system LSI is a super-multifunctional LSI manufactured by integrating a plurality of components on one chip, and more specifically, a computer system including a microprocessor, a ROM, a RAM, and the like. . A computer program is stored in the RAM. The system LSI achieves its functions by the microprocessor operating according to the computer program.

That is, in the probe information providing unit 67, the traffic information acquiring unit 69, the probe information generating unit 66, and the position detecting unit 65 of the on-vehicle apparatus 6 in each of the above embodiments, the microprocessor of the on-vehicle apparatus 6 is RAM or HDD. It is realized by executing a stored computer program. That is, the microprocessor of the in-vehicle apparatus 6 functionally includes a probe information providing unit 67, a traffic information acquisition unit 69, a probe information generation unit 66, and a position detection unit 65.
Also, in the probe information providing unit 67, the traffic information acquisition unit 69, the probe information generation unit 84, and the position detection unit 83 of the on-vehicle apparatus 8, the microprocessor of the on-vehicle apparatus 8 stores the computer program stored in the RAM or HDD. It is realized by executing. That is, the microprocessor of the on-vehicle apparatus 8 functionally includes a probe information providing unit 67, a traffic information acquisition unit 69, a probe information generation unit 84, and a position detection unit 83.
Furthermore, the probe information acquisition unit 22, the link matching unit 24, the travel lane estimation unit 25, the feature extraction unit 26, the clustering unit 27, the travel lane identification unit 28, and the traffic information calculation unit 29 of the server 2 are included in the server 2. The microprocessor is realized by executing a computer program stored in the RAM or the HDD. That is, the microprocessor included in the server 2 includes the probe information acquisition unit 22, the link matching unit 24, the travel lane estimation unit 25, the feature extraction unit 26, the clustering unit 27, the travel lane identification unit 28, and the traffic information calculation unit 29. It has functionally.

Also, the present invention may be the method described above. Furthermore, the present invention may be a computer program that implements these methods by a computer.

Furthermore, in the present invention, the computer program may be recorded in a computer readable non-transitory recording medium, such as an HDD, a CD-ROM, a semiconductor memory, or the like.

In the present invention, the computer program may be transmitted via a telecommunication line, a wireless or wired communication line, a network represented by the Internet, data broadcasting or the like.
Further, each of the above-described devices may be realized by a plurality of computers.

In addition, some or all of the functions of the above-described devices may be provided by cloud computing. That is, some or all of the functions of each device may be realized by the cloud server.
Furthermore, the above embodiment and the above modification may be combined respectively.

It should be understood that the embodiments disclosed herein are illustrative and non-restrictive in every respect. The scope of the present invention is indicated not by the meaning described above but by the scope of the claims, and it is intended to include all the modifications within the meaning and scope equivalent to the scope of the claims.

DESCRIPTION OF SYMBOLS 1 traffic information provision system 2 server 3 network 4 base station 5 target vehicle 6 in-vehicle device 7 teacher vehicle 8 in-vehicle device 21 communication I / F unit 22 probe information acquisition unit 23 storage device 23A probe information 23B map database 24 link matching unit 25 travel Lane estimation unit 26 Feature quantity extraction unit 27 Clustering unit 28 Traffic lane specification unit 29 Traffic information calculation unit 61 GPS receiver 62 Speed sensor 63 Direction sensor 64 Acceleration sensor 65 Position detection unit 66 Probe information generation unit 67 Probe information provision unit 68 Communication I / F unit 69 traffic information acquisition unit 81 reinforcement signal receiver 82 camera 83 position detection unit 84 probe information generation unit

Claims (11)

