JP6916438B2 - Flow line identification system, server device, terminal device, flow line identification method, and flow line identification program - Google Patents

Description

The present invention relates to a flow line identification system, a server device, a terminal device, a flow line identification method, and a flow line identification program.

It is important to grasp the information for each lane of the road on which the own vehicle travels in terms of safe operation and service of the vehicle.
For example, if there is information on the flatness of the road for each lane, the driver of the night bus can select a lane with good flatness and drive, so that the vehicle can operate without waking up sleeping passengers due to vibration during driving. It will be possible to improve in terms of safe operation and service.

As a technique for grasping the position of the own vehicle for each lane, for example, a map matching device for calculating the position of the own vehicle on map information has been proposed (see, for example, Patent Document 1 and the like).

JP 2015-68665

However, in the technique described in Patent Document 1, it is necessary to match the position of the own vehicle on the map information. For example, when a new road is constructed, the number of lanes that can be passed changes and the map information must be changed. Matching becomes difficult. Therefore, it may be difficult to specify the number of lanes in a form that depends on the map information, and there is a problem that it does not function without the map information.

One aspect is to provide a flow line identification system, a server device, a terminal device, a flow line identification method, and a flow line identification program capable of deriving a condition capable of specifying the number of flow lines of a movement path.

In one embodiment, the movement line identification system is mounted on a moving body and has a control unit that transmits the position information of the moving body and an image of a predetermined shooting range taken at the position indicated by the position information. Using the terminal device and the position information and the image received from each of the plurality of mobile bodies, the first condition for specifying the number of movement lines of the movement path for each predetermined range geographically adjacent to each other is set. It has a server device having a control unit for deriving and transmitting the first condition for each predetermined range to the terminal device.

On one aspect, it is possible to provide a flow line identification system, a server device, a terminal device, a flow line identification method, and a flow line identification program capable of deriving a condition capable of specifying the number of flow lines of a movement path.

FIG. 1 is a diagram showing an example of a flow line identification system. FIG. 2 is a block diagram showing an example of the hardware configuration of the server device. FIG. 3 is a block diagram showing an example of the functional configuration of the server device. FIG. 4 is a diagram showing an example of travel history data stored in the travel history DB. FIG. 5 is an explanatory diagram showing an example of LA generation processing performed by the LA generation unit. FIG. 6 is an explanatory diagram showing an example of tuning processing performed by the tuning unit. FIG. 7 is an explanatory diagram showing another example of the tuning process performed by the tuning unit. FIG. 8 is an explanatory diagram showing another example of the tuning process performed by the tuning unit. FIG. 9 is an explanatory diagram showing an example of variations in the shooting state in the reference image. FIG. 10 is an explanatory diagram showing another example of variations in the shooting state in the reference image. FIG. 11 is an explanatory diagram showing another example of variations in the shooting state in the reference image. FIG. 12 is a diagram showing an example of a road map including information on the number of lanes. FIG. 13 is a block diagram showing an example of the hardware configuration of the terminal device. FIG. 14 is an explanatory diagram showing an example of the functional configuration of the terminal device. FIG. 15 is a sequence diagram showing an example of the flow of processing performed by the flow line identification system. FIG. 16 is a flowchart showing an example of the flow of processing performed by the server device. FIG. 17 is a flowchart showing an example of the flow of processing performed by the terminal device.

The flow line identification system can specify the number of flow lines of the movement path on which the moving body is located and the position of the flow line on which the moving body is located. The flow line identification system includes a terminal device mounted on the mobile body and a server device that receives position information and a peripheral image at the position from the terminal device.

Specifically, the terminal device transmits the position information of the moving body and the image obtained by capturing a predetermined shooting range at the position indicated by the position information. The terminal device is mounted on the mobile body.

Here, examples of the moving body include a vehicle, an animal, a water moving body, and the like. Examples of vehicles include automobiles and bicycles. Animals also include humans.
Examples of the moving route include roads and waterways. Further, examples of the road include a roadway and a sidewalk.
The flow line is a line indicating a route on which the moving body moves, and examples thereof include a lane.
The position information of the moving body is information that can specify the position of the moving body, and examples thereof include longitude and latitude information acquired by a GPS (Global Positioning System) unit.
Examples of the image obtained by capturing a predetermined shooting range include an image of the periphery of the moving body taken with a digital video camera or the like with a predetermined shooting range having a radius of 60 m centered on the moving body.

Next, the server device uses the position information and the image received from each of the plurality of moving objects to specify the number of flow lines of the moving path for each predetermined range geographically adjacent to each other. Is derived. The server device transmits the first condition for each predetermined range to the terminal device.

Here, the geographically adjacent predetermined range is, for example, a region in which the moving path is sequentially divided at intervals of about 100 m in a direction orthogonal to the traveling direction. The server device specifies the number of flow lines for each predetermined range.
In the following, a predetermined range may be referred to as "LA (Local Area)".

