Vehicle-mounted device, provision device, communication system, vehicle assistance method, vehicle assistance program, and communication prediction program

Abstract

This vehicle-mounted device mounted on a vehicle comprises an acquisition unit that acquires, from a provision device, a prediction result regarding the communication quality required for a service executed in the vehicle and auxiliary information used for a determination about the prediction result.

Description

In-vehicle device, providing device, communication system, vehicle assistance method, vehicle assistance program, and communication prediction program

[0001] This application claims priority based on Japanese Patent Application No. 2024-123258, filed on July 30, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0002] Patent Literature 1 (JP 2019-24154 A) discloses the following technology. That is, a quality control device is a quality control device that predicts communication quality to be used for service control in a system in which a data distribution service is provided via a network, and includes: a collection means that collects actual measured communication quality, which is communication quality actually measured in a client terminal that uses the service, and communication environment information of the client terminal, and stores the collected information in a quality DB; and a communication quality prediction means that acquires the actual measured communication quality from the quality DB based on the communication environment information of the client terminal that is the target of communication quality prediction, calculates from the actual measured communication quality a first predicted communication quality, which is predicted communication quality for a time segment at a prediction target time, and a second predicted communication quality, which is predicted communication quality for an adjacent time segment, and calculates a final predicted communication quality to be used for the service control using the first predicted communication quality and the second predicted communication quality.

[0003] Japanese Patent Application Laid-Open No. 2019-24154

[0004] The on-board device of the present disclosure is an on-board device mounted in a vehicle, and includes an acquisition unit that acquires, from a providing device, a prediction result regarding the communication quality required for a service executed in the vehicle, and auxiliary information used to make a judgment regarding the prediction result.

[0005] One aspect of the present disclosure can be realized not only as an in-vehicle device including such a characteristic processing unit, but also as a semiconductor integrated circuit that realizes part or all of the in-vehicle device.

[0006] One aspect of the present disclosure can be realized not only as a providing device equipped with such a characteristic processing unit, but also as a method having such characteristic processing as steps, or as a semiconductor integrated circuit that realizes part or all of the providing device.

[0007] FIG. 1 is a diagram illustrating an example of the configuration of a communication system according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of the configuration of an in-vehicle device according to an embodiment of the present disclosure. FIG. 3 is a diagram illustrating an example of vehicle information and communication information included in request information transmitted by an in-vehicle device according to an embodiment of the present disclosure. FIG. 4 is a diagram illustrating an example of prediction target information included in request information transmitted by an in-vehicle device according to an embodiment of the present disclosure. FIG. 5 is a diagram illustrating a geohash value calculated by a comparative example of an in-vehicle device. FIG. 6 is a diagram illustrating a geohash value calculated by an in-vehicle device according to an embodiment of the present disclosure. FIG. 7 is a diagram illustrating an example of the configuration of a server according to an embodiment of the present disclosure. FIG. 8 is a diagram illustrating an example of auxiliary information created by a server according to an embodiment of the present disclosure. FIG. 9 is a diagram illustrating correction information created by a server according to an embodiment of the present disclosure. FIG. 10 is a diagram illustrating an example of a communication speed prediction result and correction information included in prediction result information received from a server by an in-vehicle device according to an embodiment of the present disclosure. FIG. 11 is a diagram illustrating an example of a communication speed prediction result after correction processing by an in-vehicle device according to an embodiment of the present disclosure. Fig. 12 is a diagram showing another example of a prediction result of a communication speed after correction processing by an in-vehicle device according to an embodiment of the present disclosure. Fig. 13 is a flowchart defining an example of an operation procedure when an in-vehicle device according to an embodiment of the present disclosure makes a determination on a prediction result of communication quality. Fig. 14 is a flowchart defining an example of an operation procedure when an in-vehicle device according to an embodiment of the present disclosure makes a determination on a prediction result of communication quality. Fig. 15 is a flowchart defining an example of an operation procedure when a server according to an embodiment of the present disclosure performs prediction processing. Fig. 16 is a diagram showing an example of a processing sequence of a navigation device, an in-vehicle device, and a server in a communication system according to an embodiment of the present disclosure.

[0008] Conventionally, techniques have been developed for predicting the communication quality required for providing a service.

[0009] [Problem to be Solved by the Present Disclosure] In recent years, development of vehicles called connected cars that function as ICT (Information and Communication Technology) terminals has been progressing. For example, connected cars provide various services by transmitting and receiving information to and from other devices.

[0010] In a connected car, if communication quality deteriorates during service provision or if the connected car travels in an area outside of communication range, problems such as failure to provide the service may occur. Therefore, it is desirable to predict communication quality before providing the service. However, if the accuracy of the communication quality prediction result is low, it is difficult to solve the above problem.

[0011] The present disclosure has been made to solve the above-mentioned problems, and its purpose is to provide an in-vehicle device, a providing device, a communication system, a vehicle assistance method, a vehicle assistance program, and a communication prediction program that can more reliably provide services in vehicles.

[0012] Effect of the Present Disclosure According to the present disclosure, services can be provided more reliably in vehicles.

[0013] [Description of Embodiments of the Present Disclosure] First, the contents of the embodiments of the present disclosure will be listed and described. (1) An in-vehicle device according to an embodiment of the present disclosure is an in-vehicle device mounted in a vehicle, and includes an acquisition unit that acquires, from a providing device, a prediction result of communication quality required for a service executed in the vehicle and auxiliary information used to determine the prediction result.

[0014] With this configuration, the predicted communication quality can be determined using the auxiliary information, and whether or not to provide a service can be determined based on the determined result, thereby more reliably providing the service in the vehicle.

[0015] (2) In the above (1), the auxiliary information may include information that allows the reliability of the prediction result to be determined.

[0016] With this configuration, it is possible to easily determine whether or not a service should be executed depending on the reliability of the predicted communication quality.

[0017] (3) In the above (1) or (2), the auxiliary information may include information used to correct the prediction result.

[0018] With this configuration, it is possible to correct the predicted results of communication quality, thereby more reliably reducing the possibility that the provision of a service will fail midway.

[0019] (4) In any one of (1) to (3) above, the auxiliary information may include a size of a region for predicting the communication quality.

[0020] With this configuration, the prediction result can be evaluated more accurately depending on the size of the region to be predicted.

[0021] (5) In any one of (1) to (4) above, the auxiliary information may include the number of pieces of data used to predict the communication quality.

[0022] With this configuration, the prediction result can be evaluated more accurately depending on the number of data used for the prediction.

