Information collection system, server, vehicle, method, and program

Abstract

The present invention makes it possible for a server to collect data to be collected, when a vehicle is in a situation corresponding to a specific scene. A recognition model selection means (21) selects a recognition model for identifying that a vehicle is in a situation corresponding to a specific scene on the basis of sensor information. A transmission means (22) transmits the selected recognition model to a vehicle (30). A scene determination means (31) determines whether or not the vehicle (30) is in a situation corresponding to a specific scene on the basis of the recognition model received from a server (20) and the sensor information. A data transmission means (32) transmits information to the server (20) when it is determined that the vehicle (30) is in a situation corresponding to a specific scene. A data collection means (23) collects the information transmitted from the vehicle (30).

Description

[Technical field]

[0001] The present disclosure relates to an information collection system, a server, a vehicle, an information collection method, an information transmission method, and a computer-readable medium. [Background technology]

[0002] Connected cars equipped with a communication device and capable of communicating with external servers and the like are becoming more and more popular. As a related technique, Patent Document 1 discloses a data collection system that collects road information from an on-board device mounted in each vehicle. The on-board device described in Patent Document 1 is connected to various sensors of the vehicle and detects abnormalities based on signals input from the various sensors. When the on-board device detects an abnormality, it generates collection conditions for data related to the detected abnormality. The collection conditions include target vehicle conditions, a recording trigger, and collection contents.

[0003] The in-vehicle device transmits the generated collection conditions and a data collection request that satisfies the collection conditions to the data collection device. When the data collection device accepts the data collection request, it performs data collection based on the received collection conditions. In data collection, the data collection device transmits a collection condition file including the collection conditions generated by the in-vehicle device that transmitted the data collection request to the in-vehicle device of each vehicle. The in-vehicle device of each vehicle transmits data that satisfies the collection conditions to the data collection device. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-38409 A Summary of the Invention [Problem to be solved by the invention]

[0005] In Patent Document 1, the data collection device collects data that meets the collection conditions. Therefore, data can be collected more efficiently than when all data is collected from the in-vehicle device. However, in Patent Document 1, the in-vehicle device transmits a data collection request when a condition such as a sensor value exceeding a threshold value is triggered. Therefore, the data collection device cannot collect data when the vehicle is in a situation corresponding to a specific scene.

[0006] In consideration of the above circumstances, the present disclosure aims to provide an information collection system, a server, a vehicle, an information collection method, an information transmission method, and a computer-readable medium that can cause a server to collect data when the vehicle is in a situation corresponding to a specific scene. [Means for solving the problem]

[0007] In order to achieve the above object, the present disclosure provides, as a first aspect, an information collection system. The information collection system includes a server and a vehicle connected to the server via a network. In the information collection system, the server includes a recognition model selection means for selecting a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on sensor information, a transmission means for transmitting the recognition model to the vehicle, and a data collection means for collecting information transmitted from the vehicle. The vehicle includes a scene determination means for determining whether or not the vehicle is in a situation corresponding to the specific scene based on the recognition model received from the server and the sensor information, and a data transmission means for transmitting information to the server when it is determined that the vehicle is in a situation corresponding to the specific scene.

[0008] As a second aspect, the present disclosure provides a server having a recognition model selection means for selecting a recognition model for identifying that a vehicle is in a situation corresponding to a specific scene based on sensor information, a transmission means for transmitting the recognition model to the vehicle via a network, and a data collection means for collecting information from the vehicle when it is determined in the vehicle that the vehicle is in a situation corresponding to a specific scene based on the recognition model and the sensor information.

[0009] The present disclosure provides, as a third aspect, a vehicle, which includes a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on sensor information received from a server via a network, a scene determination means for determining whether the vehicle is in a situation corresponding to the specific scene based on the sensor information, and a data transmission means for transmitting information to the server via the network when it is determined that the vehicle is in a situation corresponding to the specific scene.

[0010] The present disclosure provides, as a fourth aspect, an information collection method, which includes selecting a recognition model for identifying that a vehicle is in a situation corresponding to a specific scene based on sensor information, transmitting the recognition model to the vehicle via a network, and collecting information from the vehicle when it is determined in the vehicle that the vehicle is in a situation corresponding to the specific scene based on the recognition model and the sensor information.

[0011] The present disclosure provides, as a fifth aspect, an information transmission method, which includes determining whether or not the vehicle is in a situation corresponding to a specific scene based on a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on sensor information received from a server via a network and the sensor information, and transmitting information to the server via the network when it is determined that the vehicle is in a situation corresponding to the specific scene.

[0012] The present disclosure provides, as a sixth aspect, a computer-readable medium storing a program for causing a computer to execute a process of selecting a recognition model for identifying that a vehicle is in a situation corresponding to a specific scene based on sensor information, transmitting the recognition model to the vehicle via a network, and collecting information from the vehicle when it is determined in the vehicle that the vehicle is in a situation corresponding to the specific scene based on the recognition model and the sensor information.