The first condition for specifying the number of flow lines is not particularly limited as long as the number of flow lines can be specified, and can be appropriately selected depending on the purpose.
Specifically, as the first condition, when artificial intelligence is not used, a reference image in which the number of flow lines can be easily specified is selected from an image of a movement path around the moving body, and the selected reference image and motion are selected. The condition is that the degree of similarity with the image for which the number of lines should be specified is equal to or higher than a predetermined value. In this way, the flow line identification system can specify that the known number of flow lines in the reference image is the number of flow lines of the image to be specified if the similarity is equal to or higher than a predetermined value. Alternatively, the flow line identification system derives the similarity between the image to be specified and the image having a different number of known flow lines, and uses the condition that the similarity is equal to or higher than a predetermined value to obtain a reference image having a high degree of similarity. It can be specified that the number of flow lines of the image is the number of flow lines of the image to be specified.

Further, as the first condition, when artificial intelligence is used, the image obtained by capturing a predetermined shooting range is the teacher image data group, and the number of known movement lines in the predetermined range at the position where the image is captured is the teacher correct answer data group. Let it be a learned weight parameter etc. learned and derived by artificial intelligence. In this way, the flow line identification system can infer using the derived learned weight parameter and specify the number of flow lines.

As a result, the flow line identification system can derive a condition capable of specifying the number of flow lines of the movement path by using the position information and the image received from each of the plurality of moving bodies, and further, the number of flow lines. Can be identified.

Next, the server device specifies which flow line the one mobile body that transmitted the position information and the image taken at that position is located in the predetermined range specified by the position information. The second condition for this is derived and transmitted to the terminal device.

Here, as the second condition for specifying which flow line is located, there is no particular limitation as long as it is possible to specify which flow line is located, and it is appropriate according to the purpose. You can choose.
It should be noted that "specifying which flow line the vehicle is located in" means "specifying the position of the own vehicle traveling flow line" if the moving body is a vehicle. Further, in the following, the "driving flow line position of the own vehicle" may be referred to as the "traffic line position".

Specifically, as the second condition, when artificial intelligence is not used, a reference image that makes it easy to specify the flow line position of the moving body is selected for each predetermined range, and the selected reference image and the flow line position are specified. The condition is that the degree of similarity with the image to be used is equal to or higher than a predetermined value. In this way, the flow line identification system can identify the known flow line position in the reference image as the flow line position of the image to be specified if the similarity is equal to or higher than a predetermined value. Alternatively, the flow line identification system derives the similarity between the image to be specified and the image having a different known flow line position, and uses the condition that the similarity is equal to or higher than a predetermined value to obtain a reference image having a high degree of similarity. It is possible to specify the flow line position of the image to be specified.

As a second condition, when artificial intelligence is used, the image in which the moving body is located is the teacher image data group, and the moving line in the image is the correct answer by the teacher. As a data group, it is a learned weight parameter obtained by learning by artificial intelligence. In this way, the flow line identification system can infer using the derived learned weight parameter and specify the flow line position.

In addition, the terminal device provides information on the number of flow lines, which is the specified result, and information on which flow line the terminal device is located in, and associates the position information with the image taken at the position indicated in the position information. Send to the server device in an identifiable state.
In this way, by feeding back the result of specifying the number of movement lines and the position of the movement line to the server device, the server device uses the specified result as the teacher correct answer data group when using artificial intelligence, and the learned weight parameter. Can be updated. Further, when artificial intelligence is not used, the server device can use the specified result as a candidate for a reference image in which the number of flow lines and the flow line in which the own vehicle is located are known.

Hereinafter, an embodiment of the present invention will be described, but the present invention is not limited to this embodiment.

(First Example)
(Flow line identification system)
FIG. 1 is a diagram showing an example of a flow line specifying system 10. The flow line identification system 10 of this embodiment is a system that specifies the number of lanes and the lane position of the expressway where the vehicle as a moving body is traveling.
As shown in FIG. 1, the flow line identification system 10 includes a server device 100 and terminal devices 200a, 200b, 200c, ... Mounted on vehicles A, B, C, ..., Respectively. ..
The server device 100 is communicably connected to the terminal devices 200a, 200b, 200c, ..., Through the network 300.
Since the terminal devices 200a, 200b, 200c, ... Have the same configuration, they will be collectively referred to as "terminal device 200" below. Further, vehicles A, B, C, ... May be simply referred to as "vehicles" when it is not necessary to distinguish them.

First, the hardware configuration and the functional configuration of the server device 100 will be described.

(Server device)
<Hardware configuration of server device>
FIG. 2 is a block diagram showing an example of the hardware configuration of the server device 100.
As shown in FIG. 2, the hardware configuration of the server device 100 includes the following devices. Each device is communicably connected via a bus 109.