[0023] (6) In any of (1) to (5) above, the auxiliary information may include a ratio of the number of data in the time period for which the communication quality is to be predicted to the number of data used to predict the communication quality in a time period other than the time period for which the communication quality is to be predicted.

[0024] With this configuration, the prediction result can be evaluated more accurately depending on the magnitude relationship between the number of data in the time period to be predicted and the number of data in other time periods.

[0025] (7) In any of (1) to (6) above, the auxiliary information may include information indicating a judgment result regarding the past prediction result.

[0026] With this configuration, the current prediction result can be evaluated more accurately depending on the judgment result regarding the past prediction result.

[0027] (8) In any of (1) to (7) above, the prediction result may be an output value of the trained model when the providing device predicts the communication quality using the trained model.

[0028] With this configuration, for example, the output value of a trained model trained using the latest measurement results of communication quality can be obtained as a prediction result, thereby enabling more accurate prediction of communication quality.

[0029] (9) In any of (1) to (8) above, the service may be executed by communication conforming to TCP, and the auxiliary information may include a parameter related to a limit on the communication speed at the start of execution of the service.

[0030] With this configuration, when a service is executed through communication conforming to TCP, the prediction result can be evaluated more accurately in consideration of changes in the communication speed.

[0031] (10) In any of (1) to (9) above, the acquisition unit may acquire a plurality of prediction results corresponding to a plurality of times included in the time period for which the communication quality is predicted, and the auxiliary information for each of the times.

[0032] With this configuration, a judgment can be made about the prediction results for each time period included in the time period being predicted, making it possible to determine, for example, the time when there is a high probability that the provision of a service will fail.

[0033] (11) A providing device according to an embodiment of the present disclosure includes a prediction unit that predicts the communication quality required for a service executed in a vehicle, a creation unit that creates auxiliary information used to judge the prediction result of the prediction unit, and a transmission unit that transmits the prediction result of the prediction unit and the auxiliary information created by the creation unit.

[0034] With this configuration, the predicted communication quality can be determined using the auxiliary information, and whether or not to provide a service can be determined based on the determined result, thereby more reliably providing the service in the vehicle.

[0035] (12) A communication system according to an embodiment of the present disclosure includes an on-board device mounted in a vehicle and a providing device, and the providing device transmits to the on-board device a prediction result of the communication quality required for a service executed in the vehicle and auxiliary information used to make a judgment about the prediction result.

[0036] With this configuration, the predicted communication quality can be determined using the auxiliary information, and whether or not to provide a service can be determined based on the determined result, thereby more reliably providing the service in the vehicle.

[0037] (13) A vehicle assistance method according to an embodiment of the present disclosure is a vehicle assistance method in an on-board device mounted in a vehicle, and includes a step of obtaining, from a providing device, a prediction result regarding the communication quality required for a service executed in the vehicle and auxiliary information used to make a judgment regarding the prediction result.

[0038] This method allows the prediction result of communication quality to be determined using the auxiliary information, and it is then possible to determine whether or not to provide a service based on the prediction result, thereby enabling the service to be provided more reliably in the vehicle.

[0039] (14) A vehicle assistance program according to an embodiment of the present disclosure is a vehicle assistance program used in an on-board device mounted in a vehicle, and is a program for causing a computer to function as an acquisition unit that acquires, from a providing device, predicted results regarding the communication quality required for a service executed in the vehicle and auxiliary information used to make judgments about the predicted results.

[0040] With this configuration, the predicted communication quality can be determined using the auxiliary information, and whether or not to provide a service can be determined based on the determined result, thereby more reliably providing the service in the vehicle.

[0041] (15) A communication prediction program according to an embodiment of the present disclosure is a communication prediction program used in a providing device, and is a program for causing a computer to function as a prediction unit that predicts the communication quality required for a service executed in a vehicle, a creation unit that creates auxiliary information used to judge the prediction result of the prediction unit, and a transmission unit that transmits the prediction result of the prediction unit and the auxiliary information created by the creation unit.

[0042] With this configuration, the predicted communication quality can be determined using the auxiliary information, and whether or not to provide a service can be determined based on the determined result, thereby more reliably providing the service in the vehicle.

[0043] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the drawings, identical or corresponding parts are designated by the same reference numerals, and their description will not be repeated. Furthermore, at least some of the embodiments described below may be combined in any manner.

[0044] [Communication System] Fig. 1 is a diagram illustrating an example of the configuration of a communication system according to an embodiment of the present disclosure. Referring to Fig. 1, a communication system 501 includes one or more in-vehicle devices 101 and servers 201 and 301. Each in-vehicle device 101 and the servers 201 and 301 transmit and receive information via an external network 151 such as the Internet. The in-vehicle device 101 is mounted on a vehicle 1. The servers 201 and 301 are provided outside the vehicle 1. The server 301 is an example of a providing device.

[0045] The in-vehicle device 101 communicates with the wireless base station device 161 wirelessly according to a communication method Cm such as Wi-Fi (registered trademark), LTE (Long Term Evolution) (registered trademark), or 5G, thereby communicating with the servers 201, 301 via the external network 151.

[0046] For example, the in-vehicle device 101 executes various services S. The services S include a service for updating software used in the in-vehicle network 401, a service for detecting a fault in the vehicle 1, and the like.

[0047] More specifically, for example, the in-vehicle device 101 executes the service S by communicating with the server 201 in accordance with TCP (Transmission Control Protocol).

[0048] In addition, the in-vehicle device 101 is not limited to a configuration in which service S is executed by communicating with the server 201 in accordance with TCP, but may also be a configuration in which service S is executed by communicating with in-vehicle equipment installed in the vehicle 1.

[0049] The server 301 makes a prediction regarding the communication quality required for the service S. The server 301 is used, for example, by a business operator that manages the operation of the vehicle 1 or by an individual.

[0050] [In-Vehicle Device] Fig. 2 is a diagram showing an example of the configuration of an in-vehicle device according to an embodiment of the present disclosure. Referring to Fig. 2, the in-vehicle device 101 is connected to a plurality of in-vehicle devices 202 via, for example, an Ethernet (registered trademark) cable 51. The in-vehicle device 101 and the plurality of in-vehicle devices 202 constitute an in-vehicle network 401.

[0051] The in-vehicle devices 202 include an in-vehicle ECU (Electronic Control Unit), a sensor, a GPS (Global Positioning System) receiver, a human-machine interface, a camera, etc. The in-vehicle ECUs include a TCU (Telematics Communication Unit), an automatic driving ECU, an engine ECU, and a steering control ECU.

