Information processing system, information processing method, non-transitory computer-readable medium, and information processing device

Abstract

This information processing system: acquires vehicle detection data in which the position of a vehicle and an ID of a detected object indicated in an image captured by a camera disposed in the vehicle are associated with each other (S10); determines an ID of a road link corresponding to the position of the vehicle (S12); acquires, from map data, an ID of a registered object that is associated with the road link and has the same attribute as the detected object (S12); updates a cumulative voting point by adding a predetermined voting point to the cumulative voting point if the ID of the detected object does not match the ID of the registered object (S15); and changes the registered object associated with the road link to the detected object on the map data if the cumulative voting point is greater than a predetermined threshold (S17).

Description

Information processing system, information processing method, non-transitory computer-readable medium, and information processing device

[0001] The present disclosure relates to an information processing system, an information processing method, a non-transitory computer-readable medium, and an information processing device.

[0002] Because real-world conditions change daily, it is desirable to be able to quickly update map data to keep up with changes in the real world. In particular, increasing the frequency of map data updates to minimize the time lag between the real world and map data is an important factor for autonomous vehicles that use map data. However, updating map data is extremely time-consuming, and various methods for automating the map data update process have been proposed. For example, Patent Document 1 discloses a system that generates map change data for current high-precision map data based on observation data transmitted from a large number of autonomous vehicles operating in autonomous driving mode.

[0003] Japan Special Table No. 2023-514538

[0004] Meanwhile, when automating map data update work, it is essential to ensure the reliability of the automatic map data update process. In this regard, in a map data update system configured with a large number of vehicles and one or more servers, if the accuracy of information transmitted from some of the vehicles (e.g., vehicle position information or information about the object to be updated) is low, it is conceivable that the map data may be updated based on information with low accuracy or reliability. As a result, there is a problem in that the reliability of the updated map data may be reduced.

[0005] In view of the above, an object of the present disclosure is to provide an information processing system, an information processing method, a non-transitory computer-readable medium, and an information processing device that are capable of automatically updating map data with high reliability.

[0006] An information processing system according to one aspect of the present disclosure is an information processing system for updating map data, which acquires vehicle detection data in which the position of a vehicle and first identification information of a detected object shown in an image captured by a camera disposed on the vehicle are associated with each other, determines identification information of a road network element corresponding to the position of the vehicle, acquires second identification information of a registered object associated with the road network element and having the same attributes as the detected object from the map data, determines whether the second identification information matches the first identification information, and if the second identification information and the first identification information do not match, updates the cumulative voting points by adding a predetermined voting point to the cumulative voting points for determining whether the registered object linked to the road network element on the map data should be changed to the detected object, and if the cumulative voting points exceed a predetermined threshold, changes the registered object linked to the road network element on the map data to the detected object.

[0007] Another aspect of the present disclosure is an information processing system for updating map data, which acquires vehicle detection data in which the position of a vehicle and identification information indicating the state of a road shown in an image captured by a camera disposed on the vehicle are associated with each other, determines identification information of a road network element corresponding to the position of the vehicle, updates a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the state of the road should be associated with the road network element and registered on the map data, and when the cumulative voting point exceeds a predetermined threshold, associates the state of the road with the road network element and registers it on the map data.

[0008] Another aspect of the present disclosure is an information processing system for updating map data, which acquires vehicle detection data in which the position of a vehicle and identification information indicating dynamic information shown in an image captured by a camera disposed on the vehicle are associated with each other, determines identification information of a road network element corresponding to the position of the vehicle, updates a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the dynamic information should be associated with the road network element and registered on the map data or whether the dynamic information associated with the road network element should be changed, and when the cumulative voting point exceeds a predetermined threshold, associates the dynamic information with the road network element and registers it on the map data, or changes the dynamic information associated with the road network element on the map data.

[0009] An information processing method for updating map data according to one aspect of the present disclosure is executed by a computer and includes the steps of: acquiring vehicle detection data in which the position of a vehicle and first identification information of a detected object shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the position of the vehicle; acquiring second identification information of a registered object associated with the road network element and having the same attributes as the detected object from the map data; determining whether the second identification information and the first identification information match; if the second identification information and the first identification information do not match, updating a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the registered object linked to the road network element on the map data should be changed to the detected object; and if the cumulative voting point exceeds a predetermined threshold, changing the registered object linked to the road network element on the map data to the detected object.

[0010] An information processing method for updating map data according to another aspect of the present disclosure is executed by a computer and includes the steps of: acquiring vehicle detection data in which the position of a vehicle and identification information indicating the state of a road shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the position of the vehicle; updating a cumulative voting point by adding a predetermined voting point to the cumulative voting point for determining whether the state of the road should be associated with the road network element and registered on the map data; and, when the cumulative voting point exceeds a predetermined threshold, associating the state of the road with the road network element and registering it on the map data.

[0011] An information processing method for updating map data according to another aspect of the present disclosure is executed by a computer and includes the steps of: acquiring vehicle detection data in which a vehicle position and identification information indicating dynamic information shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the vehicle position; updating a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the dynamic information should be associated with the road network element and registered in the map data or whether the dynamic information associated with the road network element should be changed; and, when the cumulative voting point exceeds a predetermined threshold, associating the dynamic information with the road network element and registering it in the map data, or changing the dynamic information associated with the road network element in the map data.

[0012] Also provided is a non-transitory computer readable medium that holds instructions (computer program) that, when executed, cause a computer to perform the information processing method.

[0013] An information processing device according to one aspect of the present disclosure includes a processor and a memory that stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the information processing device performs the information processing method.

[0014] According to the present disclosure, it is possible to provide an information processing system, an information processing method, a non-transitory computer-readable medium, and an information processing device that are capable of automatically updating map data with high reliability.

[0015] FIG. 1 is a diagram illustrating an example of an information processing system according to an embodiment of the present disclosure (hereinafter, the present embodiment); FIG. 2 is a diagram illustrating an example of the configuration of a vehicle according to the present embodiment; FIG. 3 is a diagram illustrating an example of the configuration of a map update server according to the present embodiment; FIG. 4 is a diagram for explaining road network data; FIG. 5 is a flowchart for explaining an information processing method for updating map data according to a first example of the present embodiment; FIG. 6 is a flowchart for explaining a process for generating vehicle detection data; FIG. 7 is a diagram illustrating an example of vehicle detection data; FIG. 8 is a diagram illustrating an example of a detected object reference table; FIG. 9 is a diagram illustrating an example of people flow data; FIG. 10 is a flowchart for explaining an information processing method for updating map data according to a second example of the present embodiment; and FIG. 11 is a flowchart for explaining an information processing method for updating map data according to a third example of the present embodiment.

[0016] (Outline of this embodiment) An outline of this embodiment will be described below.

[0017] An information processing system according to one aspect of the present disclosure is an information processing system for updating map data, which acquires vehicle detection data in which the position of a vehicle and first identification information of a detected object shown in an image captured by a camera disposed on the vehicle are associated with each other, determines identification information of a road network element corresponding to the position of the vehicle, acquires second identification information of a registered object associated with the road network element and having the same attributes as the detected object from the map data, determines whether the second identification information matches the first identification information, and if the second identification information and the first identification information do not match, updates the cumulative voting points by adding a predetermined voting point to the cumulative voting points for determining whether the registered object linked to the road network element on the map data should be changed to the detected object, and if the cumulative voting points exceed a predetermined threshold, changes the registered object linked to the road network element on the map data to the detected object.