The CPU (Central Processing Unit) 101 is a processing device that performs various controls and calculations. The CPU 101 realizes various functions by executing an OS (Operating System) or a program stored in the main storage device 102 or the like. That is, in this embodiment, the CPU 101 functions as a control unit 140, which will be described later, by executing the flow line specifying program.
The flow line specifying program does not necessarily have to be stored in the main storage device 102, the auxiliary storage device 105, or the like from the beginning. Further, the flow line identification program is stored in another information processing device or the like connected to the server device 100 via the Internet, LAN, WAN, or the like, and the server device 100 acquires and executes the flow line identification program from these. You may do it.

Further, the CPU 101 controls the operation of the entire server device 100. In this embodiment, the device that controls the operation of the entire server device 100 is the CPU 101, but the device is not limited to this, and may be, for example, an FPGA (Field Programmable Gate Array).

The main storage device 102 stores various programs and stores data and the like necessary for executing various programs.
The main storage device 102 has a ROM (Read Only Memory) and a RAM (Random Access Memory) (not shown).
The ROM stores various programs such as BIOS (Basic Input / Output System).
The RAM functions as a work range expanded when various programs stored in the ROM are executed by the CPU 101. The RAM is not particularly limited and can be appropriately selected depending on the purpose. Examples of the RAM include DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory).

The GPU (Graphics Processing Unit) 103 executes necessary processing such as graphic processing and parallel numerical calculation processing.

The VRAM (Video Random Access Memory) 104 functions as a memory area for holding data necessary for displaying an image on the display 108.

The auxiliary storage device 105 is not particularly limited as long as it can store various types of information, and can be appropriately selected depending on the intended purpose. Examples thereof include a solid state drive and a hard disk drive. Further, the auxiliary storage device 105 may be a portable storage device such as a CD (Compact Disc) drive, a DVD (Digital Versailles Disc) drive, or a BD (Blu-ray (registered trademark) Disc) drive.

The communication interface 106 is not particularly limited, and a known one can be used as appropriate. Examples thereof include a communication device using wireless or wired communication.

The input device 107 is not particularly limited as long as it can accept various requests for the server device 100, and a known input device 107 can be used as appropriate. Examples thereof include a keyboard, a mouse, and a touch panel.

The display 108 is not particularly limited, and a known display 108 can be used as appropriate, and examples thereof include a liquid crystal display and an organic EL display.

The server device 100 may be a part of a cloud, which is a group of computers on the network.

<Functional configuration of server device>
FIG. 3 is a block diagram showing an example of the functional configuration of the server device 100.
As shown in FIG. 3, the functional configuration of the server device 100 includes a communication unit 120, a storage unit 130, and a control unit 140.

Based on the instruction of the control unit 140, the communication unit 120 receives the travel history data including the vehicle position information and the image from each terminal device 200 using the communication interface 106.
Further, the communication unit 120 transmits the first condition for specifying the number of lanes and the second condition for specifying the lane position to the terminal device 200.

The storage unit 130 has a travel history database (hereinafter, also referred to as “travel history DB”) 121 in the auxiliary storage device 105.
The travel history DB 121 stores the travel history data received by the communication unit 120 as a travel history data group.

FIG. 4 is a diagram showing an example of travel history data stored in the travel DB 121.
As shown in FIG. 4, the travel history data includes the items of "vehicle ID, acquisition date and time, position information (longitude, latitude)" in this embodiment, and includes an image (not shown) associated with the terminal device 200.

The data item of the "vehicle ID" is data for identifying the vehicle on which the terminal device 200 is mounted, and is set in advance.
The data items of "acquisition date and time" and "position information (longitude, latitude)" are acquired by a GPS (Global Positioning System) unit mounted on the terminal device 200.
In this embodiment, the travel history data includes position information and an image at that position, but the present invention is not limited to this, and for example, "vehicle speed, acceleration, departure date and time, departure place (longitude, latitude), etc." Data items such as "arrival date and time, destination (longitude, latitude)" may be further included.

<Control unit>
In this embodiment, the control unit 140 has a function of controlling the operation of the entire server device 100, and also functions as an LA generation unit 141, a tuning unit 142, and a condition derivation unit 143.

FIG. 5 is an explanatory diagram showing an example of the LA generation process performed by the LA generation unit 141. Here, the LA generation process performed by the LA generation unit 141 will be described with reference to FIG.
First, the LA generation unit 141 specifies the traveling direction of the vehicle based on the acquisition date / time information and the position information in the travel history data of each vehicle. Next, the LA generation unit 141 sets a reference point at a place where the travel history data of the vehicle in the same traveling direction as shown by the arrow in FIG. 5 exists. Then, the LA generation unit 141 extracts the travel history data in the vicinity of the point 100 m ahead of the reference point in the traveling direction, and calculates the center of gravity of the position of the extracted travel history data as the center point of the LA.
If there is travel history data with a relatively long time interval for acquiring data and there is no travel history data in the vicinity of a point 100 m ahead, complementary data is calculated as shown by "○" in FIG. do.