[0013] The present disclosure provides, as a seventh aspect, a computer-readable medium, which stores a program for causing a processor to execute a process of determining whether or not the vehicle is in a situation corresponding to a specific scene based on a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on sensor information received from a server via a network and the sensor information, and transmitting information to the server via the network when it is determined that the vehicle is in a situation corresponding to the specific scene. Effect of the Invention

[0014] The information collection system, server, vehicle, information collection method, information transmission method, and computer-readable medium disclosed herein enable the server to collect data when the vehicle is in a situation corresponding to a particular scene. [Brief description of the drawings]

[0015] [Figure 1] 1 is a block diagram illustrating an information collection system according to the present disclosure. [Diagram 2] FIG. 1 is a block diagram showing a data collection system according to a first embodiment of the present disclosure. [Diagram 3] FIG. 2 is a block diagram showing an example of the configuration of a server. [Figure 4] FIG. 1 is a block diagram showing an example configuration of a vehicle. [Diagram 5] 11 is a flowchart showing an operation procedure in the server. [Figure 6]4 is a flowchart showing an operation procedure in a vehicle. [Figure 7] FIG. 11 is a block diagram showing a data collection system according to a second embodiment of the present disclosure. [Figure 8] FIG. 1 is a block diagram showing an example of the configuration of a computer device. [Figure 9] FIG. 2 is a block diagram showing a hardware configuration of an electronic control device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Prior to describing the embodiments of the present disclosure, an overview of the present disclosure will be described. FIG. 1 illustrates an information collection system according to the present disclosure. The information collection system 10 includes a server 20 and a vehicle 30. In the information collection system 10, the server 20 and the vehicle 30 are connected to each other via a network. The server 20 includes a recognition model selection means 21, a transmission means 22, and a data collection means 23. The vehicle 30 includes a scene determination means 31 and a data transmission means 32.

[0017] The recognition model selection means 21 of the server 20 selects a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information. The transmission means 22 transmits the recognition model selected by the recognition model selection means 21 to the vehicle 30.

[0018] The scene determination means 31 of the vehicle 30 determines whether or not the vehicle 30 is in a situation corresponding to a specific scene based on the recognition model and the sensor information received from the server 20. The data transmission means 32 transmits information to the server 20 when it is determined that the vehicle 30 is in a situation corresponding to a specific scene. The data collection means 23 of the server 20 collects information transmitted from the vehicle 30.

[0019] In the present disclosure, the server 20 transmits the recognition model to the vehicle 30. The vehicle 30 determines whether the vehicle is in a situation corresponding to a specific scene based on the received recognition model and the sensor information. When the vehicle 30 determines that the vehicle is in a situation corresponding to a specific scene, the vehicle transmits information to the server 20. In the present disclosure, the server 20 can specify a scene for which the vehicle 30 transmits information through the recognition model transmitted to the vehicle 30. Thus, the server 20 can collect data when the vehicle 30 is in a situation corresponding to a specific scene.

[0020] Hereinafter, an embodiment of the present disclosure will be described in detail. FIG. 2 shows a data collection system according to a first embodiment of the present disclosure. The data collection system 100 includes a server 110 and one or more vehicles 200. In the data collection system 100, the server 110 is connected to the vehicle 200 via a network 150. The network 150 includes a wireless communication network using a communication line standard such as LTE (Long Term Evolution). The network 150 may include a wireless communication network such as WiFi (registered trademark) or a fifth generation mobile communication system. The data collection system 100 corresponds to the information collection system 10 shown in FIG. 1. The server 110 corresponds to the server 20 shown in FIG. 1. The vehicle 200 corresponds to the vehicle 30 shown in FIG. 1.

[0021] 3 shows an example of the configuration of the server 110. The server 110 includes a recognition model selection unit 111, a parameter determination unit 112, a transmission unit 113, a data collection unit 114, and an analysis device 115. The server 110 is disposed in, for example, a connected service infrastructure.

[0022] The recognition model selection unit 111 selects a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information. The recognition model includes at least one of a recognition model for determining a scene of aggressive driving, a recognition model for determining a scene of ignoring a traffic signal, and a recognition model for determining a scene of drowsiness. The recognition model is configured as, for example, an artificial intelligence (AI) model. The recognition model includes, for example, a convolutional neural network (CNN).

[0023] The recognition model selection unit 111 selects a recognition model to be transmitted to the vehicle 200, for example, based on the position information of the vehicle 200. The recognition model selection unit 111 selects a recognition model, for example, using regional characteristic information that associates a geographical position with a specific scene. The regional characteristic information stores information indicating a specific scene to be recognized, for example, for each section divided into a mesh shape at a predetermined distance. The recognition model selection unit 111 acquires information indicating a specific scene associated with the position of the vehicle 200 from the regional characteristic information. The recognition model selection unit 111 selects a recognition model for identifying the specific scene indicated by the acquired information as the recognition model to be transmitted to the vehicle 200.