[0018] According to the above configuration, if the second identification information of the registered object and the first identification information of the detected object do not match, the cumulative voting points are updated. Then, if the cumulative voting points exceed a predetermined threshold, the registered object associated with the road network element (specifically, a road link or a road node) in the map data is changed to a detected object. In this way, when the reliability or accuracy of vehicle detection data is low, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy, thereby sufficiently improving the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing system that can automatically update map data with high reliability.

[0019] The information processing system may also acquire image data in which the image captured by the camera and the time of capture are associated with each other, acquire location data in which the position of the vehicle and the time the position of the vehicle was captured are associated with each other, identify first identification information of the detected object shown in the image using an image recognition model, and generate the vehicle detection data in which the position of the vehicle and the first identification information of the detected object are associated with each other.

[0020] According to the above configuration, vehicle detection data in which the vehicle position and the first identification information of the detected object are associated with each other can be automatically generated based on image data, position data, and an image recognition model.

[0021] The information processing system may also determine the predetermined voting point depending on at least one of a traffic volume associated with the road network element and an accuracy of the vehicle's position.

[0022] According to the above configuration, the predetermined voting points are determined according to at least one of traffic volume and vehicle position accuracy, so that the voting points can be optimized according to traffic volume and vehicle position accuracy. For example, if the voting points for roads with low traffic volume are higher than the voting points for roads with high traffic volume, it is possible to effectively prevent a situation in which updates of information related to road network elements with low traffic volume are significantly delayed compared to updates of information related to road network elements with high traffic volume. Furthermore, if the voting points are set lower as the vehicle position accuracy decreases, it is possible to sufficiently ensure the reliability of the automatic map data update process.

[0023] The information processing system may also determine the predetermined voting point according to the traffic volume associated with the road network element in a time period that includes the time when the image showing the detected object was taken.

[0024] According to the above configuration, since the traffic volume associated with a specific road network element changes depending on the time period, it is possible to further optimize voting points according to the traffic volume in the time period including the time of shooting.

[0025] Another aspect of the present disclosure is an information processing system for updating map data, which acquires vehicle detection data in which the position of a vehicle and identification information indicating the state of a road shown in an image captured by a camera disposed on the vehicle are associated with each other, determines identification information of a road network element corresponding to the position of the vehicle, updates a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the state of the road should be associated with the road network element and registered on the map data, and when the cumulative voting point exceeds a predetermined threshold, associates the state of the road with the road network element and registers it on the map data.

[0026] According to the above configuration, when vehicle detection data includes identification information indicating road conditions (e.g., road surface deterioration), cumulative voting points are updated. Thereafter, when the cumulative voting points exceed a predetermined threshold, the road conditions are associated with road network elements (specifically, road links and road nodes) and registered in the map data. In this way, when the reliability or accuracy of vehicle detection data is low, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy (specifically, a situation in which the road conditions are immediately registered in the map data), thereby sufficiently improving the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing system capable of automatically updating map data with high reliability. In particular, it is possible to provide an information processing system capable of reflecting road conditions in map data with high reliability.

[0027] The information processing system may also delete the road status associated with the road network element from the map data if the process of registering the road status on the map data is not performed for a predetermined period of time.

[0028] According to the above configuration, after a road condition changes (e.g., after road conditions such as road surface deterioration are improved due to road construction or the like), the process of registering the road condition in the map data is not executed for a predetermined period of time. Therefore, the road condition associated with the road network element is automatically deleted from the map data. In this way, it is possible to effectively prevent a situation in which information regarding the road condition continues to remain in the map data after the road condition (e.g., road surface deterioration) has changed.

[0029] The information processing system may also acquire image data in which the image captured by the camera and the time of capture are associated with each other, acquire location data in which the vehicle's position is associated with the time the vehicle's position was captured, use an image recognition model to identify identification information indicating the state of the road shown in the image, and generate the vehicle detection data in which the vehicle's position and the identification information indicating the state of the road are associated with each other.

[0030] According to the above configuration, vehicle detection data in which the vehicle position and identification information indicating the road condition are mutually associated can be automatically generated based on image data, position data, and an image recognition model.

[0031] The information processing system may also determine the predetermined voting point depending on at least one of a traffic volume associated with the road network element and an accuracy of the vehicle's position.

[0032] According to the above configuration, the predetermined voting points are determined according to at least one of traffic volume and vehicle position accuracy, so that the voting points can be optimized according to traffic volume and vehicle position accuracy. For example, if the voting points for roads with low traffic volume are higher than those for roads with high traffic volume, it is possible to effectively prevent a situation in which updates of information related to road network elements with low traffic volume are significantly delayed compared to updates of information related to road network elements with high traffic volume. Furthermore, if the voting points are set lower as the vehicle position accuracy decreases, it is possible to sufficiently improve the reliability of the automatic map data update process.

[0033] Another aspect of the present disclosure is an information processing system for updating map data, which acquires vehicle detection data in which the position of a vehicle and identification information indicating dynamic information (e.g., the presence of an emergency vehicle or a variable sign) shown in an image captured by a camera arranged on the vehicle are associated with each other, determines identification information of a road network element corresponding to the position of the vehicle, updates a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the dynamic information should be associated with the road network element and registered on the map data or whether the dynamic information associated with the road network element should be changed, and when the cumulative voting point exceeds a predetermined threshold, associates the dynamic information with the road network element and registers it on the map data, or changes the dynamic information associated with the road network element on the map data.

[0034] According to the above configuration, when vehicle detection data includes identification information indicating dynamic information such as the presence of an emergency vehicle or a variable sign, cumulative voting points are updated. Then, when the cumulative voting points exceed a predetermined threshold, the dynamic information is associated with a road network element (specifically, a road link or a road node) and registered in the map data, or the dynamic information associated with the road network element is changed. In this way, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy when the reliability or accuracy of the vehicle detection data is low (specifically, a situation in which dynamic information is immediately reflected in the map data based on vehicle detection data with low reliability or accuracy), thereby sufficiently improving the reliability of the automatic map data update process. Therefore, an information processing system capable of automatically updating map data with high reliability can be provided. In particular, an information processing system capable of reflecting dynamic information in map data with high reliability can be provided.

[0035] In addition, the information processing system may delete the dynamic information associated with the road network element from the map data if the process of associating the dynamic information with the road network element and registering it in the map data is not executed for a predetermined period of time.

[0036] According to the above configuration, after passing a road associated with a road network element of dynamic information, the registration process is not executed for a predetermined period of time. Therefore, the dynamic information associated with the specific road network element is automatically deleted from the map data. In this way, it is possible to effectively prevent a situation in which, for example, information about an emergency vehicle remains associated with the specific road network element in the map data after the emergency vehicle has passed a road associated with the specific road network element.

[0037] The information processing system may also acquire image data in which the image captured by the camera and the time of capture are associated with each other, acquire location data in which the vehicle's position is associated with the time the vehicle's position was captured, identify identification information indicating the dynamic information shown in the image using an image recognition model, and generate the vehicle detection data in which the vehicle's position and the identification information indicating the dynamic information are associated with each other.

[0038] According to the above configuration, vehicle detection data in which the vehicle position and identification information indicating dynamic information are associated with each other can be automatically generated based on image data, position data, and an image recognition model.

[0039] An information processing method for updating map data according to one aspect of the present disclosure is executed by a computer and includes the steps of: acquiring vehicle detection data in which the position of a vehicle and first identification information of a detected object shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the position of the vehicle; acquiring second identification information of a registered object associated with the road network element and having the same attributes as the detected object from the map data; determining whether the second identification information and the first identification information match; if the second identification information and the first identification information do not match, updating a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the registered object linked to the road network element on the map data should be changed to the detected object; and if the cumulative voting point exceeds a predetermined threshold, changing the registered object linked to the road network element on the map data to the detected object.