6 to 8 are explanatory views showing an example of tuning processing performed by the tuning unit 142. Here, with reference to FIGS. 6 to 8, the tuning process performed by the tuning unit 142 will be described separately for the case where the LA is generated and the case where the road is newly constructed.

First, the tuning process when LA is generated will be described.
In this case, the tuning unit 142 derives the number of lanes based on the image of each acquisition point of the travel history data in the generated LA, and even if the number of lanes differs for each acquisition point, the number of lanes is within the LA. The "largest number of lanes" is specified as the number of lanes in the LA.
Specifically, as shown in FIG. 6, in the LAy where the number of lanes increases from "3" to "4", the number of lanes derived at each acquisition point is a mixture of "3" and "4". think. In this case, the tuning unit 142 specifies the number of lanes of the LAy as "4" because the number of lanes "4" is larger than the number of lanes "3" in the LAy.

Next, the tuning process when a new road is constructed will be described.
In this case, the tuning unit 142 compares the number of lanes before and after the derivation in the LA. The tuning unit 142 determines that the tuning process is unnecessary if the number of lanes before and after the derivation is the same, and determines that the tuning process is performed if the number of lanes before and after the derivation does not match.
Specifically, consider a case where a road as shown in FIG. 8 is newly constructed when the number of lanes is derived in a plurality of LAs as shown in FIG. 7 before the number of lanes is derived. In this case, since the number of lanes in LA2 and LA3 has increased from "4" to "5", the tuning unit 142 can perform tuning processing and respond even when a new road is constructed.

The condition derivation unit 143 derives a first condition for specifying the number of lanes and a second condition for specifying the lane position.
In this embodiment, the condition derivation unit 143 is a reference in which the white line on the road and the variation of the positional relationship with other vehicles in the vicinity are provided in the images taken from the own vehicle in the state shown in FIGS. 9 to 11. Using the image, the similarity that specifies the number of lanes and the lane position is derived.

The condition derivation unit 143 may determine whether or not at least one specific error between the information on the number of lanes received from the vehicle and the information on the lane position is equal to or higher than a certain threshold value. .. When the condition derivation unit 143 determines that the threshold value is equal to or higher than a certain threshold value, the condition derivation unit 143 derives the first condition and / or the second condition again. As a result, the control unit 140 can acquire information on the number of lanes and the lane position with few errors.

The control unit 140 may output the information on the number of lanes specified for each LA based on the first condition as a road map including the information on the number of lanes, as shown in FIG. This makes it possible to visualize the number of lanes on the road.

Next, the hardware configuration and the functional configuration of the terminal device 200 will be described.

(Terminal device)
In this embodiment, the terminal device 200 is a digital tachograph mounted on a commercial vehicle such as a truck, and has a GPS (Global Positioning System) unit and a camera capable of photographing the surroundings of the vehicle. The terminal device 200 sequentially transmits the travel history data acquired by associating the position information and the image around the vehicle at that position with the acquisition date and time by the GPS unit and the camera to the server device 100 every 1 second or every 10 seconds. Send.
<Hardware configuration of terminal device>
FIG. 13 is a block diagram showing an example of the hardware configuration of the terminal device 200.
As shown in FIG. 13, the hardware configuration of the terminal device 200 includes the following devices. Each device is connected so as to be able to communicate with each other via the bus 211.

The CPU 201 is a processing device that performs various controls and calculations. The CPU 201 realizes various functions by executing an OS or a program stored in the main storage device 202 or the like. That is, in this embodiment, the CPU 201 functions as the control unit 240, which will be described later, by executing the flow line specifying program.
The flow line specifying program does not necessarily have to be stored in the main storage device 202, the auxiliary storage device 205, or the like from the beginning. Further, the flow line identification program is stored in another information processing device or the like connected to the terminal device 200 via the Internet, LAN, WAN, or the like, and the terminal device 200 acquires and executes the flow line identification program from these. You may do it.

Further, the CPU 201 controls the operation of the entire terminal device 200. In this embodiment, the device that controls the operation of the entire terminal device 200 is the CPU 201, but the device is not limited to this, and may be, for example, an FPGA.

The main storage device 202 stores various programs and stores data and the like necessary for executing various programs.
The main storage device 202 has a ROM and a RAM (not shown).
The ROM stores various programs such as BIOS.
The RAM functions as a work range expanded when various programs stored in the ROM are executed by the CPU 201. The RAM is not particularly limited and can be appropriately selected depending on the purpose. Examples of the RAM include DRAM and SRAM.