[0052] 2, Ethernet cables 51A, 51B, and 51C are provided as the Ethernet cable 51. Also, in the example shown in Fig. 2, in-vehicle devices 202A, 202B, and 202C are provided as the in-vehicle devices 202.

[0053] The in-vehicle device 101 is connected to in-vehicle devices 202A, 202B, and 202C via Ethernet cables 51A, 51B, and 51C, respectively.

[0054] The in-vehicle device 202 creates an Ethernet frame including various information described later, and transmits the created Ethernet frame to the in-vehicle device 101.

[0055] 2, the in-vehicle devices 202A, 202B, and 202C are a GPS receiver, a vehicle speed sensor, and a navigation device, respectively. In the following description, the in-vehicle devices 202A, 202B, and 202C are also referred to as the GPS receiver 202A, the vehicle speed sensor 202B, and the navigation device 202C, respectively.

[0056] The GPS receiver 202A receives GPS signals from one or more satellites and, based on the received GPS signals, detects the position of the vehicle 1. The position of the vehicle 1 is indicated by, for example, latitude and longitude.

[0057] The GPS receiver 202A transmits position information P including the detection result and the detection time tp to the in-vehicle device 101. The GPS receiver 202A performs position detection and transmits the position information P, for example, periodically.

[0058] The vehicle speed sensor 202B measures the speed of the vehicle 1, for example, periodically, and transmits vehicle speed information V including the measurement result and the measurement time tv to the in-vehicle device 101.

[0059] The navigation device 202C accepts input of the departure point, destination, and scheduled departure time ta of the vehicle 1 by the user of the vehicle 1. Upon accepting the input of the departure point, destination, and scheduled departure time ta, the navigation device 202C creates route information indicating a planned driving route from the departure point to the destination, the scheduled departure time ta, passing points between the departure point and the destination, scheduled passage times tb which are the scheduled times at which the vehicle will pass the passing points, and scheduled arrival time tc which is the scheduled time at which the vehicle will arrive at the destination.

[0060] For example, when the navigation device 202C creates route information, it displays a screen on its display unit prompting the driver to input whether or not they would like to receive service S when the vehicle 1 travels along the planned route indicated by the route information.

[0061] When the navigation device 202C receives an input indicating a request for the provision of service S, the navigation device 202C transmits the created route information to the in-vehicle device 101.

[0062] On the other hand, when the navigation device 202C receives an input indicating that the provision of the service S is not desired, the navigation device 202C does not transmit the created route information to the in-vehicle device 101.

[0063] The in-vehicle network 401 is not limited to a configuration in which four in-vehicle devices 202 are provided, but may be a configuration in which one to three or five or more in-vehicle devices 202 are provided.

[0064] Furthermore, the in-vehicle device 101 and the in-vehicle equipment 202 may be configured to perform communication in accordance with a communication protocol such as CAN (Controller Area Network), CAN FD (CAN with Flexible Data Rate), FlexRay (registered trademark), MOST (Media Oriented System Transport) (registered trademark), LIN (Local Interconnect Network), and CXPI (Clock Extension Peripheral Interface) (registered trademark), instead of or in addition to communication in accordance with the Ethernet standard.

[0065] The in-vehicle device 101 includes an in-vehicle communication unit 11, a service management unit 12, an external communication unit 13, a determination unit 14, a correction unit 15, and a storage unit 16. Some or all of the in-vehicle communication unit 11, the service management unit 12, the external communication unit 13, the determination unit 14, and the correction unit 15 are realized, for example, by a processing circuit including one or more processors. The storage unit 16 is, for example, a non-volatile memory included in the processing circuit. The external communication unit 13 is an example of an acquisition unit.

[0066] (In-vehicle communication unit) The in-vehicle communication unit 11 acquires various information from the in-vehicle device 202. More specifically, the in-vehicle communication unit 11 stores the received position information P in the storage unit 16 every time the in-vehicle communication unit 11 receives the position information P from the GPS receiver 202A.

[0067] In addition, the in-vehicle communication unit 11 stores the received vehicle speed information V in the storage unit 16 every time it receives the vehicle speed information V from the vehicle speed sensor 202B.

[0068] Furthermore, when the in-vehicle communication unit 11 receives route information from the navigation device 202C, it outputs the received route information to the service management unit 12 with a timestamp indicating the time tr at which the route information was received.

[0069] The in-vehicle device 101 is not limited to a configuration that acquires route information indicating a planned driving route of the vehicle 1 from the navigation device 202C, but may also be a configuration that predicts the planned driving route. In this case, for example, the in-vehicle device 101 predicts the planned driving route based on the latest vehicle speed and latest traveling direction of the vehicle 1. Specifically, for example, the in-vehicle device 101 predicts the planned driving route by assuming that the vehicle 1 moves at a constant speed in a straight line based on the vehicle speed and traveling direction.

[0070] (Service Management Unit) The service management unit 12 requests the server 301 to predict the communication quality required for the service S.

[0071] More specifically, for example, the service management unit 12 transmits request information to the server 301 via the external vehicle communication unit 13, the request information including vehicle information regarding the vehicle 1, communication information regarding communication between the vehicle 1 and the server 201, and prediction target information indicating the type of communication quality that the server 301 is requested to predict.

[0072] FIG. 3 is a diagram illustrating an example of vehicle information and communication information included in request information transmitted by the in-vehicle device according to the embodiment of the present disclosure.

[0073] 3 , the vehicle information includes, for example, identification information for identifying the vehicle 1 (hereinafter also referred to as a vehicle ID (Identifier)), vehicle model information indicating the vehicle model of the vehicle 1, modem information indicating the model number of a modem mounted in the vehicle 1, and antenna information indicating the model number and number of antennas mounted in the vehicle 1. The vehicle ID is an ID unique to the vehicle 1.

[0074] For example, the storage unit 16 stores a vehicle ID, vehicle model information, modem information, and antenna information.

[0075] When the service management unit 12 receives the route information from the in-vehicle communication unit 11 , it acquires the vehicle ID, vehicle model information, modem information, and antenna information from the storage unit 16 .

[0076] The vehicle information further includes the predicted start time ts of the communication quality requested from the server 301, the latest position of the vehicle 1, the geohash value corresponding to the position, the latest direction of travel of the vehicle 1, and the latest vehicle speed of the vehicle 1.