[0040] According to the above method, if the second identification information of the registered object and the first identification information of the detected object do not match, the cumulative voting points are updated. Then, if the cumulative voting points exceed a predetermined threshold, the registered object associated with the road network element in the map data is changed to a detected object. In this way, when the reliability or accuracy of vehicle detection data is low, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy, thereby sufficiently increasing the reliability of the automatic map data update process. Therefore, an information processing method capable of automatically updating map data with high reliability can be provided.

[0041] An information processing method for updating map data according to another aspect of the present disclosure is executed by a computer and includes the steps of: acquiring vehicle detection data in which the position of a vehicle and identification information indicating the state of a road shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the position of the vehicle; updating a cumulative voting point by adding a predetermined voting point to the cumulative voting point for determining whether the state of the road should be associated with the road network element and registered on the map data; and, when the cumulative voting point exceeds a predetermined threshold, associating the state of the road with the road network element and registering it on the map data.

[0042] According to the above method, when vehicle detection data includes identification information indicating road conditions (e.g., road surface deterioration), cumulative voting points are updated. Then, when the cumulative voting points exceed a predetermined threshold, the road conditions are associated with road network elements and registered in map data. In this way, when the reliability or accuracy of vehicle detection data is low, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy (specifically, a situation in which road conditions are immediately registered in map data), thereby sufficiently improving the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing method that can automatically update map data with high reliability. In particular, it is possible to provide an information processing method that can reliably reflect road conditions in map data.

[0043] An information processing method for updating map data according to one aspect of the present disclosure is executed by a computer and includes the steps of: acquiring vehicle detection data in which the position of a vehicle and identification information indicating dynamic information shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the position of the vehicle; updating a cumulative voting point by adding a predetermined voting point to a cumulative voting point for determining whether the dynamic information should be associated with the road network element and registered on the map data or whether the dynamic information associated with the road network element should be changed; and, when the cumulative voting point exceeds a predetermined threshold, associating the dynamic information with the road network element and registering it on the map data, or changing the dynamic information associated with the road network element on the map data.

[0044] According to the above method, when identification information indicating dynamic information is included in the vehicle detection data, the cumulative voting points are updated. Then, when the cumulative voting points exceed a predetermined threshold, the dynamic information is associated with a road network element (specifically, a road link or a road node) and registered in the map data, or the dynamic information associated with the road network element in the map data is changed. In this way, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy when the reliability or accuracy of the vehicle detection data is low (specifically, a situation in which dynamic information is immediately reflected in the map data based on vehicle detection data with low reliability or accuracy), thereby sufficiently improving the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing system that can automatically update map data with high reliability. In particular, it is possible to provide an information processing method that can reflect dynamic information in map data with high reliability.

[0045] Also provided is a non-transitory computer readable medium that holds instructions (computer program) that, when executed, cause a computer to perform the information processing method.

[0046] According to one aspect of the present disclosure, there is provided an information processing device including a processor and a memory for storing computer-readable instructions, the computer-readable instructions being executed by the processor to cause the information processing device to perform the information processing method according to any one of claims 12 to 14.

[0047] (Configuration of Information Processing System 1) Hereinafter, an information processing system 1 according to this embodiment will be described with reference to the drawings. FIG. 1 is a diagram showing an example of the configuration of the information processing system 1 according to this embodiment. As shown in FIG. 1, the information processing system 1 includes a vehicle 2, a map update server 3, a people flow data server 4, a map data server 5, a vehicle 6, and a user terminal 7. These are connected to a communication network 8. The vehicle 2, the vehicle 6, and the user terminal 7 are communicatively connected to the map update server 3 and / or the map data server 5 via the communication network 8. The communication network 8 is formed by, for example, the Internet. The information processing system 1 functions as a map update system that can automatically update map data based on vehicle detection data acquired by the vehicle 2.

[0048] (Configuration of Vehicle 2) Next, the hardware configuration of the vehicle 2 will be described below with reference to Fig. 2. Fig. 2 is a diagram showing an example of the configuration of the vehicle 2 according to this embodiment. As shown in Fig. 2, the vehicle 2 includes a vehicle control unit 20, a camera 21, a wireless communication unit 22, a GPS (Global Positioning System) receiver 23, an HMI (Human Machine Interface) 24, a storage device 25, a drive system 26, and a direction sensor 27.

[0049] The vehicle 2 may be a vehicle (e.g., an autonomous vehicle) capable of running in an autonomous driving mode. In this example, a four-wheeled vehicle is used as an example of a vehicle, but the number of wheels of the vehicle 2 is not particularly limited. The vehicle control unit 20 is configured to control various components provided in the vehicle 2, and is configured by, for example, at least one electronic control unit (ECU). The electronic control unit includes a computer system including one or more processors and one or more memories.

[0050] The camera 21 is configured to capture an image of the surrounding environment in front of the vehicle 2. The camera 21 is arranged at a predetermined position on the vehicle 2 so as to capture an image of the surrounding environment in front of the vehicle 2 through, for example, the windshield of the vehicle 2. The image captured by the camera 21 may be a still image or a moving image. The frame rate of the moving image is not particularly limited. The camera 21 may also be detachably mounted on the vehicle 2. In this regard, the camera 21 may be portable by the driver or the like. The camera 21 may also be configured to capture an image of the surrounding environment behind the vehicle 2. In this case, the camera 21 may be arranged at a predetermined position on the vehicle 2 so as to capture an image of the surrounding environment behind the vehicle 2.

[0051] The wireless communication unit 22 is configured to connect the vehicle 2 to the communication network 8, and includes a transmitting / receiving antenna and a wireless transmitting / receiving circuit. The wireless communication unit 22 may be a wireless communication module compatible with a short-range wireless communication standard such as Wi-Fi (registered trademark) or Bluetooth (registered trademark), or may be a wireless communication module compatible with a fourth-generation mobile communication system or a fifth-generation mobile communication system such as LTE.

[0052] The GPS receiver 23 is configured to acquire information related to the current position of the vehicle 2. The HMI 24 is configured to include an input unit that accepts input operations from the driver and an output unit that outputs information related to the driving of the vehicle 2 to the driver. The storage device 25 is an external storage device such as a hard disk drive (HDD) or a solid state drive (SSD). Map data and vehicle control programs may be stored in the storage device 25. The drive system 26 is configured to control the driving state of the vehicle 2. For example, the drive system 26 is configured to control the driving of the vehicle 2 by controlling the accelerator, brake, and steering of the vehicle 2. The direction sensor 27 is configured to detect the traveling direction of the vehicle 2.

[0053] The vehicle 2 transmits vehicle detection data (see FIG. 7 ), which will be described later, to the map update server 3 via the communication network 8. The vehicle 2 also acquires image data and position data, and generates vehicle detection data based on the image data and position data. The image data includes multiple images (still images or frames of video) captured by a camera 21 mounted on the vehicle 2 traveling on a road, and the capture times of the multiple images. In the image data, each of the multiple images is associated with a corresponding one of the multiple capture times. The position data includes multiple positions (longitude and latitude) of the vehicle 2 and the acquisition times of the multiple positions of the vehicle 2. In the position data, each of the multiple positions is associated with a corresponding one of the multiple position acquisition times. Although FIG. 1 illustrates a single vehicle 2, multiple vehicles 2 may transmit vehicle detection data to the map update server 3.