[0024] The recognition model selection unit 111 may select a recognition model according to the type of road on which the vehicle 200 is traveling. The types of roads include, for example, expressways and other motorways, trunk roads, urban roads, suburban roads, etc. For example, when the vehicle 200 is traveling on a continuous single road without intersections, such as an expressway, the recognition model selection unit 111 may select a rear-end collision accident scene discrimination model.

[0025] Alternatively, the recognition model selection unit 111 may select a recognition model according to a route (route name) on which the vehicle is traveling, such as National Route 1 or Prefectural Route 55. Furthermore, the recognition model selection unit 111 may select a recognition model according to a combination of a region, such as a prefecture, on which the vehicle 200 is traveling, and a road type or a route name. The recognition model selection unit 111 may select a recognition model specified by an operator as the recognition model to be transmitted to the vehicle 200.

[0026] The recognition model selection unit 111 may select a recognition model for each vehicle. For example, the recognition model selection unit 111 may select a different recognition model for one vehicle 200 and another vehicle 200. The recognition model selection unit 111 may select a plurality of recognition models for one vehicle 200 as the recognition models to be transmitted to the vehicle 200. For example, the recognition model selection unit 111 may select a recognition model for discriminating a scene of aggressive driving and a recognition model for discriminating a scene of drowsiness as the recognition models to be transmitted to one vehicle 200. The recognition model selection unit 111 corresponds to the recognition model selection means 21 shown in FIG. 1.

[0027] The parameter determination unit (parameter determination means) 112 determines parameters specifying information to be acquired from the vehicle 200 based on the recognition model selected by the recognition model selection unit 111. The parameter determination unit 112 holds, for example, a table in which identification models or situations identified using the identification models are associated with the types of information to be acquired. The parameter determination unit 112 refers to the table and determines parameters according to the selected recognition model.

[0028] The transmission unit 113 transmits the recognition model selected by the recognition model selection unit 111 to the vehicle 200. The transmission unit 113 also transmits a parameter sheet describing the parameters (type of information) determined by the parameter determination unit 112 to the vehicle 200. The transmission unit 113 places the recognition model and the parameter sheet in the vehicle 200 using, for example, an OTA (Over The Air) technology. The transmission unit 113 corresponds to the transmission means 22 shown in Fig. 1.

[0029] The data collection unit 114 collects information transmitted from each vehicle 200 from each vehicle 200. The data collection unit 114 corresponds to the data collection means 23 shown in FIG. 1. The analysis device 115 performs analysis using the information collected by the data collection unit 114. The analysis device 115 performs analysis using the information collected from the vehicle 200, for example, for each situation identified using the recognition model. The analysis device 115 does not necessarily need to be configured inside the server 110, and may be configured as a device physically different from the server 110.

[0030] 4 shows an example of the configuration of a vehicle 200. The vehicle 200 includes a surroundings monitoring sensor 201, a vehicle sensor 202, a vehicle control ECU (Electric Control Unit) 203, a scene recognition unit 204, and a communication device 205. In the vehicle 200, these components are configured to be able to communicate with each other via an in-vehicle LAN (Local Area Network) or a CAN (Controller Area Network), etc.

[0031] The periphery monitoring sensor 201 is a sensor that monitors the surrounding conditions of the vehicle 200. The periphery monitoring sensor 201 includes, for example, a camera, a radar, and a LiDAR (Light Detection and Ranging). The periphery monitoring sensor 201 may include, for example, a plurality of cameras that capture images of the front, rear, right side, and left side of the vehicle. The periphery monitoring sensor 201 may include a camera that captures images of the interior of the vehicle 200.

[0032] The vehicle sensor 202 is a sensor for detecting various conditions of the vehicle 200. The vehicle sensor 202 includes, for example, sensors such as a vehicle speed sensor for detecting the vehicle speed, a steering sensor for detecting the steering angle, an accelerator pedal position sensor for detecting the accelerator pedal position, and a brake pedal pressure sensor for detecting the amount of depression of the brake pedal.

[0033] The vehicle control ECU 203 is an electronic control device that performs driving control of the vehicle 200. In general, an electronic control device has a processor, a memory, an I / O (Input / Output), and a bus connecting these. Based on sensor information output by the vehicle sensor 202, the vehicle control ECU 203 performs various controls such as control of the fuel injection amount, control of the engine ignition timing, and control of the amount of assist of the power steering.

[0034] The communication device 205 is configured as a device for performing wireless communication between the vehicle 200 and the network 150 (see FIG. 2). The communication device 205 includes an antenna, a transmitter, and a receiver for wireless communication.

[0035] The scene recognition unit 204 is a functional unit that transmits information specified in a parameter sheet to the server 110 when the vehicle 200 is in a situation corresponding to a specific scene identified using the recognition model. The scene recognition unit 204 receives the recognition model and the parameter sheet from the server 20 through the communication device 205. In addition, the scene recognition unit 204 transmits information (data) specified in the parameter sheet to the server 20 through the communication device 205.