[0077] More specifically, when the service management unit 12 receives route information from the in-vehicle communication unit 11, it acquires, from among the multiple pieces of location information P stored in the memory unit 16, location information P (hereinafter also referred to as "location information P1") that includes the detection time tp closest to the predicted start time ts indicated by the timestamp attached to the route information, and location information P (hereinafter also referred to as "location information P2") that includes the detection time tp second closest to the predicted start time ts.

[0078] The geohash value is a value obtained by converting a pair of latitude and longitude indicating the location of vehicle 1 into a character string using a method called geohash.

[0079] When the service management unit 12 acquires the location information P1 from the memory unit 16, it calculates a geohash value corresponding to the location of the vehicle 1 indicated by the location information P1 by substituting the pair of latitude and longitude indicated by the location information P1 into a predetermined conversion formula.

[0080] In addition, the service management unit 12 calculates the latest traveling direction of the vehicle 1 using the position information P1 and position information P2 acquired from the storage unit 16.

[0081] In addition, when the service management unit 12 receives route information from the in-vehicle communication unit 11, it acquires vehicle speed information V (hereinafter also referred to as ``vehicle speed information V1'') from the multiple vehicle speed information V stored in the memory unit 16, which includes the measurement time tv that is closest to the predicted start time ts indicated by the timestamp attached to the route information.

[0082] Then, the service management unit 12 creates vehicle information including the predicted start time ts, the latest position of vehicle 1 indicated by the position information P1, the calculated geohash value and direction of travel, and the latest vehicle speed of vehicle 1 indicated by the vehicle speed information V1.

[0083] <Communication Information> The communication information indicates identification information for identifying the telecommunications carrier that manages the wireless base station device 161 (hereinafter also referred to as the "telecommunications carrier ID"), identification information for identifying the cell Ce formed by the wireless base station device 161 (hereinafter also referred to as the "cell ID"), and the bandwidth used in communication between the in-vehicle device 101 and the server 201.

[0084] More specifically, the used band indicates at least one of a center frequency and a bandwidth used in wireless communication between the in-vehicle device 101 and the wireless base station device 161. In the present embodiment, for example, the used band indicates the center frequency of the wireless communication according to the above-mentioned communication method Cm, i.e., a communication method such as Wi-Fi, LTE, or 5G.

[0085] When the service management unit 12 receives the route information from the in-vehicle communication unit 11, it outputs a request notification Re1 to the out-vehicle communication unit 13, requesting notification of the carrier ID, cell ID, and bandwidth used.

[0086] When the exterior-vehicle communication unit 13 receives the request notification Re1 from the service management unit 12, it notifies the service management unit 12 of the communication carrier ID, cell ID, and required bandwidth acquired from the wireless base station device 161 of the communication partner.

[0087] The communication information further indicates a received signal strength indicator (RSSI), a signal-to-interference-plus-noise ratio (SINR), a reference signal received power (RSRP), and a reference signal received quality (RSRQ).

[0088] When the service management unit 12 receives the route information from the in-vehicle communication unit 11, it outputs a request notification Re2 to the out-vehicle communication unit 13, requesting notification of RSSI, SINR, RSRP, and RSRQ.

[0089] The exterior-vehicle communication unit 13 measures the RSSI, SINR, RSRP, and RSRQ in wireless communication with the wireless base station device 161, for example, periodically or irregularly.

[0090] When the exterior communication unit 13 receives the request notification Re2 from the service management unit 12, it notifies the service management unit 12 of the latest measurement results of the RSSI, SINR, RSRP, and RSRQ.

[0091] The service management unit 12 generates communication information including the carrier ID, cell ID, required band, RSSI, SINR, RSRP, and RSRQ notified by the exterior communication unit 13 .

[0092] <Prediction Target Information> FIG. 4 is a diagram illustrating an example of prediction target information included in request information transmitted by the in-vehicle device according to the embodiment of the present disclosure.

[0093] Referring to Figure 4, for example, the prediction target information indicates the communication speed Vc between vehicle 1 and server 201, the communication delay time Td in the communication between vehicle 1 and server 201, and whether or not a handover H occurs during the communication between vehicle 1 and server 201, as the type of communication quality that server 301 is requested to predict.

[0094] For example, the storage unit 16 stores map information for a planned travel area A including the planned travel route indicated by the route information received from the in-vehicle communication unit 11 .

[0095] The service management unit 12 divides the planned travel area A into a plurality of grid areas Rg using the map information stored in the storage unit 16 and the route information received from the in-vehicle communication unit 11 .

[0096] For example, the prediction target information includes communication speed information for requesting prediction of the communication speed Vc, delay time information for requesting prediction of the communication delay time Td, and handover information for requesting prediction of whether or not a handover H will occur. Note that the prediction target information may include at least one of the communication speed information, delay time information, and handover information.

[0097] Each of the communication speed information and delay time information indicates the ID of the server 201, the communication direction D during execution of the service S, the time period Tp to be predicted, and the geohash value G, the bandwidth used F, and the vehicle speed for each grid area.

[0098] For example, the storage unit 16 stores the ID of the server 201 and the communication direction D. When the service management unit 12 divides the planned travel area A into a plurality of grid areas, it acquires the ID of the server 201 and the communication direction D.

[0099] In addition, when the service management unit 12 divides the planned travel area A into multiple grid areas Rg, it calculates a geohash value G for each grid area Rg by substituting the pair of latitude and longitude contained in that grid area Rg into a conversion formula.

[0100] FIG. 5 is a diagram for explaining geohash values ​​calculated by a comparative example of an in-vehicle device.

[0101] 5, the comparative example of the on-board device divides the planned travel area A into multiple grid areas Rg, each of which has a distance of approximately 152 m in the east-west direction and a distance of approximately 152 m in the north-south direction. In this case, the geohash value G calculated by the comparative example of the on-board device has seven digits.

[0102] FIG. 6 is a diagram illustrating geohash values ​​calculated by an in-vehicle device according to an embodiment of the present disclosure.

[0103] 2 and 6, in the in-vehicle device 101, the service management unit 12 divides the planned travel area A into multiple grid areas Rg whose east-west and north-south distances are approximately 38 m and 19 m, respectively. In this case, the geohash value G calculated by the service management unit 12 has eight digits. In the example shown in Fig. 10, each grid area Rg is assigned a code represented by numbers or letters.

[0104] After calculating the geohash value G, the service management unit 12 stores the number of digits of the geohash value G, i.e., eight digits, in the storage unit 16 .

[0105] In addition, the service management unit 12 calculates the time period Tp to be predicted as the time period from the prediction start time ts indicated by the route information received from the in-vehicle communication unit 11 to the prediction end time tf at which the server 301 ends the prediction regarding communication quality.