[0054] (Configuration of Map Update Server 3) Next, the hardware configuration of the map update server 3 will be described below with reference to FIG. 3. FIG. 3 is a diagram showing an example of the configuration of the map update server 3. The map update server 3 is configured to determine updates to map data based on vehicle detection data transmitted from the vehicle 2. As shown in FIG. 3, the map update server 3 includes a control unit 30, a storage device 31, an input / output interface 32, a communication unit 33, an input operation unit 34, and a display unit 35. These elements are connected to a communication bus 36. The map update server 3 may be configured by a single server or by multiple servers. The map update server 3 may be constructed on-premise or may be a cloud server.

[0055] The control unit 30 includes a memory and a processor. The memory is configured to store computer-readable instructions (programs). For example, the memory may include a ROM (Read Only Memory) storing various programs and a RAM (Random Access Memory) having multiple work areas for storing various programs executed by the processor. The processor may include at least one of a CPU (Central Processing Unit), an MPU (Micro Processing Unit), and a GPU (Graphics Processing Unit). The CPU may include multiple CPU cores. The GPU may include multiple GPU cores. The processor may be configured to load a specified program from various programs stored in the storage device 31 or the ROM onto the RAM and execute various processes in cooperation with the RAM. The memory may store an information processing program for causing the processor to execute a series of processes (information processing method) executed by the map update server 3 .

[0056] The storage device 31 is a storage device (storage) such as an HDD, SSD, etc., and is configured to store programs and various data. The storage device 31 stores cumulative voting point data and a detected object reference table (see FIG. 8), which will be described later.

[0057] The input / output interface 32 is an interface that enables connection between an external device and the map update server 3, and includes an interface that complies with a predetermined communication standard such as the USB standard or the HDMI (registered trademark) standard. The communication unit 33 may include various wired communication modules for communicating with external terminals on the communication network 8. The input operation unit 34 is, for example, a touch panel, a mouse, and / or a keyboard, and is configured to accept input operations by an operator and to generate operation signals in response to the input operations by the operator. The display unit 35 is, for example, configured by a video display and a video display circuit.

[0058] The people flow data server 4 stores people flow data (see FIG. 9 ). As shown in FIG. 9 , the people flow data includes road link identification information (road link ID), date and time period information, and information indicating traffic volume (number of passing vehicles). In the people flow data, traffic volume is associated with the road link ID and the date and time period information, so traffic volume according to these conditions can be extracted. The people flow data server 4 may have the same hardware configuration as the map update server 3 shown in FIG. 3 .

[0059] The map data server 5 stores map data. The map data includes road network data for route search, background data, note data, address data, and store data. The map data server 5 may have the same hardware configuration as the map update server 3 shown in FIG. 3.

[0060] As shown in FIG. 4 , the road network data is composed of multiple road links (lines) and multiple road nodes (points). A road node is connected to multiple road links. Each road node is connected to other road nodes via one road link. Each road link is assigned unique identification information (ID). Each road link is associated with, for example, road attribute information and road regulation information (e.g., one-way streets, etc.). The road attribute information includes, for example, road type information, route cost information (e.g., distance cost information), number of lanes information, and road width information. Each road node is assigned unique identification information (ID). Each road node is associated with, for example, lane information, direction guidance information, traffic light information, and intersection name information. In this way, the road network data includes information related to road links and information related to road nodes. By using the road network data, a desired route search from a departure point to a destination can be performed.

[0061] The background data includes illustration data related to the visual background of the map (e.g., map shapes such as buildings, ocean, forests, roads, etc.). The annotation data includes text information to be displayed on the map (e.g., names of buildings and mountains, etc.). The address data includes address information associated with each building on the map. The store information data includes information related to stores on the map (e.g., business hours, etc.).

[0062] 1 , the user terminal 7 is a terminal operated by the user U. The user terminal 7 is communicatively connected to the map data server 5 via the communication network 8, and receives map data from the map data server 5. A map application or a web browser for displaying a map may be installed on the user terminal 7.

[0063] Vehicle 6 may be a vehicle capable of running in an autonomous driving mode. Vehicle 6 may have, for example, the hardware configuration of vehicle 2 shown in FIG. 2. Vehicle 6 is communicably connected to map data server 5. Vehicle 6 may receive map data from map data server 5 and update the map data stored in a storage device of vehicle 6. Vehicle 6 is equipped with a car navigation system, and can execute a route search from a departure point to a destination by using road network data included in the map data.

[0064] (First Example of Map Data Update Processing) Next, a first example of map data update processing stored in the map data server 5 will be described below, mainly with reference to FIG. 5 . FIG. 5 is a flowchart for explaining an information processing method for updating map data according to a first example of this embodiment. In the update processing according to this example, static information of the map data is updated. The static information of the map data includes information on roads, structures on the roads, permanent regulations, and logical information virtually generated from roads and structures on the roads. In particular, examples of static information of the map data include signs, directional signs, store signs, buildings, road shapes, traffic lights, crosswalks, etc.

[0065] As shown in Fig. 5, in step S10, the map update server 3 (specifically, the control unit 30 of the map update server 3) receives vehicle detection data from the vehicle 2. As shown in Fig. 7, the vehicle detection data includes position information (longitude and latitude) of the vehicle 2, accuracy information of the vehicle position (in this example, accuracy radius), identification information (ID) of the detected object detected by the vehicle 2, and time information (date and time). The process of generating vehicle detection data will be described below with reference to Fig. 6. Fig. 6 is a flowchart for explaining the process of generating vehicle detection data.

[0066] 6, in step S20, the vehicle control unit 20 of the vehicle 2 acquires image data through the camera 21. The image data is associated with the image captured by the camera 21 and the time of capture. Next, in step S21, the vehicle control unit 20 acquires position data through the GPS receiver 23. The position data is associated with the position of the vehicle 2 and the time the vehicle position was acquired. Furthermore, the image data and the position data are associated with each other through time information.

[0067] In step S22, the vehicle control unit 20 uses an image recognition model to identify the identification information (an example of first identification information) of the detected object shown in the image. The image recognition model is a trained model constructed by machine learning. For example, the image recognition model is constructed using various training images showing detected objects such as signs, directional signs, store signs, buildings, road shapes, traffic lights, crosswalks, road conditions, and emergency vehicles. Images are input to the input layer of the image recognition model, and each unit in the output layer of the image recognition model is associated with the identification information of the detected object. When an image captured by the camera 21 is input to the image recognition model, the ID (identification information) of the detected object shown in the image is output. For example, if a road sign indicating a maximum speed of 40 km / h is shown in the image, the image recognition model outputs the ID corresponding to the road sign indicating a maximum speed of 40 km / h. The image recognition model is stored in the storage device 25 of the vehicle 2.

[0068] Next, in step S23, the vehicle control unit 20 generates vehicle detection data based on the position of the vehicle 2, the position accuracy (position accuracy radius), the ID of the detected object, and time information (photographing time or position acquisition time). In step S24, the vehicle control unit 20 transmits the vehicle detection data to the map update server 3 via the communication network 8. In this way, vehicle detection data in which the position of the vehicle 2 and the ID of the detected object are associated with each other can be automatically generated based on the image data, position data, and image recognition model.

[0069] Returning to FIG. 5 , in step S11, the map update server 3 confirms that the detected object in the vehicle detection data is static information by referring to the detected object reference table (see FIG. 8 ). As shown in FIG. 8 , the detected object reference table includes the detected object ID, detected object information, detected object attributes, a road network element (road link or road node) associated with the detected object, and a dynamic level. Here, the dynamic level indicates whether the detected object belongs to static information, quasi-dynamic information, or dynamic information. For example, if the detected object ID indicated in the vehicle detection data is 10040, the map update server 3 confirms that the detected object is associated with a road link and that the detected object is static information by referring to the detected object reference table. Note that if the detected object is quasi-dynamic information, a map data update process according to a second example (see FIG. 10 ) is executed. If the detected object is dynamic information, a map data update process according to a third example (see FIG. 11 ) is executed. The map data update processes according to the second and third examples will be described later.