[0036] The scene recognition unit 204 has a recognition model storage unit 241, a scene determination unit 242, a data transmission unit 243, and a recognition model update unit 244. The recognition model storage unit 241 stores one or more recognition models received from the server 110. The scene determination unit 242 acquires a recognition model from the recognition model storage unit 241. In addition, the scene determination unit 242 acquires sensor information from the periphery monitoring sensor 201 and the vehicle sensor 202. The scene determination unit 242 determines whether or not the vehicle 200 is in a situation corresponding to a specific scene based on the recognition model and the sensor information.

[0037] The scene determination unit 242 obtains, as sensor information, information obtained using, for example, a camera included in the perimeter monitoring sensor 201 and at least one of a speed sensor and an acceleration sensor included in the vehicle sensor 202. The scene determination unit 242 inputs the sensor information to, for example, a CNN constituting a recognition model. The recognition model outputs a determination result indicating whether or not the vehicle is in a situation corresponding to a specific scene. When the determination result indicates that the vehicle is in a situation corresponding to a specific scene, the scene determination unit 242 notifies the data transmission unit 243 of that effect. The scene determination unit 242 corresponds to the scene determination means 31 shown in FIG. 1.

[0038] When the scene determination unit 242 determines that the vehicle is in a situation corresponding to a specific scene, the data transmission unit 243 transmits data specified in the parameter sheet to the server 110. The data transmission unit 243 transmits data specified in the parameter sheet, for example, among data obtainable from the perimeter monitoring sensor 201, the vehicle sensor 202, and the vehicle control ECU 203, to the server 110. When, for example, outside-vehicle video data is specified in the parameter sheet, the data transmission unit 243 transmits video of the camera included in the perimeter monitoring sensor 201 to the server 110. When the scene determination unit 242 determines that the vehicle is not in a situation corresponding to a specific scene, the data transmission unit 243 does not transmit data specified in the parameter sheet. The data transmission unit 243 may always transmit specific information, such as vehicle position information, to the server 110.

[0039] The data transmission unit 243 acquires data specified in the parameter sheet from the periphery monitoring sensor 201, the vehicle sensor 202, and the vehicle control ECU 203, for example, regardless of the judgment result of the scene judgment unit 242. If the scene judgment unit 242 judges that the vehicle is in a situation corresponding to a specific scene, the data transmission unit 243 transmits the acquired data to the server 110. If the scene judgment unit 242 does not judge that the vehicle is in a situation corresponding to a specific scene, the data transmission unit 243 discards the acquired data. The data transmission unit 243 corresponds to the data transmission means 32 shown in FIG. 1.

[0040] In this embodiment, the transmission of the parameter sheet from the server 110 to the vehicle 200 may be omitted. In that case, the data transmission unit 243 may transmit, for example, predetermined information to the server 110. For example, in the data transmission unit 243, a recognition model is associated with information to be transmitted when the recognition model is used. The data transmission unit 243 may transmit information corresponding to the recognition model used in the scene determination unit 242 to the server 110.

[0041] The recognition model update unit (recognition model update means) 244 receives a recognition model from the server 110. When the recognition model update unit 244 receives a recognition model from the server 110, it stores the received recognition model in the recognition model storage unit 241. When the recognition model storage unit 241 has already stored a recognition model, the recognition model update unit 244 may update the recognition model stored in the recognition model storage unit 241 with the recognition model received from the server 110. When the recognition model stored in the recognition model storage unit 241 is updated, the scene determination unit 242 uses the updated recognition model to determine whether or not the vehicle is in a situation corresponding to a specific scene.

[0042] Next, an operation procedure in the data collection system will be described. Fig. 5 shows an operation procedure (information collection method) in the server 110. The recognition model selection unit 111 (see Fig. 3) of the server 110 selects a recognition model to be transmitted to the vehicle 200 (step A1). In step A1, the recognition model selection unit 111 selects a scene to be recognized in the vehicle 200 based on, for example, position information of the vehicle 200, and selects a recognition model corresponding to the selected scene.

[0043] The parameter determination unit 112 determines parameters corresponding to the recognition model determined in step A1 as parameters to be transmitted to the vehicle 200 (step A2). The transmission unit 113 transmits a parameter sheet describing the recognition model selected in step A1 and the parameters determined in step A2 to the vehicle 200 via the network 150 (see FIG. 2) (step A3).

[0044] 6 shows an operation procedure (information transmission method) in vehicle 200. Scene recognition unit 204 (see FIG. 4) of vehicle 200 receives a recognition model and a parameter sheet from server 110 via network 150 (step B1). Scene recognition unit 204 may receive the recognition model and parameters while the vehicle is traveling.

[0045] The recognition model update unit 244 stores the recognition model received in step B1 in the recognition model storage unit 241 (step B2). If a recognition model has already been stored in the recognition model storage unit 241, the recognition model update unit 244 updates the recognition model stored in the recognition model storage unit 241 with the recognition model received in step B1.

[0046] The scene determination unit 242 acquires sensor information from the surroundings monitoring sensor 201 and the vehicle sensor 202 of the vehicle 200. The scene determination unit 242 applies the acquired sensor information to the recognition model and determines whether or not the vehicle is in a situation corresponding to a specific scene (step B3). If it is determined in step B3 that the vehicle is in a situation corresponding to a specific scene, the data transmission unit 243 transmits data specified in the parameter sheet to the server 110 via the network 150 (step B4).