[0106] Then, the service management unit 12 creates communication speed information and delay time information indicating the ID of the server 201, the communication direction D, the bandwidth F and vehicle speed used for each grid area Rg, as well as the calculated geohash value G and time period Tp for each grid area Rg.

[0107] Note that the communication speed information and delay time information included in the prediction target information may indicate the bandwidth F, vehicle speed, and coordinates represented by latitude and longitude for each passing point of the vehicle 1, instead of the bandwidth F, vehicle speed, and geohash value G for each grid area Rg. In this case, the server 301 uses the prediction target information received from the on-board device 101 to divide the planned travel area A of the vehicle 1 into multiple grid areas Rg, and calculates the bandwidth F, vehicle speed, and geohash value G for each grid area Rg.

[0108] The handover information indicates the time period Tp to be predicted and the stay time Tc of the vehicle 1 in the cell Ce.

[0109] For example, the storage unit 16 stores the positions of each of a plurality of points included in the cell Ce. When the service management unit 12 receives route information from the in-vehicle communication unit 11, the service management unit 12 calculates the stay time Tc using the route information and the positions of each of the plurality of points.

[0110] Then, the service management unit 12 creates handover information using the calculated stay time and time period Tp.

[0111] After creating the vehicle information, communication information, and prediction target information, the service management unit 12 outputs request information including the created vehicle information, communication information, and prediction target information to the exterior communication unit 13.

[0112] When the exterior communication unit 13 receives the request information from the service management unit 12 , it transmits an IP packet including the request information (hereinafter also referred to as a “request packet”) to the server 301 .

[0113] Specifically, for example, the exterior-vehicle communication unit 13 creates a request packet that includes the request information received from the service management unit 12 and the vehicle ID stored in the storage unit 16, and that includes the IP address of the vehicle 1 and the IP address of the server 301 as the source IP address and the destination IP address, respectively. Then, the exterior-vehicle communication unit 13 transmits the created request packet to the server 301.

[0114] [Server] Fig. 7 is a diagram illustrating an example of a configuration of a server according to an embodiment of the present disclosure. Referring to Fig. 7, the server 301 includes a communication unit 21, a prediction unit 22, a creation unit 23, an update unit 24, and a storage unit 25. Some or all of the communication unit 21, the prediction unit 22, the creation unit 23, and the update unit 24 are realized, for example, by a processing circuit including one or more processors. The storage unit 25 is, for example, a non-volatile memory included in the processing circuit. The communication unit 21 is an example of a transmission unit.

[0115] When the communication unit 21 receives a request packet from the in-vehicle device 101 via the wireless base station device 161 and the external network 151 , the communication unit 21 outputs the request information included in the received request packet to the prediction unit 22 .

[0116] (Prediction Unit) The prediction unit 22 performs a prediction process to predict the communication quality required for the service S executed in the in-vehicle device 101 .

[0117] In the prediction process, the prediction unit 22 predicts the communication quality at each time included in the time period Tp to be predicted, which is included in the request information received from the communication unit 21 .

[0118] For example, the prediction unit 22 performs a prediction process using machine learning techniques (hereinafter also referred to as "prediction process R1") and a prediction process using past communication history in the vehicle 1 (hereinafter also referred to as "prediction process R2").

[0119] <Prediction Process R1> For example, the prediction unit 22 performs prediction process R1 using a trained model. In this case, the prediction result of the prediction unit 22 is the output value of the trained model.

[0120] For example, the memory unit 25 stores a trained model E1 for making predictions regarding the communication speed Vc, a trained model E2 for making predictions regarding the communication delay time Td, and a trained model E3 for making predictions regarding handover H.

[0121] More specifically, for example, the trained model E1 outputs a prediction result C11 of the communication speed Vc at regular intervals during the time period Tp to be predicted, from the prediction start time ts to the prediction end time tf, and a probability Q1 that the prediction result C11 will be less than a predetermined tolerance value K1.

[0122] For example, the trained model E2 outputs a prediction result C12 of the communication delay time Td at regular intervals during the time period Tp to be predicted, and a probability Q2 that the prediction result C12 will be greater than a predetermined tolerance value K2.

[0123] For example, the trained model E3 outputs a prediction result C13 regarding whether or not a handover H will occur at regular intervals during the time period Tp to be predicted, and a classification probability Q3 indicating the reliability of the prediction result C13.

[0124] When the prediction unit 22 receives request information from the communication unit 21, it acquires the trained models E1, E2, and E3 from the storage unit 25.

[0125] When the prediction unit 22 acquires the trained model E1 from the memory unit 25, it inputs to the trained model E1 the vehicle information, communication information, and communication speed information included in the request information received from the communication unit 21. Then, the prediction unit 22 acquires the prediction result C11 and the probability Q1 output from the trained model E1.

[0126] When the prediction unit 22 acquires the trained model E2 from the memory unit 25, it inputs to the trained model E2 the vehicle information, communication information, and delay time information included in the request information received from the communication unit 21. Then, the prediction unit 22 acquires the prediction result C12 and the probability Q2 output from the trained model E2.

[0127] When the prediction unit 22 acquires the trained model E3 from the memory unit 25, it inputs to the trained model E3 the vehicle information, communication information, and handover information included in the request information received from the communication unit 21. Then, the prediction unit 22 acquires the prediction result C13 and the classification probability Q3 output from the trained model E3.

[0128] <Prediction process R2> For example, the storage unit 25 stores communication history information indicating past communication history of the vehicle 1. The communication history information indicates a history value J1 of the communication speed Vc at a passing point and a history value J2 of the communication delay time Td at the passing point. The communication history information is updated by the update unit 24, which will be described later.

[0129] Specifically, for example, the communication history information indicates the correspondence between a geohash value G indicating a passing point, a time period, a history value J1 of a communication speed Vc, and a history value J2 of a communication delay time Td.

[0130] For example, the time periods indicated by the communication performance information are morning, mid-morning, noon, afternoon, evening, night, and late night. In the following description, the morning time period is from 5:00 AM to 7:59 AM, the morning time period is from 8:00 AM to 10:59 AM, the daytime period is from 11:00 AM to 12:59 PM, the afternoon time period is from 1:00 PM to 3:59 PM, the evening time period is from 4:00 PM to 5:59 PM, the night time period is from 6:00 PM to 11:59 PM, and the late night time period is from 12:00 AM to 4:59 AM.