[0070] In step S12, the map update server 3 acquires from the map data server 5 a road link ID (an example of identification information of a road network element) corresponding to the position of the vehicle 2 and an ID of a registered object (an example of second identification information) that has the same attribute as the detected object. Specifically, first, the map update server 3 transmits the position information of the vehicle 2 and the ID of the detected object to the map data server 5. Next, the map data server 5 identifies the road link ID corresponding to the position of the vehicle 2 based on the position of the vehicle 2 and the road network data. Each road link has a start point position, an end point position, and the positions of one or more constituent points that are located between the start point and the end point. In this way, the map data server 5 identifies the road link ID that is most relevant to the position of the vehicle 2 from the position information associated with each road link. In particular, the map data server 5 identifies the ID of the road link that has position information that matches the vehicle position or the ID of the road link that has position information that is closest to the vehicle position.

[0071] Next, the map data server 5 acquires from the map data the ID of a registered object that is associated with the identified road link and has the same attribute as the detected object. For example, if the detected object is a road sign indicating a maximum speed of 40 km / h, the attribute of the detected object is the maximum speed road sign. In this case, the map data server 5 acquires from the map data the ID of the maximum speed road sign associated with the identified road link ID. As shown in FIG. 8 , if the detected object is a road sign indicating a maximum speed of 40 km / h, the ID of the detected object is 10040. If the detected object is a road sign indicating a maximum speed of 30 km / h, the ID of the detected object is 10030. In this regard, if the maximum speed road sign on the road associated with the identified road link ID has recently changed from 30 km / h to 40 km / h, the ID of the detected object detected by the vehicle 2 will be 10040, while the ID of the registered object registered in the map data will be 10030.

[0072] In this way, the map data server 5 transmits the road link ID and the ID of the registered object to the map update server 3. In step S13, the map update server 3 determines whether the ID of the detected object matches the ID of the registered object. For example, if the road signs on the road associated with the identified road link ID have not been changed, the ID of the road sign as the detected object detected by the vehicle 2 will match the ID of the road sign as the registered object registered in the map data. On the other hand, if the road signs on the road associated with the road link ID in the real world have recently been changed, the ID of the road sign detected by the vehicle 2 will differ from the ID of the road sign registered in the map data.

[0073] If the ID of the detected object matches the ID of the registered object (YES in step S13), this process ends. On the other hand, if the ID of the detected object does not match the ID of the registered object (NO in step S13), the map update server 3 determines voting points to be added to the cumulative voting points of the cumulative voting point data stored in the storage device 31 (step S14). In particular, the map update server 3 may determine voting points according to at least one of the vehicle traffic volume associated with the road link and the position accuracy of the vehicle 2 (e.g., accuracy radius).

[0074] In this regard, the map update server 3 obtains traffic volume associated with the road link from the people flow data server 4. Specifically, the map update server 3 transmits a road link ID and time information (date and time) to the people flow data server 4. Here, the time information corresponds to the time when the image showing the detected object was captured by the camera 21 of the vehicle 2. The people flow data server 4 identifies information related to traffic volume based on the road link ID and the time information by referencing the people flow data. More specifically, the people flow data server 4 obtains the traffic volume of vehicles associated with the road link ID for the time period including the time when the image showing the detected object was captured. For example, as shown in FIG. 9 , the traffic volume (number of passing vehicles) on the road associated with road link ID: X00001 from 02:00 to 03:00 on January 1st is 150. The people flow data server 4 then transmits the traffic volume information to the map update server 3.

[0075] Furthermore, the position accuracy (accuracy radius) of vehicle 2 is included in the vehicle detection data. When the voting points are determined based on both the vehicle traffic volume and the position accuracy of vehicle 2, the map update server 3 may determine the voting points P according to the following formulas (1) to (3): P = 1 × γ1 × γ2 (1) γ1 = 1 / R × a1 (2) γ2 = 1 / V × a2 (3) Here, γ1 is a weighting coefficient related to the accuracy radius. γ2 is a weighting coefficient related to traffic volume. R is the accuracy radius. V is traffic volume. a1 and a2 are constants.

[0076] In this way, since the voting points are determined according to the traffic volume and the position accuracy of the vehicle 2, it is possible to optimize the voting points according to the traffic volume and the position accuracy of the vehicle 2. For example, if the voting points for roads with low traffic volume are higher than the voting points for roads with high traffic volume, it is possible to effectively prevent a situation in which the update of information related to road link IDs with low traffic volume is significantly delayed compared to the update of information related to road link IDs with high traffic volume. Furthermore, if the voting points are set lower as the position accuracy of the vehicle 2 becomes lower, it is possible to sufficiently increase the reliability of the automatic update process of the map data.

[0077] Furthermore, since the traffic volume associated with a road link varies depending on the time period, it is possible to further optimize voting points according to the traffic volume during the time period including the time of the photo shoot. Note that the people flow data does not need to include date information. In this case, the people flow data is composed of a road link ID, time period information, and traffic volume information.

[0078] Next, in step S15, the map update server 3 updates the cumulative voting points by adding the voting points determined in step S14 to the cumulative voting points included in the cumulative voting point data. The cumulative voting points are used to determine whether a registered object associated with a road link ID or a road node ID should be changed to a detected object. In the cumulative voting point data stored in the storage device 31, the road link ID and / or road node ID, the detected object ID, the registered object ID, information indicating the cumulative voting points, and information indicating a predetermined threshold may be associated with each other. For example, in the case where the road link ID is X00001, the detected object ID is 10040, and the voting points are 0.8, 0.8 voting points are added to the cumulative voting points associated with the road link ID: X00001 and the detected object ID: 10040 in the cumulative voting point data.

[0079] Next, in step S16, the map update server 3 determines whether the cumulative voting points included in the cumulative voting point data exceed a predetermined threshold Pth1. If the determination result in step S16 is NO, this process ends. On the other hand, if the cumulative voting points exceed the threshold Pth1 (if the determination result in step S16 is YES), the map update server 3 changes the registered objects linked to the road link IDs in the map data to detected objects (step S17). More specifically, the map update server 3 transmits a change request to the map data server 5 to change the registered objects to detected objects. In response to the received change request, the map data server 5 updates the map data so as to change the registered objects linked to the road link IDs in the map data to detected objects. For example, if the registered object in road link: X00001 is a road sign indicating a maximum speed of 30 km / h, while the detected object is a road sign indicating a maximum speed of 40 km / h, the registered object ID: 10030 associated with road link ID: X00001 is changed to the detected object ID: 10040. In this way, the map data update process is automated. The updated map data may be periodically distributed to the user terminal 7 and the vehicle 6.

[0080] Furthermore, after the map data is updated, the value of the cumulative voting points associated with the road link ID and the registered object ID may be reset.

[0081] According to a first example of the map data update process, if the registered object ID and the detected object ID do not match, the cumulative voting points are updated. Then, if the cumulative voting points exceed a predetermined threshold, the registered object associated with the road link or road node in the map data is changed to a detected object. In this way, when the reliability or accuracy of vehicle detection data is low, it is possible to effectively prevent a situation in which map data is immediately updated based on vehicle detection data with low reliability or accuracy, thereby sufficiently improving the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing system that can automatically update map data with high reliability.