[0047] Returning to FIG. 5, the data collection unit 114 of the server 110 receives data transmitted from the vehicle 200 (step A4). The data collection unit 114 collects data from a plurality of vehicles 200 when it is determined that the vehicle is in a situation corresponding to a specific scene. The data collection unit 114, for example, associates the collected data with the recognition model transmitted to the vehicle 200 and outputs the data to the analysis device 115. The analysis device 115 accumulates the data received from the vehicle 200 for each scene recognized using the recognition model, for example. The analysis device 115 analyzes the accumulated data.

[0048] In this embodiment, the server 110 transmits the recognition model and parameters to the vehicle 200. The vehicle 200 determines whether the vehicle is in a situation corresponding to a specific scene based on the received recognition model and sensor information. When the vehicle 200 determines that the vehicle is in a situation corresponding to a specific scene, the server 110 transmits information specified in the parameters to the server 110. In the present disclosure, the server 110 can specify a scene for which the vehicle 200 is to transmit information and the information to be transmitted through the recognition model and parameters transmitted to the vehicle 200. Therefore, the server 110 can collect data to be collected when the vehicle 200 is in a situation corresponding to a specific scene.

[0049] In this embodiment, the server 110 can select a recognition model and parameters according to data to be acquired. If the server 110 acquires all data from the vehicle 200, the server 110 needs to receive a large amount of data from a large number of vehicles 200 and process a large amount of data. In this embodiment, the server 110 can specify the situation in which data is transmitted and the data to be acquired by using the recognition model and parameters. Therefore, in this embodiment, the data to be collected from the vehicle 200 can be narrowed down, and the storage cost in the server 110 can be reduced. In addition, in this embodiment, the amount of data transferred between the server 110 and the vehicle 200 can be reduced, and the communication cost can be reduced.

[0050] Next, a second embodiment of the present disclosure will be described. FIG. 7 shows a data collection system according to the second embodiment of the present disclosure. A data collection system 100a according to this embodiment differs from the data collection system 100 according to the first embodiment shown in FIG. 2 in that the data collection system 100a further includes a traffic information system 300. In this embodiment, the configuration of the server 110 may be the same as the configuration of the server 110 in the first embodiment shown in FIG. 3. Also, the configuration of the vehicle 200 may be the same as the configuration of the vehicle 200 in the first embodiment shown in FIG. 4.

[0051] The traffic information system 300 is a system that provides regional characteristic information related to traffic. The traffic information system 300 holds information that associates specific traffic-related events with locations where the events frequently occur. The traffic information system 300 holds, for example, collision frequent location information 310 including information indicating locations where rear-end collisions frequently occur. The traffic information system 300 may further hold, for example, information indicating locations where aggressive driving frequently occurs. The traffic information system 300 provides the held collision frequent location information 310 to the server 110.

[0052] The recognition model selection unit 111 of the server 110 refers to information held by the traffic information system 300 and selects a recognition model. For example, the recognition model selection unit 111 refers to collision frequent location information 310 and determines whether the vehicle 200 is located at a location where collisions frequently occur. When the server 110 determines that collisions frequently occur near the current location of the vehicle 200, it selects a recognition model for identifying collision scenes as the recognition model to be transmitted to the vehicle 200.

[0053] In this embodiment, the server 110 cooperates with the traffic information system 300 and selects a recognition model using information held by the traffic information system 300. For example, the server 110 selects a recognition model using collision frequent spot information 310, and can transmit to the vehicle 200 a recognition model that identifies a scene of a rear-end collision when rear-end collisions frequently occur near the current location of the vehicle 200. In this way, in this embodiment, a recognition model that discriminates events such as accidents that frequently occur near the current location of the vehicle 200 can be arranged in the vehicle 200. Other effects are similar to those described in the first embodiment.

[0054] In each of the above embodiments, the parameter sheet may include information indicating the priority of data. The priority includes, for example, "high", "medium", and "low". The data transmission unit 243 (see FIG. 4) may monitor the communication band in the network 150 (see FIG. 2) and transmit data with a high priority to the server 110 preferentially according to the communication band. For example, when the communication band is greater than a first threshold, the data transmission unit 243 transmits data of all priorities to the server 110. When the communication band is equal to or less than the first threshold, the data transmission unit 243 may transmit data with a priority set to "high" or "medium" to the server 110 and discard data with a priority set to "low". When the communication band is equal to or less than a second threshold that is smaller than the first threshold, the data transmission unit 243 may transmit data with a priority set to "high" to the server 110 and discard data with a priority set to "medium" or "low". The parameter sheet may include information specifying data to be always transmitted to the server 110 regardless of the determination result of the scene determination unit 242.