[0131] When the prediction unit 22 receives request information from the communication unit 21, the prediction unit 22 refers to the communication performance information in the storage unit 25, and for each geohash value G indicated by the communication speed information included in the request information, identifies an performance value J1 corresponding to the geohash value G. In addition, for each geohash value G indicated by the delay time information included in the request information received from the communication unit 21, the prediction unit 22 identifies an performance value J2 corresponding to the geohash value G.

[0132] The prediction unit 22 then acquires the identified performance value J1 and performance value J2 as a prediction result C21 of the communication speed Vc and a prediction result C22 of the communication delay time Td, respectively.

[0133] Each time the prediction unit 22 obtains a prediction result C21, it stores a set W1 of the prediction result C21 and the geohash value G corresponding to the prediction result C21 in the storage unit 25. The prediction unit 22 also stores a set W2 of the prediction result C22 and the geohash value G corresponding to the prediction result C22 in the storage unit 25.

[0134] Upon completing the prediction processes R1 and R2, the prediction unit 22 stores the vehicle ID included in the request information received from the communication unit 21, as well as the prediction result information including the acquired prediction results C11, C12, C13, C21, and C22, the probabilities Q1 and Q2, and the classification probability Q3, in the storage unit 25. Then, the prediction unit 22 outputs the request information received from the communication unit 21 to the creation unit 23.

[0135] Note that the server 301 is not limited to a configuration that performs the prediction process R1 using the trained models E1, E2, and E3, but may also be configured to perform the prediction process R1 using communication history information in addition to the trained models E1, E2, and E3. In this case, the server 301 updates the trained models E1, E2, and E3 in the storage unit 25, for example, periodically, using the latest communication history information stored in the storage unit 25. Then, the server 301 performs the prediction process R1 using the updated trained models E1, E2, and E3.

[0136] (Auxiliary Information) FIG. 8 is a diagram illustrating an example of auxiliary information created by the server according to the embodiment of the present disclosure.

[0137] 7 and 8, the creation unit 23 creates auxiliary information to be used for determining the prediction result of the prediction unit 22. For example, the creation unit 23 creates auxiliary information for each time included in the time slot Tp to be predicted.

[0138] For example, the auxiliary information includes determination information that allows determination of the reliability of the prediction result of the prediction section 22 and correction information that is used to correct the prediction result of the prediction section 22 .

[0139] <Judgment information> More specifically, for example, the judgment information includes judgment information M1 that can judge the reliability of the prediction result of the communication speed Vc, judgment information M2 that can judge the reliability of the prediction result of the communication delay time Td, and judgment information M3 that can judge the reliability of the prediction result regarding whether or not a handover H will occur.

[0140] (a1) Size of the prediction target area for communication quality: For example, the determination information M1 includes the size La of the prediction target area for the communication speed Vc. Specifically, the determination information M1 includes the number of digits of the geohash value G corresponding to the prediction result C21 for the communication speed Vc as the size La of the prediction target area for the communication speed Vc.

[0141] For example, when the creation unit 23 receives request information from the prediction unit 22, it checks whether a set W1 corresponding to the geohash value G (hereinafter also referred to as "geohash value G1") with the same number of digits as the geohash value G indicated by the communication speed information included in the request information is stored in the memory unit 25.

[0142] Then, if a set W1 corresponding to a geohash value G with the same number of digits as the geohash value G1 is stored in the memory unit 25, the creation unit 23 includes the number of digits of the geohash value G as size La in the judgment information M1.

[0143] On the other hand, if a set W1 corresponding to a geohash value G with the same number of digits as the geohash value G1 is not stored in the memory unit 25, the creation unit 23 checks whether a set W1 corresponding to a geohash value G that is one digit smaller than the number of digits of the geohash value G1 is stored in the memory unit 25.

[0144] Then, if a set W1 corresponding to a geohash value G that is one digit smaller than the number of digits of the geohash value G1 is stored in the memory unit 25, the creation unit 23 includes the number of digits of the geohash value G as size La in the judgment information M1.

[0145] On the other hand, if a set W1 corresponding to a geohash value G that is one digit smaller than the number of digits of the geohash value G1 is not stored in the memory unit 25, the creation unit 23 includes, for example, a value obtained by subtracting 2 from the number of digits of the geohash value G1 as size La in the judgment information M1.

[0146] For example, the determination information M2 includes the size Lb of the prediction target area for the communication delay time Td. Specifically, the determination information M2 includes the number of digits of the geohash value G corresponding to the prediction result C22 for the communication delay time Td as the size Lb of the prediction target area for the communication delay time Td.

[0147] For example, when the creation unit 23 receives request information from the prediction unit 22, it checks whether a set W2 corresponding to the geohash value G (hereinafter also referred to as "geohash value G2") having the same number of digits as the geohash value G indicated by the communication speed information included in the request information is stored in the memory unit 25.

[0148] Then, if a set W2 corresponding to a geohash value G with the same number of digits as the geohash value G2 is stored in the memory unit 25, the creation unit 23 includes the number of digits of the geohash value G as size Lb in the judgment information.

[0149] On the other hand, if a set W2 corresponding to a geohash value G with the same number of digits as the geohash value G2 is not stored in the memory unit 25, the creation unit 23 checks whether a set W2 corresponding to a geohash value G that is one digit smaller than the number of digits of the geohash value G2 is stored in the memory unit 25.

[0150] Then, if a set W2 corresponding to a geohash value G that is one digit smaller than the number of digits of the geohash value G2 is stored in the memory unit 25, the creation unit 23 includes the number of digits of the geohash value G as size Lb in the judgment information.

[0151] On the other hand, if a set W2 corresponding to a geohash value G that is one digit smaller than the number of digits of the geohash value G2 is not stored in the memory unit 25, the creation unit 23 includes, for example, a value obtained by subtracting 2 from the number of digits of the geohash value G2 as size Lb in the judgment information M2.

[0152] (a2) Number of Data Used to Predict Communication Quality For example, the determination information M1 includes the number of data used to predict the communication speed Vc, specifically the number of actual values ​​J1.

[0153] For example, when the creation unit 23 receives request information from the prediction unit 22, the creation unit 23 refers to the communication performance information in the storage unit 25 to confirm, for each geohash value G1 indicated by the communication speed information included in the request information, the number of performance values ​​J1 corresponding to the same geohash value as the geohash value G1. The creation unit 23 then includes the confirmed number of performance values ​​J1 in the determination information M1.

[0154] For example, the determination information M2 includes the number of data used to predict the communication delay time Vd, specifically the number of actual results J2.