[0082] In this example, a registered object associated with a road link on the map data is changed to a detected object, but a registered object associated with a road node may also be changed to a detected object. For example, if a predetermined sign placed at an intersection is associated with a road node, the predetermined sign may be changed on the map data through this process.

[0083] (Second Example of Map Data Update Processing) Next, a second example of the update processing of map data stored in the map data server 5 will be described below, mainly with reference to FIG. 10 . FIG. 10 is a flowchart for explaining an information processing method for updating map data according to the second example of this embodiment. In the update processing according to this example, semi-dynamic information of the map data is updated. In the semi-dynamic information of the map data, the position, range, and appearance time (time period) of the target event are not constant, but appear, disappear, move, expand, contract, etc. In particular, one example of the semi-dynamic information of the map data is information indicating the condition of the road, such as deterioration of the road surface, road construction, flooding of the road, traffic restrictions, fallen objects on the road, and broken-down vehicles parked on the road.

[0084] As shown in Fig. 10 , in step S30, the map update server 3 (specifically, the control unit 30 of the map update server 3) receives vehicle detection data from the vehicle 2. As shown in Fig. 7 , the vehicle detection data includes position information (longitude and latitude) of the vehicle 2, accuracy information of the vehicle position (in this example, accuracy radius), an ID of a detected object detected by the vehicle 2, and time information (date and time). In this example, an ID indicating the state of the road is included as a detected object in the vehicle detection data. As shown in Fig. 8 , for example, ID: 80000 indicating road surface deterioration is included as a detected object in the vehicle detection data.

[0085] The vehicle detection data is generated by the vehicle 2 through a series of processes shown in FIG. 6 . The series of processes shown in FIG. 6 has already been described in detail in the first example of the map data update process, and therefore will not be described in detail here. Meanwhile, in step S22, the vehicle control unit 20 uses an image recognition model to identify identification information indicating the state of the road as a detected object shown in the image. For example, the image recognition model is constructed using various learning images indicating detected objects such as signs, direction signs, store signs, buildings, road shapes, traffic lights, crosswalks, road conditions (e.g., road surface deterioration, road construction, flooded roads, etc.), and emergency vehicles. An image is input to the input layer of the image recognition model, and each unit in the output layer of the image recognition model is associated with the identification information of the detected object. When an image captured by the camera 21 is input to the image recognition model, the ID of the detected object shown in the image is output. For example, if road surface deterioration is shown in the image, the image recognition model outputs an ID indicating the road surface deterioration.

[0086] 10 , in step S31, the map update server 3 refers to the detected object reference table (see FIG. 8 ) to confirm that the detected object in the vehicle detection data is quasi-dynamic information. As shown in FIG. 8 , for example, when the ID of the detected object indicated in the vehicle detection data is 80000, the map update server 3 refers to the detected object reference table to confirm that the detected object indicates road surface deterioration, that the attribute of the detected object is the road surface condition, that the detected object is associated with a road link, and that the detected object is quasi-dynamic information.

[0087] In step S32, the map update server 3 acquires a road link ID (an example of identification information of a road network element) corresponding to the position of the vehicle 2 from the map data server 5. Specifically, first, the map update server 3 transmits position information of the vehicle 2 to the map data server 5. Next, the map data server 5 identifies the road link ID corresponding to the position of the vehicle 2 based on the position of the vehicle 2 and the road network data. In particular, the map data server 5 identifies the ID of the road link having position information that matches the vehicle position or the ID of the road link having position information that is closest to the vehicle position.

[0088] Next, in step S33, the map update server 3 determines the voting points to be added to the cumulative voting points of the cumulative voting point data stored in the storage device 31. In particular, the map update server 3 may determine the voting points according to at least one of the traffic volume associated with the road link and the position accuracy of the vehicle 2 (e.g., accuracy radius).

[0089] In this regard, the map update server 3 obtains traffic volume associated with the road link from the people flow data server 4. Specifically, the map update server 3 transmits a road link ID and time information (date and time) to the people flow data server 4. Here, the time information corresponds to the time when the image showing road surface deterioration was taken by the camera 21 of the vehicle 2. The people flow data server 4 identifies information related to traffic volume based on the road link ID and the time information by referring to the people flow data. When the voting point is determined based on both the traffic volume and the position accuracy of the vehicle 2, the map update server 3 may determine the voting point P based on the above formulas (1) to (3).

[0090] Next, in step S34, the map update server 3 updates the cumulative voting points by adding the voting points determined in step S33 to the cumulative voting points included in the cumulative voting point data. The cumulative voting points are used to determine whether road conditions such as road surface deterioration should be associated with road link IDs and registered in the map data. In the cumulative voting point data stored in the storage device 31, road link IDs and / or road node IDs, detected object IDs, information indicating cumulative voting points, and information indicating a predetermined threshold may be associated with each other. For example, if the road link ID is X00001, the ID indicating road surface deterioration is 80000, and the voting points are 0.7, 0.7 voting points are added to the cumulative voting points associated with the road link ID: X00001 and the ID indicating road surface deterioration: 80000 in the cumulative voting point data. Furthermore, the threshold Pth2 associated with IDs indicating quasi-dynamic information such as road conditions is set lower than the threshold Pth1 associated with IDs indicating static information such as road signs. That is, since the update timing of the semi-dynamic information needs to be earlier than the update timing of the static information, the threshold value Pth2 related to the semi-dynamic information is set lower than the threshold value Pth1 related to the static information. In other words, since the static information continues to be registered in the map data on an annual basis, the threshold value Pth1 related to the static information is set higher than the threshold value Pth2.

[0091] Next, in step S35, the map update server 3 determines whether the cumulative voting points included in the cumulative voting point data exceed the threshold value Pth2. If the determination result in step S35 is NO, this process ends. On the other hand, if the cumulative voting points exceed the threshold value Pth2 (if the determination result in step S35 is YES), the map update server 3 associates road conditions such as road surface deterioration with road links and registers them in the map data (step S36). More specifically, the map update server 3 transmits a registration request to the map data server 5 to register road conditions such as road surface deterioration in the map data. In response to the received registration request, the map data server 5 updates the map data by registering road conditions such as road surface deterioration with road links in the map data. For example, ID: 80000 indicating road surface deterioration is associated with road link ID: X00001 and registered in the map data. In this way, the map data update process is automated. The updated map data may be periodically distributed to the user terminal 7 and the vehicle 6. The road administrator can check the information indicating road surface deterioration that has been newly registered on the map data via the user terminal 7, and can therefore quickly identify roads with deteriorated road surfaces.

[0092] Furthermore, after the map data is updated, the values ​​of the cumulative voting points associated with the road link ID and the ID indicating the road condition such as road surface deterioration, which are included in the cumulative voting point data, may be reset. Furthermore, if the process of registering the road condition such as road surface deterioration in the map data (i.e., the process of step S36) is not executed for a predetermined period (e.g., several weeks), the map update server 3 deletes the road surface condition such as road surface deterioration associated with the road link from the map data.

[0093] After the road conditions change (for example, after the road conditions, such as road surface deterioration, have been improved by road construction or the like), the vehicle 2 does not detect the road surface deterioration, and therefore the process of registering the road conditions in the map data is not executed for a predetermined period of time. Therefore, the road conditions associated with the road link are automatically deleted from the map data. In this way, it is possible to effectively prevent a situation in which information related to the road conditions continues to remain in the map data after the road conditions (for example, road surface deterioration) have changed.