[0055] In each of the above embodiments, the data transmission unit 243 may determine data that is important for the analysis based on the purpose and scene of the analysis performed by the analysis device 115 (see FIG. 3), and discard data that is not important. For example, when the scene determination unit 242 determines a following scene, little information can be obtained from the video data. When sensor data is sufficient for the analysis, the data transmission unit 243 may discard the video data on the vehicle side without transmitting it to the server 110.

[0056] Next, a physical configuration will be described. In this disclosure, the server 110 may be configured as a computer device. Fig. 8 shows an example of the configuration of a computer device that may be used as the server 110. The computer device 500 has a control unit (CPU: Central Processing Unit) 510, a storage unit 520, a ROM (Read Only Memory) 530, a RAM (Random Access Memory) 540, a communication interface (IF: Interface) 550, and a user interface 560.

[0057] The communication interface 550 is an interface for connecting the computer device 500 to a communication network via a wired communication means or a wireless communication means, etc. The user interface 560 includes a display unit such as a display, and an input unit such as a keyboard, a mouse, and a touch panel.

[0058] The storage unit 520 is an auxiliary storage device that can hold various data. The storage unit 520 does not necessarily need to be a part of the computer device 500, and may be an external storage device or a cloud storage connected to the computer device 500 via a network.

[0059] The ROM 530 is a non-volatile storage device. For example, a semiconductor storage device having a relatively small capacity, such as a flash memory, is used for the ROM 530. The programs executed by the CPU 510 may be stored in the storage unit 520 or the ROM 530. The storage unit 520 or the ROM 530 stores various programs for implementing the functions of each unit in the server 110, for example.

[0060] The RAM 540 is a volatile storage device. Various semiconductor memory devices such as a dynamic random access memory (DRAM) or a static random access memory (SRAM) are used for the RAM 540. The RAM 540 can be used as an internal buffer for temporarily storing data and the like. The CPU 510 loads a program stored in the storage unit 520 or the ROM 530 into the RAM 540 and executes it. The CPU 510 executes a program to realize the functions of each unit in the server 110. The CPU 510 may have an internal buffer for temporarily storing data and the like.

[0061] Also, in the present disclosure, the scene recognition unit 204 (see FIG. 4) may be configured as an electronic control unit (electronic control device). FIG. 9 shows an example of a hardware configuration of an electronic control device that may be used for the scene recognition unit 204. The electronic control device 600 has a processor 601, a ROM 602, and a RAM 603. In the electronic control device 600, the processor 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. The electronic control device 600 may include other circuits such as peripheral circuits, communication circuits, and interface circuits, although not shown.

[0062] The ROM 602 is a non-volatile storage device. For example, a semiconductor storage device with a relatively small capacity such as a flash memory is used for the ROM 602. The ROM 602 stores the programs executed by the processor 501. The RAM 603 is a volatile storage device. For the RAM 603, various semiconductor memory devices such as DRAM or SRAM are used. The RAM 640 can be used as an internal buffer for temporarily storing data, etc.

[0063] The processor 601 loads a program stored in the ROM 602 into the RAM 603 and executes the program. When the CPU 601 executes the program, the functions of the various parts of the scene recognition unit 204 can be realized.

[0064] The above-mentioned program can be stored using various types of non-transitory computer-readable media and can be supplied to the computer device 500 or the electronic control device 600. The non-transitory computer-readable medium includes various types of tangible storage media. Examples of the non-transitory computer-readable medium include magnetic recording media such as a flexible disk, a magnetic tape, or a hard disk, magneto-optical recording media such as a magneto-optical disk, optical disk media such as a compact disc (CD) or a digital versatile disk (DVD), and semiconductor memories such as a mask ROM, a programmable ROM (PROM), an erasable PROM (EPROM), a flash ROM, or a RAM. The program may also be supplied to the computer using various types of temporary computer-readable media. Examples of the temporary computer-readable medium include an electric signal, an optical signal, and an electromagnetic wave. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or via a wireless communication path.

[0065] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the above-described embodiments, and changes and modifications to the above-described embodiments that do not deviate from the spirit of the present disclosure are also included in the present disclosure.

[0066] For example, some or all of the above embodiments may be described as, but are not limited to, the following supplementary notes.

[0067] [Appendix 1] A server; a vehicle connected to the server via a network; The server, a recognition model selection means for selecting a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information; a transmission means for transmitting the recognition model to the vehicle; A data collection means for collecting information transmitted from the vehicle, The vehicle is a scene determination means for determining whether or not the vehicle is in a situation corresponding to a specific scene based on the recognition model received from the server and the sensor information; and a data transmission means for transmitting information to the server when it is determined that the vehicle is in a situation corresponding to a specific scene.

[0068] [Appendix 2] The server further includes a parameter determination means for determining a parameter specifying information to be acquired from the vehicle based on the selected recognition model; The transmitting means further transmits the determined parameters to the vehicle; An information collection system as described in Appendix 1, wherein the data transmission means transmits information specified in the parameters received from the server to the server when it is determined that the vehicle is in a situation corresponding to a specific scene.

[0069] [Appendix 3] 3. The information collection system according to claim 1, wherein the recognition model selection means selects a recognition model to be transmitted to the vehicle based on position information of the vehicle.