The GPU 203 executes necessary processing such as graphic processing and parallel numerical calculation processing.

The VRAM 204 functions as a memory area for holding data necessary for displaying an image on the display 208.

The auxiliary storage device 205 is not particularly limited as long as it can store various types of information, and can be appropriately selected depending on the intended purpose. Examples thereof include a solid state drive and a hard disk drive. Further, the auxiliary storage device 205 may be a portable storage device such as a CD drive, a DVD drive, or a BD drive.

The communication interface 206 is not particularly limited, and a known one can be used as appropriate. Examples thereof include a wireless communication device.

The input device 207 is not particularly limited as long as it can accept various requests for the terminal device 200, and a known input device 207 can be used as appropriate, and examples thereof include a keyboard, a mouse, and a touch panel.

The display 208 is not particularly limited, and a known display 208 can be used as appropriate, and examples thereof include a liquid crystal display and an organic EL display.

The GPS unit 209 acquires latitude and longitude information as position information and time information as acquisition date and time.

The camera 210 is, for example, a digital video camera, etc., which is installed in a vehicle and photographs a predetermined shooting range centered on the vehicle. The number of cameras 210 may be singular or plural.

<Functional configuration of terminal device>
FIG. 14 is an explanatory diagram showing an example of the functional configuration of the terminal device 200.
As shown in FIG. 14, the functional configuration of the terminal device 200 includes a communication unit 220, a storage unit 230, and a control unit 240.

Based on the instruction of the control unit 240, the communication unit 220 transmits the travel history data including the vehicle position information and the image to the server device 100 using the communication interface 206.
Further, the communication unit 220 receives the first condition for specifying the number of lanes and the second condition for specifying the lane position from the server device 100.

The storage unit 230 stores the first condition and the second condition received from the server device 100 in the auxiliary storage device 205.

<Control unit>
In this embodiment, the control unit 240 has a function of controlling the operation of the entire terminal device 200, and also functions as a position information acquisition unit 241, an image acquisition unit 242, and a specific unit 243.

The position information acquisition unit 241 acquires latitude / longitude information as position information and time information as acquisition date / time using the GPS unit 209.

The image acquisition unit 242 uses a camera 210 installed in the vehicle to take a picture of a predetermined shooting range centered on the vehicle so that a white line on the road or another vehicle is reflected.

The identification unit 243 specifies the number of lanes based on the first condition in which the vehicle equipped with the terminal device 200 is located, and also specifies which lane is based on the second condition. Further, the specifying unit 243 transmits the specified result of the number of lanes and the lane position information to the server device 100 in a state in which the correspondence between the position information and the image taken at the position indicated by the position information can be identified. do.
By feeding back the result of specifying the number of lanes and the lane position to the server device 100 in this way, the server device 100 uses the specified result as the teacher correct answer data group when using artificial intelligence, and the learned weight parameter. Can be updated. Further, when artificial intelligence is not used, the server device can use the specified result as a candidate for a reference image in which the number of lanes and the lane in which the own vehicle is located are known.

The control unit 240 may change a predetermined shooting range of an image when it is determined that a specific error in the number of lanes in LA is equal to or higher than a certain threshold value.

In this embodiment, the terminal device 200 is a digital tachograph, but the present invention is not limited to this, and examples thereof include an in-vehicle device of a car navigation system which is mounted on a passenger car and has a camera capable of photographing the surroundings of the vehicle. Further, the traveling history data may be composed of, for example, position information acquired by ETC (Electronic Toll Collection System) and an image taken by a camera mounted on the vehicle and capable of photographing the surroundings of the vehicle. ..

FIG. 15 is a sequence diagram showing an example of the flow of processing performed by the flow line specifying system 10.
Here, with reference to the sequence diagram of FIG. 15, the process from when the terminal device 200 transmits the position information and the image of the own vehicle to the server device 100 to when the number of vehicles and the lane position are specified will be described.

In step S101, the terminal device 200 acquires the travel history data (position information and image) of the mobile body on which the terminal device 200 is mounted, transmits the data to the server device 100, and shifts the process to S102. In this embodiment, the terminal device 200 acquires travel history data every second.

In step S102, the server device 100 performs an LA estimation process for estimating which LA range the received travel history data is within.
Further, as a result of performing the LA estimation process, the server device 100 determines whether or not the acquisition point of the travel history data is "LA not set". When the server device 100 determines that the acquisition point of the travel history data is "LA not set", the server device 100 shifts the process to S103, and when it determines that the acquisition point is not "LA not set", the server device 100 shifts the process to S105.

In step S103, the server device 100 acquires a certain amount of travel history data of a plurality of vehicles in an area where LA is not set, then performs an LA generation process for newly generating LA, and shifts the process to S104.

In step S104, the server device 100 performs tuning processing on the LA generated in S103, and shifts the processing to S105 when the number of vehicles in the LA is specified.