Computer,
A first acquisition unit that acquires speed transition information in the predetermined section of a teacher vehicle that is a vehicle whose traveling lane in the predetermined section is known;
A second acquisition unit for acquiring speed transition information in the predetermined section of a target vehicle whose vehicle is an unknown vehicle in the predetermined section;
A feature amount extraction unit for extracting feature amounts of each vehicle from the speed transition information of each of the teacher vehicle and the target vehicle;
A classification unit that classifies the target vehicle by classifying at least the characteristic quantity of the target vehicle among the characteristic quantity of the teacher vehicle and the characteristic quantity of the target vehicle into one or more groups;
A computer program for causing a vehicle to function as a traveling lane identification unit that identifies a traveling lane of the target vehicle included in each group based on the feature amount of the teacher vehicle. The computer program according to claim 1, wherein the feature quantity of each vehicle extracted by the feature quantity extraction unit includes at least one of an average speed, a maximum speed, a minimum speed, a deceleration point and an acceleration point of the vehicle. The first acquisition unit further acquires type information of the teacher vehicle.
The second acquisition unit further acquires type information of the target vehicle,
The classification unit classifies at least the feature amount of the target vehicle into one or more groups for each type of the target vehicle.
The driving lane identification unit identifies the traveling lane of the target vehicle included in each group based on the feature amount of the teacher vehicle for each type of the teacher vehicle. Computer program. The said classification | category part classifies the representative feature-value for every driving lane of the said teacher vehicle, and the feature-value of the said target vehicle into one or more groups, according to any one of claims 1 to 3. Computer program. The classification unit performs one or more of the feature amount of the teacher vehicle and the feature amount of the target vehicle under a constraint that the feature amounts of the teacher vehicle traveling in the same lane are classified into the same group. The computer program according to any one of claims 1 to 3, wherein the computer program is classified into the following group: The traveling lane identification unit selects one of the feature quantities of the teacher vehicle based on the comparison result of the representative feature quantity of each driving lane of the teacher vehicle and the center of gravity of the feature quantities of the target vehicles belonging to each group. The computer program according to any one of claims 1 to 3, wherein a travel lane of the target vehicle included in each group is specified by applying to the group of. The classification unit further classifies at least the feature quantities of the target vehicle into one or more groups under the constraint that the distance between groups is equal to or greater than a predetermined threshold. The computer program according to any one of the items. The travel lane identification unit is included in the travel lane of the target vehicle included in the first group, when there is a second group whose distance to the first group is equal to or less than a predetermined threshold. The computer program according to any one of claims 1 to 5, wherein a travel lane of the target vehicle is added. The driving lane identification unit is the teacher vehicle in which the traveling lanes of the target vehicles included in the plurality of groups are classified into the plurality of groups for a plurality of groups in which the distance between the groups is a predetermined distance or less The computer program according to any one of claims 1 to 5, wherein the computer program is specified to be any one of the following travel lanes. A first acquisition unit that acquires speed transition information in the predetermined section of a teacher vehicle that is a vehicle whose traveling lane in the predetermined section is known;
A second acquisition unit for acquiring speed transition information in the predetermined section of a target vehicle whose vehicle is an unknown vehicle in the predetermined section;
A feature amount extraction unit for extracting feature amounts of each vehicle from the speed transition information of each of the teacher vehicle and the target vehicle;
A classification unit that classifies the target vehicle by classifying at least the characteristic quantity of the target vehicle among the characteristic quantity of the teacher vehicle and the characteristic quantity of the target vehicle into one or more groups;
A travel lane identification device comprising: a travel lane identification unit that identifies a travel lane of the target vehicle included in each group based on the feature amount of the teacher vehicle. A travel lane identification device for identifying a travel lane of a target vehicle whose vehicle is an unknown vehicle in a predetermined section;
An on-vehicle device installed in the target vehicle that transmits probe information in the predetermined section of the target vehicle to the traffic lane identification device;
The traffic lane identification device
A first acquisition unit that acquires speed transition information in the predetermined section of a teacher vehicle that is a vehicle whose traveling lane in the predetermined section is known;
A second acquisition unit that acquires the probe information from the in-vehicle apparatus and acquires speed transition information of the target vehicle in the predetermined section based on the acquired probe information;
A feature amount extraction unit for extracting feature amounts of each vehicle from the speed transition information of each of the teacher vehicle and the target vehicle;
A classification unit that classifies the target vehicle by classifying at least the characteristic quantity of the target vehicle among the characteristic quantity of the teacher vehicle and the characteristic quantity of the target vehicle into one or more groups;
A travel lane identification system comprising: a travel lane identification unit that identifies a travel lane of the target vehicle included in each group based on the feature amount of the teacher vehicle.

Images