[0070] [Appendix 4] The information collection system described in Appendix 3, wherein the recognition model selection means acquires information indicating a specific scene associated with the position of the vehicle from regional characteristic information that associates geographical positions with the specific scenes, and selects a recognition model for identifying the specific scene indicated by the acquired information as the recognition model to be transmitted to the vehicle.

[0071] [Appendix 5] 5. The information collection system according to claim 1, wherein the recognition model selection means selects the recognition model depending on a type of road on which the vehicle is traveling.

[0072] [Appendix 6] The information collection system according to any one of appendices 1 to 5, wherein the sensor information includes at least one of information acquired using a camera mounted on the vehicle, information acquired using a speed sensor, and information acquired using an acceleration sensor.

[0073] [Appendix 7] The information gathering system according to any one of appendices 1 to 6, wherein the recognition model includes a Convolutional Neural Network (CNN).

[0074] [Appendix 8] An information collection system as described in any one of Appendices 1 to 7, wherein the vehicle further has a recognition model update means for receiving the recognition model from the server and updating the recognition model used by the scene determination means with the received recognition model.

[0075] [Appendix 9] 9. The information collection system according to any one of claims 1 to 8, wherein the server further comprises an analysis device that performs analysis using the information collected by the data collection means.

[0076] [Appendix 10] a recognition model selection means for selecting a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information; a transmission means for transmitting the recognition model to the vehicle via a network; A server comprising a data collection means for collecting information from the vehicle when it is determined that the vehicle is in a situation corresponding to a specific scene based on the recognition model and sensor information in the vehicle.

[0077] [Appendix 11] The server described in Appendix 10, further comprising a parameter determination means for determining parameters specifying information to be transmitted to the vehicle when it is determined in the vehicle that the vehicle is in a situation corresponding to a specific scene based on the selected recognition model.

[0078] [Appendix 12] 12. The server according to claim 10, wherein the recognition model selection means selects a recognition model to be transmitted to the vehicle based on position information of the vehicle.

[0079] [Appendix 13] The server according to claim 12, wherein the recognition model selection means acquires information indicating a specific scene associated with the position of the vehicle from regional characteristic information that associates geographical positions with the specific scenes, and selects a recognition model for identifying the specific scene indicated by the acquired information as the recognition model to be transmitted to the vehicle.

[0080] [Appendix 14] 14. The server according to any one of claims 10 to 13, wherein the recognition model selection means selects the recognition model depending on a type of road on which the vehicle is traveling.

[0081] [Appendix 15] 15. The server according to any one of appendices 10 to 14, further comprising an analysis device that performs analysis using information collected by the data collection means.

[0082] [Appendix 16] A vehicle, a recognition model for identifying whether the vehicle is in a situation corresponding to a specific scene based on sensor information received from a server via a network, and a scene determination means for determining whether the vehicle is in a situation corresponding to the specific scene based on the sensor information; and a data transmission means for transmitting information to the server via a network when it is determined that the vehicle is in a situation corresponding to a specific scene.

[0083] [Appendix 17] The vehicle described in Appendix 16, wherein the data transmission means transmits to the server, when it is determined that the vehicle is in a situation corresponding to a specific scene, information specified in a parameter received from the server that specifies the information to be transmitted to the server.

[0084] [Appendix 18] 18. The vehicle of claim 16, wherein the sensor information includes at least one of information obtained using a camera mounted on the vehicle, information obtained using a speed sensor, and information obtained using an acceleration sensor.

[0085] [Appendix 19] The vehicle described in any one of Appendices 16 to 18, further comprising a recognition model update means for receiving the recognition model from the server and updating the recognition model used by the scene determination means with the received recognition model.

[0086] [Appendix 20] Selecting a recognition model for identifying that the vehicle is in a situation corresponding to a particular scene based on the sensor information; Transmitting the recognition model to the vehicle via a network; An information gathering method for gathering information from a vehicle when it is determined that the vehicle is in a situation corresponding to a specific scene based on the recognition model and sensor information in the vehicle.

[0087] [Appendix 21] determining whether the vehicle is in a situation corresponding to a specific scene based on a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information received from the server via the network and the sensor information; An information transmitting method for transmitting information to the server via a network when it is determined that the vehicle is in a situation corresponding to a specific scene.

[0088] [Appendix 22] Selecting a recognition model for identifying that the vehicle is in a situation corresponding to a particular scene based on the sensor information; Transmitting the recognition model to the vehicle via a network; A non-transitory computer-readable medium storing a program for causing a computer to execute a process of collecting information from the vehicle when it is determined that the vehicle is in a situation corresponding to a specific scene based on the recognition model and sensor information in the vehicle.