[0155] For example, the creation unit 23 refers to the communication performance information in the storage unit 25 and checks, for each geohash value G2 indicated by the delay time information included in the request information received from the prediction unit 22, the number of performance values ​​J2 corresponding to the same geohash value G2. Then, the creation unit 23 includes the checked number of performance values ​​J2 in the determination information M2.

[0156] (a3) Proportion of data in the time period Tp to be predicted For example, the judgment information M1 includes the proportion of the number of actual values ​​J1 in the time period Tp to the number of actual values ​​J1 used to predict the communication speed Vc in other time periods different from the time period Tp to be predicted.

[0157] For example, the creation unit 23 obtains the number of actual values ​​J1 in each time period for each geohash value G1 indicated by the communication speed information included in the request information received from the prediction unit 22 by referring to the communication performance information in the memory unit 25.

[0158] Then, for each geohash value G1, the creation unit 23 selects, from the number of actual values ​​J1 in each time zone, the number N11 of actual values ​​J1 in the time zone Tp to be predicted and the maximum number N12 of actual values ​​J1 in other time zones other than the time zone Tp.

[0159] After selecting the number N11 and the maximum value N12, the creating unit 23 calculates the ratio Ya of the number N11 to the maximum value N12, and then includes the calculated ratio Ya in the determination information M1.

[0160] For example, the determination information M2 includes the ratio of the number of performance values ​​J2 in the time period Tp to the number of performance values ​​J2 used to predict the communication delay time Td in other time periods different from the time period Tp to be predicted.

[0161] For example, the creation unit 23 obtains the number of actual values ​​J2 in each time period for each geohash value G2 indicated by the delay time information included in the request information received from the prediction unit 22 by referring to the communication performance information in the memory unit 25.

[0162] Then, for each geohash value G2, the creation unit 23 selects, from the number of actual values ​​J2 in each time zone, the number N21 of actual values ​​J1 in the time zone Tp to be predicted and the maximum number N22 of actual values ​​J1 in other time zones other than the time zone Tp.

[0163] After selecting the number N21 and the maximum value N22, the creating unit 23 calculates the ratio Yb of the number N21 to the maximum value N22, and then includes the calculated ratio Yb in the determination information M2.

[0164] (a4) Judgment result on past prediction result For example, the judgment information M1 includes a judgment result on a past prediction result C11 for the communication speed Vc. More specifically, the judgment information M1 includes an error rate U1 of a past prediction result C21 in the in-vehicle device 101 as the judgment result.

[0165] For example, the storage unit 25 stores an error rate U1. When the creation unit 23 receives request information from the prediction unit 22, it acquires the error rate U1 from the storage unit 25. Then, the creation unit 23 includes the acquired error rate U1 in the determination information M1.

[0166] For example, the determination information M2 includes a determination result for the past prediction result C12 for the communication delay time Td. More specifically, the determination information M2 includes an error rate U2 of the past prediction result C12 in the in-vehicle device 101 as the determination result.

[0167] For example, the storage unit 25 stores an error rate U2. When the creation unit 23 receives request information from the prediction unit 22, it acquires the error rate U2 from the storage unit 25. Then, the creation unit 23 includes the acquired error rate U2 in the determination information M2.

[0168] For example, the determination information M3 includes a determination result for a past prediction result C13 regarding whether or not a handover H will occur. More specifically, the determination information M3 includes, as the determination result, an accuracy rate U3 of the past prediction result C13 in the in-vehicle device 101.

[0169] For example, the storage unit 25 stores the accuracy rate U3. When the creation unit 23 receives request information from the prediction unit 22, it acquires the accuracy rate U3 from the storage unit 25. Then, the creation unit 23 includes the acquired accuracy rate U3 in the determination information M3.

[0170] <Correction Information> Fig. 9 is a diagram for explaining correction information created by the server according to the embodiment of the present disclosure. Fig. 9 shows the change over time in the communication speed Vc between the vehicle 1 and the server 201. In Fig. 9, the horizontal axis represents time t, and the vertical axis represents the communication speed Vc.

[0171] 9, for example, when communication conforming to TCP is performed between the in-vehicle device 101 and the server 201, the communication speed Vc at the start of execution of the service S is limited.

[0172] More specifically, for example, slow start control is performed between the in-vehicle device 101 and the server 201 in accordance with TCP. Specifically, when a communication connection is established between the in-vehicle device 101 and the server 201, the in-vehicle device 101 and the server 201 set the window size to a minimum value and transmit and receive packets containing various information in accordance with the window size. Then, the in-vehicle device 101 and the server 201 transmit and receive packets by gradually increasing the window size as time passes.

[0173] 9, the change in communication speed Vc over time when slow start control is performed between the in-vehicle device 101 and the server 201 is expressed, for example, by an exponential function. For example, the change in time is expressed by the following equation (1): In equation (1), Va is the communication speed at the start of execution of service S, and B is a coefficient.

[0174]

[0175] In the server 301 shown in FIG. 7, the creation unit 23 creates correction information used to correct the predicted result of the communication speed Vc.

[0176] More specifically, for example, the creation unit 23 creates correction information including parameters p1 and p2 relating to the limit on the communication speed Vc.

[0177] For example, the storage unit 25 stores the value Va and the value B in the formula (1). When the creation unit 23 receives the request information from the prediction unit 22, the creation unit 23 creates correction information that includes the value Va and the value B stored in the storage unit 25 as parameters p1 and p2, respectively.

[0178] The creation unit 23 then outputs to the prediction unit 22 auxiliary information including the vehicle ID included in the request information received from the prediction unit 22, the created correction information, and the determination information M1, M2, and M3.

[0179] When the prediction unit 22 receives the auxiliary information from the creation unit 23, the prediction unit 22 acquires prediction result information including the same vehicle ID as the vehicle ID indicated in the auxiliary information from the storage unit 25. Then, the prediction unit 22 outputs the acquired prediction result information, including the auxiliary information received from the creation unit 23, to the communication unit 21.

[0180] (Communication Unit) The communication unit 21 transmits the prediction result of the prediction unit 22 and the auxiliary information created by the creation unit 23 .

[0181] More specifically, for example, when the communication unit 21 receives prediction result information from the prediction unit 22 , it transmits a packet including the prediction result information (hereinafter also referred to as a “prediction packet”) to the in-vehicle device 101 .

[0182] Specifically, for example, the communication unit 21 creates a prediction packet that includes the prediction result information received from the prediction unit 22, and that includes the IP address of its own server 301 and the IP address of the vehicle 1 as the source IP address and the destination IP address, respectively. Then, the communication unit 21 transmits the created prediction packet to the in-vehicle device 101 via the external network 151.