In step S105, the terminal device 200 performs a lane number derivation process for deriving the lane number by determining whether or not the first condition for specifying the lane number is satisfied. The terminal device 200 transmits the derived information on the number of lanes to the server device 100, and shifts the processing to S106.

In step S106, the terminal device 200 performs a lane position derivation process for deriving the lane position by determining whether or not the second condition for specifying the lane position is satisfied. The terminal device 200 transmits the derived lane position information to the server device 100.
Further, the server device 100 determines whether or not the acquisition of the travel history data in the LA is completed based on the received travel history data. When the server device 100 determines that the acquisition of the travel history data in the LA has been completed based on the received travel history data, the server device 100 shifts the process to S107. Further, if the tuning process is performed in S104, the server device 100 shifts the process to S109, and returns the process to S101 when it is determined that the acquisition of the travel history data in the LA has not been completed.

In step S107, the server device 100 performs a process of deriving the number of lanes in the LA.
Further, the server device 100 compares the number of lanes in the LA derived in S107 with the number of lanes in the LA before derivation, returns the process to S101 if they match, and performs the process in S108 if they do not match. Move to.

In step S108, the server device 100 performs a tuning process in the same manner as the process of S104, and shifts the process to S109.

In step S109, when the server device 100 performs the tuning process in S104 or S108, the server device 100 derives the first condition and the second condition and transmits them to the terminal device 200, and returns the process to S101.
When the server device 100 determines that at least one specific error between the information on the number of lanes received from the vehicle and the information on which lane is located is equal to or higher than a certain threshold value. The first condition and / or the second condition may be derived again.
Further, the terminal device 200 may change the predetermined shooting range of the image when it is determined that the specific error of the number of lanes in the predetermined range is equal to or more than a certain threshold value.

Next, the processing performed by the server device 100 will be described.

FIG. 16 is a flowchart showing an example of the flow of processing performed by the server device 100.
Here, the flow of processing performed by the server device 100 will be described with reference to the flowchart shown in FIG.

In step S201, the server device 100 receives the travel history data acquired by the terminal device 200, and shifts the processing to S202.

In step S202 (corresponding to S102), the server device 100 performs an LA estimation process for estimating which LA range the received travel history data is within, and shifts the process to S203.

In step S203, the server device 100 determines whether or not the acquisition point of the travel history data is "LA set" as a result of performing the LA estimation process. The server device 100 shifts the process to S206 when it determines that the acquisition point of the travel history data is "LA set", and shifts the process to S204 when it determines that the acquisition point is not "LA set".

In step S204 (corresponding to S103), the server device 100 acquires a certain amount of travel history data of a plurality of vehicles in an area where LA is not set, then performs an LA generation process for newly generating LA, and shifts the process to S205. do.

In step S205 (corresponding to S104), the server device 100 performs tuning processing on the LA generated in S204, and when the number of vehicles in the LA is specified, the processing shifts to S206.

In step S206, the server device 100 receives the data of the number of lanes derived by the terminal device 200, and shifts the processing to S207.

In step S207, the server device 100 receives the lane position data derived from the terminal device 200, and shifts the process to S208.

In step S208 (corresponding to S107), when the server device 100 performs a process of deriving the number of lanes in the LA, the process shifts to S209.

In step S209, the server device 100 compares the number of lanes in the LA derived in S107 with the number of lanes in the LA before derivation, returns the process to S201 if they match, and processes if they do not match. To S210.

In step S210 (corresponding to S108), the server device 100 performs a tuning process in the same manner as the process of S205, and shifts the process to S211.

In step S211 (corresponding to S109), the server device 100 derives the first condition and the second condition, transmits them to the terminal device 200, and returns the process to S201.

In step S212, the server device 100 determines whether or not the tuning process has been performed in S205. When it is determined that the tuning process has been performed in S205, the server device 100 shifts the process to S211 and returns the process to S201 when it determines that the tuning process has not been performed in S205.

Next, the processing performed by the terminal device 200 will be described.

FIG. 17 is a flowchart showing an example of the flow of processing performed by the terminal device 200.
Here, the flow of processing performed by the terminal device 200 will be described with reference to the flowchart shown in FIG.
The terminal device 200 is in a state of receiving the first condition and the second condition from the server device 100 in advance. Further, when the server device 100 performs the tuning process, the terminal device 200 receives the first condition and the second condition from the server device 100.

In step S301 (corresponding to S101), the terminal device 200 acquires the travel history data (position information and image) of the moving body on which the terminal device 200 is mounted, transmits the data to the server device 100, and shifts the process to S302.

In step S302 (corresponding to S105), the terminal device 200 performs a lane number derivation process for deriving the lane number by determining whether or not the first condition for specifying the lane number is satisfied. The terminal device 200 transmits the derived information on the number of lanes to the server device 100, and shifts the processing to S302.