[0094] According to the second example of the map data update process, if the vehicle detection data includes an ID indicating a road condition (e.g., road surface deterioration), the cumulative voting points are updated. Then, if the cumulative voting points exceed a threshold Pth2, the road condition is associated with the road link and registered in the map data. In this way, it is possible to effectively prevent a situation in which the map data is immediately updated based on vehicle detection data with low reliability or accuracy (specifically, a situation in which the road condition is immediately registered in the map data) when the reliability or accuracy of the vehicle detection data is low. This makes it possible to sufficiently improve the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing system 1 that can automatically update map data with high reliability. In particular, it is possible to provide an information processing system 1 that can reliably reflect road conditions in the map data.

[0095] (Third Example of Map Data Update Processing) Next, a third example of the update processing of map data stored in the map data server 5 will be described below, mainly with reference to FIG. 11 . FIG. 11 is a flowchart for explaining an information processing method for updating map data according to the third example of this embodiment. In the update processing according to this example, dynamic information of the map data is updated. In the dynamic information of the map data, the position of an object is not fixed but moves. Furthermore, even if the position of an object is fixed, the update cycle of the attributes held is short, and changes in position and updates of attributes can change independently of the object. In particular, examples of dynamic information of the map data include the presence of emergency vehicles and variable signs.

[0096] As shown in FIG. 11 , in step S40, the map update server 3 (specifically, the control unit 30 of the map update server 3) receives vehicle detection data from the vehicle 2. As shown in FIG. 7 , the vehicle detection data includes position information (longitude and latitude) of the vehicle 2, accuracy information of the vehicle position (in this example, the accuracy radius), the ID of the detected object detected by the vehicle 2, and time information (date and time). In this example, the vehicle detection data includes an ID indicating the presence of an emergency vehicle (for example, an ambulance, fire engine, or police car traveling in an emergency) as the detected object. As shown in FIG. 8 , for example, the vehicle detection data includes an ID: 90000 indicating the presence of an emergency vehicle as the detected object.

[0097] The vehicle detection data is generated by the vehicle 2 through a series of processes shown in FIG. 6 . The series of processes shown in FIG. 6 has already been described in detail in the first example of the map data update process, and therefore will not be described in detail here. Meanwhile, in step S22, the vehicle control unit 20 uses an image recognition model to identify identification information indicating the presence of an emergency vehicle as a detected object shown in the image. For example, the image recognition model is constructed using various training images indicating detected objects such as signs, directional signs, store signs, buildings, road shapes, traffic signals, crosswalks, road conditions (e.g., road surface deterioration, road construction, flooded roads, etc.), and emergency vehicles. An image is input to the input layer of the image recognition model, and each unit in the output layer of the image recognition model is associated with the identification information of the detected object. When an image captured by the camera 21 is input to the image recognition model, the ID of the detected object shown in the image is output. For example, if an emergency vehicle is shown in the image, the image recognition model outputs an ID indicating the presence of the emergency vehicle.

[0098] 10 , in step S41, the map update server 3 refers to the detected object reference table (see FIG. 8 ) to confirm that the detected object in the vehicle detection data is dynamic information. As shown in FIG. 8 , for example, when the ID of the detected object indicated in the vehicle detection data is 90000, the map update server 3 refers to the detected object reference table to confirm that the detected object indicates the presence of an emergency vehicle traveling in an emergency, that the attribute of the detected object is an emergency vehicle, that the detected object is associated with a road link / road node, and that the detected object is dynamic information.

[0099] In step S42, the map update server 3 acquires a road link ID or a road node ID (an example of identification information of a road network element) corresponding to the position of the vehicle 2 from the map data server 5. Specifically, first, the map update server 3 transmits position information of the vehicle 2 to the map data server 5. Next, the map data server 5 identifies a road link ID or a road node ID corresponding to the position of the vehicle 2 based on the position of the vehicle 2 and the road network data. In particular, the map data server 5 identifies a road link or a road node having position information that matches the vehicle position, or a road link or a road node having position information that is closest to the vehicle position.

[0100] Next, in step S43, the map update server 3 updates the cumulative voting points by adding the voting points to the cumulative voting points included in the cumulative voting point data. The cumulative voting points are used to determine whether the presence of an emergency vehicle should be registered in the map data in association with a road link ID or a road node ID. In the cumulative voting point data stored in the storage device 31, the road link ID and / or road node ID, the ID of the detected object, information indicating the cumulative voting points, and information indicating a predetermined threshold may be associated with each other. For example, if the road link ID is X00001, the ID indicating the presence of an emergency vehicle is 90000, and the voting point is 1, in the cumulative voting point data, 1 voting point is added to the cumulative voting points associated with the road link ID: X00001 and the ID: 90000 indicating the presence of an emergency vehicle. In addition, the threshold Pth3 associated with the ID indicating dynamic information such as the presence of an emergency vehicle is set lower than the threshold Pth2 associated with the ID indicating quasi-dynamic information such as road surface conditions. In other words, since the timing of updating dynamic information requires more immediacy than the timing of updating semi-dynamic information, the threshold value Pth3 related to dynamic information is set lower than the threshold value Pth1 related to static information and the threshold value Pth2 related to semi-dynamic information (Pth3 < Pth2 < Pth1).

[0101] Next, in step S44, the map update server 3 determines whether the cumulative voting points included in the cumulative voting point data exceed the threshold value Pth3. If the determination result in step S44 is NO, this process ends. On the other hand, if the cumulative voting points exceed the threshold value Pth3 (the determination result in step S44 is YES), the map update server 3 registers the presence of the emergency vehicle in the map data by associating it with a road link or a road node (step S45). More specifically, the map update server 3 transmits a registration request to the map data server 5 to register the presence of the emergency vehicle in the map data. In response to the received registration request, the map data server 5 updates the map data by registering the presence of the emergency vehicle in the map data by associating it with the road link. For example, ID: 90000 indicating the presence of the emergency vehicle is registered in the map data by being associated with road link ID: X00001. In this way, the map data update process is automated. The updated map data may be periodically distributed to the user terminal 7 and the vehicle 6. The vehicle 6 can check the information indicating the presence of the emergency vehicle newly registered on the map data, and can therefore determine a travel route that takes the presence of the emergency vehicle into consideration. For example, the vehicle 6 can predict the travel route of the emergency vehicle and then determine a travel route that does not overlap with any part of the travel route of the emergency vehicle.

[0102] Furthermore, after the map data is updated, the values ​​of the cumulative voting points associated with the road link ID and the ID indicating the presence of the emergency vehicle included in the cumulative voting point data may be reset. Furthermore, if the process of registering the presence of an emergency vehicle in the map data (i.e., the process of step S45) is not executed for a predetermined period (e.g., several minutes), the map update server 3 deletes the emergency vehicle associated with the road link or road node from the map data.

[0103] After the emergency vehicle passes through the intersection associated with the road or road node associated with the road link, the registration process of step S45 is not executed for a predetermined period of time, and the existence of the emergency vehicle associated with the road link or road node is deleted from the map data. In this way, it is possible to effectively prevent a situation in which information about the emergency vehicle remains in the map data while being associated with the road link or road node.

[0104] According to the third example of the map data update process, if the vehicle detection data includes an ID indicating the presence of an emergency vehicle, the cumulative voting points are updated. Then, if the cumulative voting points exceed a threshold Pth3, the presence of the emergency vehicle is associated with a road link or road node and registered in the map data. In this way, it is possible to effectively prevent a situation in which the map data is immediately updated based on vehicle detection data with low reliability or accuracy when the vehicle detection data is unreliable or accurate (specifically, a situation in which the driving status of an emergency vehicle is immediately reflected in the map data based on vehicle detection data with low reliability or accuracy), thereby sufficiently improving the reliability of the automatic map data update process. Therefore, it is possible to provide an information processing system 1 that can automatically update map data with high reliability. In particular, it is possible to provide an information processing system 1 that can reliably reflect the driving status of an emergency vehicle in the map data.