[0089] [Appendix 23] determining whether the vehicle is in a situation corresponding to a specific scene based on a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information received from the server via the network and the sensor information; A non-transitory computer-readable medium storing a program for causing a processor to execute a process of transmitting information to the server via a network when it is determined that the vehicle is in a situation corresponding to a particular scene. [Explanation of symbols]

[0090] 10: Information gathering system 20: Server 21: Recognition model selection means 22: Transmission means 23: Data collection methods 30: Vehicle 31: Scene determination means 32: Data transmission means 100: Data Acquisition System 110: Server 111: Recognition model selection unit 112: Parameter determination unit 113: Transmitter 114: Data collection department 115: Analyzer 150: Network 200:Vehicle 201: Periphery monitoring sensor 202: Vehicle sensor 203: Vehicle control ECU 204: Scene Recognition Unit 205: Communication equipment 241: Recognition model memory unit 242: Scene determination section 243: Data transmission unit 244: Recognition model update unit 300: Traffic Information System 310: Information on locations where rear-end collisions occur frequently

Claims

1. A server; a vehicle connected to the server via a network; The server, a recognition model selection means for selecting a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information; a transmission means for transmitting the recognition model to the vehicle; A data collection means for collecting information transmitted from the vehicle, The vehicle is a scene determination means for determining whether or not the vehicle is in a situation corresponding to a specific scene based on the recognition model received from the server and the sensor information; a data transmission means for transmitting information to the server when it is determined that the vehicle is in a situation corresponding to a specific scene; The recognition model selection means acquires information indicating a scene related to a specific event associated with the position of the vehicle from regional characteristic information that associates geographical positions with points where specific events frequently occur, and selects a recognition model for identifying a scene related to the specific event indicated by the acquired information as the recognition model to be transmitted to the vehicle.

2. The sensor information includes at least information acquired using a perimeter monitoring sensor, The server further includes a parameter determination means for determining a parameter specifying information to be acquired from the vehicle based on the selected recognition model; The transmitting means further transmits the determined parameters to the vehicle; 2. The information collection system according to claim 1, wherein the data transmission means transmits to the server information specified in the parameters received from the server when it is determined that the vehicle is in a situation corresponding to a specific scene.

3. 3. The information gathering system according to claim 1, wherein the recognition model selection means selects the recognition model depending on a type of road on which the vehicle is traveling.

4. 4. The information collection system according to claim 1, further comprising a recognition model update means for receiving the recognition model from the server and updating the recognition model used by the scene determination means with the received recognition model.

5. a recognition model selection means for selecting a recognition model for identifying that the vehicle is in a situation corresponding to a specific scene based on the sensor information; a transmission means for transmitting the recognition model to the vehicle via a network; a data collection means for collecting information from the vehicle when it is determined that the vehicle is in a situation corresponding to a specific scene based on the recognition model and sensor information in the vehicle; The recognition model selection means acquires information indicating a scene related to a specific event associated with the position of the vehicle from regional characteristic information that associates geographical positions with points where specific events frequently occur, and selects a recognition model for identifying a scene related to the specific event indicated by the acquired information as the recognition model to be transmitted to the vehicle.

6. A vehicle, a recognition model for identifying whether the vehicle is in a situation corresponding to a specific scene based on sensor information received from a server via a network, the recognition model being for identifying a scene related to a specific event associated with the position of the vehicle, the scene being selected using regional characteristic information that associates geographical positions with locations where specific events frequently occur, and a scene determination means for determining whether the vehicle is in a situation corresponding to a scene related to the specific event based on the sensor information; and a data transmission means for transmitting information to the server via a network when it is determined that the vehicle is in a situation corresponding to a scene related to a specific event.

7. A recognition model is selected for acquiring information indicating a scene related to a specific event associated with the vehicle's position from regional characteristic information that associates geographical positions with locations where a specific event frequently occurs, and for identifying that the vehicle is in a situation corresponding to the specific scene based on sensor information, the recognition model being for identifying a scene related to the specific event indicated by the acquired information; transmitting the selected recognition model to the vehicle via a network; An information collection method for collecting information from a vehicle when it is determined that the vehicle is in a situation corresponding to a scene related to a specific event based on the recognition model and sensor information in the vehicle.

8. a recognition model for identifying whether the vehicle is in a situation corresponding to a specific scene based on sensor information received from a server via a network, the recognition model being for identifying a scene related to a specific event associated with the position of the vehicle, the scene being selected using regional characteristic information that associates geographical positions with locations where specific events frequently occur, and determining whether the vehicle is in a situation corresponding to a scene related to a specific event based on the sensor information; An information transmission method for transmitting information to the server via a network when it is determined that the vehicle is in a situation corresponding to a scene related to a specific event.

9. A recognition model for acquiring information indicating a scene related to a specific event associated with the vehicle's position from regional characteristic information that associates geographical positions with locations where a specific event frequently occurs, and for identifying that the vehicle is in a situation corresponding to the specific scene based on sensor information, the recognition model for identifying the scene related to the specific event indicated by the acquired information being selected; transmitting the selected recognition model to the vehicle via a network; A program for causing a computer to execute a process of collecting information from the vehicle when it is determined that the vehicle is in a situation corresponding to a scene related to a specific event based on the recognition model and sensor information in the vehicle.

Images