In step S303 (corresponding to S106), the terminal device 200 performs a lane position derivation process for deriving the lane position by determining whether or not the second condition for specifying the lane position is satisfied. The terminal device 200 transmits the derived lane position information to the server device 100.

As described above, the flow line identification system specifies the number of lanes for each predetermined range from the terminal device that transmits the position information of the vehicle and the peripheral image of the position and the position information and images received from a plurality of vehicles. The number of lanes is specified by a device that derives the conditions for the operation and transmits the conditions to the terminal device.

(Second Example)
In the second embodiment, the number of lanes and the lane position derivation method are different from those in the first embodiment, and the condition derivation unit 143 of the server device 100 in FIG. 3 and the specific unit of the terminal device 200 in FIG. 243 has artificial intelligence.
In the condition derivation in S109 of FIG. 15, the condition derivation unit 143 of the server device 100 uses a deep learning method using the teacher image data group and the teacher correct answer data group to perform a lane number learning process for deriving the first condition. , The lane position learning process for deriving the second condition is performed. This point is different from the first embodiment. The condition derivation unit 143 derives the first learned weight parameter as the first condition and the second learned weight parameter as the second condition by these learning processes, and derives the second learned weight parameter as the second condition of the terminal device 200. It is transmitted to the specific unit 243.
The first point is that the identification unit 243 of the terminal device 200 infers and specifies the number of lanes and the lane position by using the first learned weight parameter and the second learned weight parameter received from the condition derivation unit 143. It is different from the embodiment of.

[Learning process]
First, in the server device 100, the learning process performed by the condition derivation unit 143 using the GPU 103 and the VRAM 104 of FIG. 2 will be described.
The condition derivation unit 143 extracts, for example, white lines on the road, other vehicles in the vicinity, etc. from the images taken from the own vehicle in the state shown in FIGS. 9 to 11, the number of extracted white lines, and other vehicles in the vicinity. The first learned weight parameter is derived so that the number of lanes can be specified by the position of.
The condition derivation unit 143 derives the second learned weight parameter so that the lane position can be specified as well as the number of lanes.
The condition derivation unit 143 transmits the derived two learned weight parameters to the specific unit 243 of the terminal device 200. The server device 100 may specify the number of lanes and the lane position using two learned weight parameters.

[Inference processing]
Next, the inference processing performed by the specific unit 243 of the terminal device 200 will be described.
The identification unit 243 uses the two learned weight parameters obtained by the condition derivation unit 143 to perform similarity of the features of the generated image, and specifies the number of lanes and the lane position.
As a result, the terminal device 200 can accurately specify the number of lanes and the lane position by inference processing using the two learned weight parameters.

Regarding the above embodiments, the following additional notes will be further disclosed.
(Appendix 1)
A terminal device mounted on a moving body and having a control unit for transmitting the position information of the moving body and an image of a predetermined shooting range taken at a position indicated by the position information.
Using the position information and the image received from each of the plurality of moving bodies, a first condition for specifying the number of flow lines of the moving path for each predetermined range geographically adjacent to each other is derived, and the said A server device having a control unit that transmits the first condition for each predetermined range to the terminal device, and a server device.
A flow line identification system characterized by having.
(Appendix 2)
The server device uses the position information and the image received from each of the plurality of the moving bodies, and the one said moving body that has transmitted the position information and the image is specified by the position information. The flow line identification system according to Appendix 1, wherein a second condition derived for identifying which flow line is located in the above range is transmitted to the terminal device.
(Appendix 3)
The terminal device specifies the number of flow lines in which the moving body equipped with the terminal device is located based on the first condition, and specifies which flow line is based on the second condition. ,
A state in which the correspondence between the position information and the image taken at the position indicated by the position information can be identified by the information on the number of flow lines as a result of the identification and the information on which flow line the information is located on. The flow line identification system according to Appendix 2, wherein the information is transmitted to the server device in the above.
(Appendix 4)
When the server device determines that at least one specific error of the information on the number of flow lines received from the mobile body and the information on which flow line is located is equal to or higher than a certain threshold value. The flow line identification system according to Appendix 3, wherein the first condition and / or the timing for deriving the second condition again is determined.
(Appendix 5)
The terminal device is characterized in that, when it is determined that a specific error in the number of flow lines in the predetermined range is equal to or higher than a certain threshold value, the predetermined shooting range of the image is changed. The flow line identification system according to 3 or 4.
(Appendix 6)
The server device is characterized in that the information on the number of flow lines specified for each predetermined range based on the first condition is output as a movement route map including the information on the number of flow lines. The flow line identification system according to any one of 3 to 5.
(Appendix 7)
Using the position information of the moving body received from each of the terminal devices mounted on the plurality of moving bodies and the image of a predetermined shooting range taken at the position indicated by the position information, a predetermined geographically adjacent predetermined position is used. A server having a control unit that derives a first condition for specifying the number of movement lines of a movement path for each range and transmits the first condition for each predetermined range to the terminal device. Device.
(Appendix 8)
It is mounted on a moving body, and the position information of the moving body and an image obtained by capturing a predetermined shooting range at the position indicated by the position information are transmitted.
Using the position information received from each of the plurality of moving bodies and the image, a first condition derived for specifying the number of flow lines of a moving path for each predetermined range geographically adjacent to each other is determined. A terminal device characterized by receiving.
(Appendix 9)
It is mounted on a moving body, and the position information of the moving body and an image obtained by capturing a predetermined shooting range at the position indicated by the position information are transmitted.
Using the position information received from each of the plurality of moving bodies and the image, a first condition derived for specifying the number of flow lines of a moving path for each predetermined range geographically adjacent to each other is determined. Receive,
A flow line identification program characterized by having a computer execute processing.
(Appendix 10)
It is mounted on a moving body, and the position information of the moving body and an image obtained by capturing a predetermined shooting range at the position indicated by the position information are transmitted.
Using the position information received from each of the plurality of moving bodies and the image, a first condition derived for specifying the number of flow lines of a moving path for each predetermined range geographically adjacent to each other is determined. Receive,
A flow line identification method characterized in that a computer executes processing.