[0105] In this example, the presence of an emergency vehicle is cited as an example of dynamic information of map data, but the dynamic information may also be a variable sign. Here, a variable sign is a sign (e.g., a sign indicating a speed limit or a direction of travel) that changes depending on the time of day or road conditions. The display of the variable sign may be controlled in response to an external control signal. If the dynamic information is a variable sign, in step S45, the map update server 3 may change the variable sign associated with the road link or road node. More specifically, the map update server 3 transmits a change request to the map data server 5 to change the variable sign in the map data. In response to the received change request, the map data server 5 updates the map data by changing the variable sign associated with the road link or road node in the map data. In this way, it is possible to provide an information processing system 1 that can automatically update map data in accordance with the state of a variable sign.

[0106] Although the embodiments of the present invention have been described above, the technical scope of the present invention should not be construed as being limited by the description of the present embodiments. The present embodiments are merely examples, and it will be understood by those skilled in the art that various modifications of the embodiments are possible within the scope of the invention described in the claims. The technical scope of the present invention should be determined based on the scope of the invention described in the claims and its equivalents.

[0107] This application appropriately incorporates the contents disclosed in Japanese Patent Application No. 2024-027022 filed on February 26, 2024.

Claims

1. An information processing system for updating map data, comprising: acquiring vehicle detection data in which the position of a vehicle and first identification information of a detected object shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the position of the vehicle; acquiring second identification information of a registered object associated with the road network element and having the same attributes as the detected object from the map data; determining whether the second identification information and the first identification information match; if the second identification information and the first identification information do not match, updating the cumulative voting points for determining whether the registered object linked to the road network element on the map data should be changed to the detected object by adding a predetermined voting point to the cumulative voting point; and if the cumulative voting point exceeds a predetermined threshold, changing the registered object linked to the road network element on the map data to the detected object.

2. The information processing system of claim 1, wherein the information processing system acquires image data in which the image captured by the camera and the time of capture are associated with each other, acquires location data in which the position of the vehicle and the time the position of the vehicle was acquired are associated with each other, identifies first identification information of the detected object shown in the image using an image recognition model, and generates the vehicle detection data in which the position of the vehicle and the first identification information of the detected object are associated with each other.

3. The information processing system according to claim 1 or 2, wherein the information processing system determines the predetermined voting point depending on at least one of the traffic volume associated with the road network element and the accuracy of the vehicle's position.

4. The information processing system according to claim 3, wherein the information processing system determines the predetermined voting point according to the traffic volume associated with the road network element during a time period including the time when the image showing the detected object was taken.

5. An information processing system for updating map data, comprising: acquiring vehicle detection data in which a vehicle position and identification information indicating a road condition shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the vehicle position; updating an accumulated voting point for determining whether the road condition should be associated with the road network element and registered on the map data by adding a predetermined voting point to the accumulated voting point; and, when the accumulated voting point exceeds a predetermined threshold, associating the road condition with the road network element and registering it on the map data.

6. The information processing system according to claim 5, wherein the information processing system deletes the road state associated with the road network element from the map data if the process of registering the road state on the map data is not executed for a predetermined period of time.

7. The information processing system according to claim 5 or 6, wherein the information processing system acquires image data in which the image captured by the camera and the time of capture are associated with each other, acquires location data in which the position of the vehicle and the time the vehicle's position was acquired are associated with each other, identifies identification information indicating the state of the road shown in the image using an image recognition model, and generates the vehicle detection data in which the position of the vehicle and the identification information indicating the state of the road are associated with each other.

8. The information processing system according to claim 5 or 6, wherein the information processing system determines the predetermined voting point depending on at least one of the traffic volume associated with the road network element and the accuracy of the vehicle's position.

9. An information processing system for updating map data, comprising: acquiring vehicle detection data in which a vehicle position and identification information indicating dynamic information shown in an image captured by a camera disposed on the vehicle are associated with each other; determining identification information of a road network element corresponding to the vehicle position; updating an accumulated voting point by adding a predetermined voting point to the accumulated voting point for determining whether the dynamic information should be associated with the road network element and registered in the map data, or whether the dynamic information associated with the road network element should be changed; and, when the accumulated voting point exceeds a predetermined threshold, associating the dynamic information with the road network element and registering it in the map data, or changing the dynamic information associated with the road network element in the map data.

10. The information processing system according to claim 9, wherein the information processing system deletes the dynamic information associated with the road network element from the map data if the process of associating the dynamic information with the road network element and registering it in the map data is not executed for a predetermined period of time.

11. The information processing system according to claim 9 or 10, wherein the information processing system acquires image data in which the image captured by the camera and the time of capture are associated with each other, acquires location data in which the position of the vehicle and the time the position of the vehicle was acquired are associated with each other, identifies identification information indicative of the dynamic information shown in the image using an image recognition model, and generates the vehicle detection data in which the position of the vehicle and the identification information indicative of the dynamic information are associated with each other.

12. An information processing method for updating map data, the information processing method being executed by a computer, comprising: a step of acquiring vehicle detection data in which the position of a vehicle and first identification information of a detected object shown in an image captured by a camera disposed on the vehicle are associated with each other; a step of determining identification information of a road network element corresponding to the position of the vehicle; a step of acquiring second identification information of a registered object associated with the road network element and having the same attributes as the detected object from the map data; a step of determining whether the second identification information and the first identification information match; a step of updating a cumulative voting point for determining whether the registered object linked to the road network element in the map data should be changed to the detected object by adding a predetermined voting point to the cumulative voting point when the second identification information and the first identification information do not match; and a step of changing the registered object linked to the road network element in the map data to the detected object when the cumulative voting point exceeds a predetermined threshold.

13. An information processing method for updating map data, the information processing method being executed by a computer, comprising: a step of acquiring vehicle detection data in which the position of a vehicle and identification information indicating the state of a road shown in an image captured by a camera disposed on the vehicle are associated with each other; a step of determining identification information of a road network element corresponding to the position of the vehicle; a step of updating an accumulated voting point for determining whether the state of the road should be associated with the road network element and registered in the map data by adding a predetermined voting point to the accumulated voting point; and a step of associating the state of the road with the road network element and registering it in the map data when the accumulated voting point exceeds a predetermined threshold.

14. An information processing method for updating map data, the information processing method being executed by a computer, comprising: a step of acquiring vehicle detection data in which a vehicle position and identification information indicating dynamic information shown in an image captured by a camera disposed on the vehicle are associated with each other; a step of determining identification information of a road network element corresponding to the vehicle position; a step of updating an accumulated voting point by adding a predetermined voting point to an accumulated voting point for determining whether the dynamic information should be associated with the road network element and registered in the map data, or whether the dynamic information associated with the road network element should be changed; and a step of associating the dynamic information with the road network element and registering it in the map data, or changing the dynamic information associated with the road network element in the map data, when the accumulated voting point exceeds a predetermined threshold.

15. A non-transitory computer-readable medium bearing instructions that, when executed, cause a computer to perform the information processing method of any one of claims 12 to 14.

16. An information processing device comprising: a processor; and a memory that stores computer-readable instructions, wherein when the computer-readable instructions are executed by the processor, the information processing device performs the information processing method according to any one of claims 12 to 14.

Images