10 Lane Judgment System 100 Server Device 140, 240 Control Unit 200 Terminal Device A, B, C Vehicle (Mobile)

Claims (10)

A terminal device mounted on a moving body and having a control unit for transmitting the position information of the moving body and an image of a predetermined shooting range taken at a position indicated by the position information.
Using the position information and the image received from each of the plurality of moving bodies, a first condition for specifying the number of flow lines of the moving path for each predetermined range geographically adjacent to each other is derived, and the said A server device having a control unit that transmits the first condition for each predetermined range to the terminal device, and a server device.
A flow line identification system characterized by having. The server device uses the position information and the image received from each of the plurality of the moving bodies, and the one said moving body that has transmitted the position information and the image is specified by the position information. The flow line specifying system according to claim 1, wherein a second condition derived for identifying which flow line is located in the above range is transmitted to the terminal device. The terminal device specifies the number of flow lines in which the moving body equipped with the terminal device is located based on the first condition, and specifies which flow line is based on the second condition. ,
A state in which the correspondence between the position information and the image taken at the position indicated by the position information can be identified by the information on the number of flow lines as a result of the identification and the information on which flow line the information is located on. The flow line specifying system according to claim 2, wherein the information is transmitted to the server device. When the server device determines that at least one specific error of the information on the number of flow lines received from the mobile body and the information on which flow line is located is equal to or higher than a certain threshold value. The flow line specifying system according to claim 3, wherein the first condition and / or the timing for deriving the second condition again is determined. The terminal device is characterized in that when it is determined that a specific error in the number of flow lines in the predetermined range is equal to or more than a certain threshold value, the predetermined shooting range of the image is changed. Item 6. The flow line identification system according to item 3 or 4. The server device is characterized in that it outputs information on the number of flow lines specified for each predetermined range based on the first condition as a movement route map including the information on the number of flow lines. Item 5. The flow line identification system according to any one of Items 3 to 5. Using the position information of the moving body received from each of the terminal devices mounted on the plurality of moving bodies and the image of a predetermined shooting range taken at the position indicated by the position information, a predetermined geographically adjacent predetermined position is used. A server having a control unit that derives a first condition for specifying the number of movement lines of a movement path for each range and transmits the first condition for each predetermined range to the terminal device. Device. It is mounted on the moving body, and the position information of the moving body and the image obtained by taking a predetermined shooting range at the position indicated by the position information are transmitted to the server device.
Using the position information received from each of the plurality of mobile bodies and the image, the first derived from the server device in order to specify the number of flow lines of the movement path for each predetermined range geographically adjacent to each other. A terminal device characterized in that the condition 1 is received from the server device. It is mounted on the moving body, and the position information of the moving body and the image obtained by taking a predetermined shooting range at the position indicated by the position information are transmitted to the server device.
Using the position information received from each of the plurality of mobile bodies and the image, the first derived from the server device in order to specify the number of flow lines of the movement path for each predetermined range geographically adjacent to each other. Receiving the condition 1 from the server device,
A flow line identification program characterized by having a computer execute processing. It is mounted on the moving body, and the position information of the moving body and the image obtained by taking a predetermined shooting range at the position indicated by the position information are transmitted to the server device.
Using the position information received from each of the plurality of mobile bodies and the image, the first derived from the server device in order to specify the number of flow lines of the movement path for each predetermined range geographically adjacent to each other. Receiving the condition 1 from the server device,
A flow line identification method characterized in that a computer executes processing.

Images