[0183] [In-Vehicle Device] Referring again to FIG. 2 , in the in-vehicle device 101, for example, the external communication unit 13 acquires a plurality of prediction results corresponding to a plurality of times included in the time period Tp to be predicted, and auxiliary information for each time.

[0184] More specifically, when the exterior communication unit 13 receives a prediction packet from the server 301 , the exterior communication unit 13 outputs the prediction result information included in the received prediction packet to the determination unit 14 .

[0185] (Determination Unit) The determination unit 14 uses the correction information acquired from the server 301 to make a determination on the predicted result of the communication quality.

[0186] More specifically, for example, when the judgment unit 14 receives prediction result information from the exterior communication unit 13, it uses auxiliary information included in the prediction result information to make a judgment on the prediction result of communication quality included in the prediction result information.

[0187] For example, when the determination unit 14 receives prediction result information from the exterior communication unit 13, it checks whether or not correction information is included in the auxiliary information.

[0188] If the correction information is included in the auxiliary information, the determination unit 14 outputs the prediction result information received from the exterior communication unit 13 to the correction unit 15 .

[0189] On the other hand, if the correction information is not included in the auxiliary information, the determination unit 14 makes a determination on the predicted result of the communication quality using the determination information M1, M2, and M3 included in the auxiliary information.

[0190] (b1) Determination of the predicted result of the communication speed Vc For example, the determination unit 14 makes a determination of the predicted result C21 of the communication speed Vc using the determination information M1 included in the auxiliary information.

[0191] Specifically, for example, the judgment unit 14 makes a judgment on the predicted result C21 of the communication speed Vc using at least one of the number of digits of the geohash value G, the number of actual values ​​J1, the ratio Ya, and the error rate U1 contained in the judgment information M1.

[0192] For example, for each grid region Rg, the determination unit 14 compares the number of digits of the geohash value G included in the determination information M1 with the number of digits of the geohash value G stored in the memory unit 16 by the service management unit 12. Then, if the number of digits of all of the multiple geohash values ​​G included in the determination information M1 is equal to or greater than the number of digits of the geohash value G stored in the memory unit 16, the determination unit 14 makes a positive determination about the prediction result C21.

[0193] On the other hand, if the number of digits of at least one of the multiple geohash values ​​G included in the judgment information M1 is less than the number of digits of the geohash value G stored in the memory unit 16, the judgment unit 14 makes a negative judgment about the prediction result C21.

[0194] Furthermore, for example, the determination unit 14 compares the number of performance values ​​J1 included in the determination information M1 with a predetermined threshold value Th11 for each grid region Rg.

[0195] For example, when the vehicle 1 travels through the grid areas Rg of codes "2", "8", "9", "d", "f" and "g" in the order shown in FIG. 6, the judgment unit 14 checks whether the number of actual values ​​J1 in each grid area Rg is greater than or equal to the threshold value Th11.

[0196] Then, when the number of actual results values ​​J1 in each grid region Rg is equal to or greater than the threshold value Th11, the determination unit 14 makes a positive determination regarding the prediction result C21.

[0197] On the other hand, when the number of actual values ​​J1 in at least one grid region Rg among the plurality of grid regions Rg is less than the threshold value Th11, the determination unit 14 makes a negative determination regarding the prediction result C21.

[0198] Furthermore, for example, the determination unit 14 compares the ratio Ya included in the determination information M1 with a predetermined threshold Th12 for each grid region Rg. Then, when the ratio Ya is equal to or greater than the threshold Th12 for each grid region Rg, the determination unit 14 makes a positive determination regarding the prediction result C21.

[0199] On the other hand, when the proportion Ya in at least one grid region Rg among the plurality of grid regions Rg is less than the threshold value Th12, the determination unit 14 makes a negative determination regarding the prediction result C21.

[0200] Furthermore, for example, the determination unit 14 compares the error rate U1 included in the determination information M1 with a predetermined threshold value Th13. Then, if the error rate U1 is equal to or greater than the threshold value Th13, the determination unit 14 makes a positive determination regarding the prediction result C21.

[0201] On the other hand, when the error rate U1 included in the determination information M1 is less than the threshold value Th13, the determination unit 14 makes a negative determination regarding the prediction result C21.

[0202] (b2) Determination of the Prediction Result of the Communication Delay Time Td For example, the determination unit 14 determines the reliability of the prediction result C22 of the communication delay time Td using the determination information M2 included in the auxiliary information.

[0203] Specifically, for example, the judgment unit 14 makes a judgment on the predicted result C22 of the communication delay time Td using at least one of the number of digits of the geohash value G, the number of actual values ​​J2, the ratio Yb, and the error rate U2 contained in the judgment information M2.

[0204] For example, for each grid region Rg, the determination unit 14 compares the number of digits of the geohash value G included in the determination information M2 with the number of digits of the geohash value G stored in the memory unit 16 by the service management unit 12. Then, if the number of digits of all of the multiple geohash values ​​G included in the determination information M2 is equal to or greater than the number of digits of the geohash value G stored in the memory unit 16, the determination unit 14 makes a positive determination about the prediction result C22.

[0205] On the other hand, if the number of digits of at least one of the multiple geohash values ​​G included in the judgment information M2 is less than the number of digits of the geohash value G stored in the memory unit 16, the judgment unit 14 makes a negative judgment about the prediction result C22.

[0206] Furthermore, for example, the determination unit 14 compares the number of actual values ​​J2 included in the determination information M2 with a predetermined threshold Th21 for each grid region Rg. Then, when the number of actual values ​​J2 in each grid region Rg is equal to or greater than the threshold Th21, the determination unit 14 makes a positive determination regarding the prediction result C22.

[0207] On the other hand, when the number of actual results values ​​J2 in at least one grid region Rg among the plurality of grid regions Rg is less than the threshold value Th21, the determination unit 14 makes a negative determination regarding the prediction result C22.

[0208] Furthermore, for example, the determination unit 14 compares the ratio Yb included in the determination information M2 with a predetermined threshold Th22 for each grid region Rg. Then, if the ratio Yb for each grid region Rg is equal to or greater than the threshold Th22, the determination unit 14 makes a positive determination regarding the prediction result C22.

[0209] On the other hand, when the proportion Yb in at least one grid region Rg among the plurality of grid regions Rg is less than the threshold value Th22, the determination unit 14 makes a negative determination regarding the prediction result C22.

[0210] Furthermore, for example, the determination unit 14 compares the error rate U2 included in the determination information M2 with a predetermined threshold Th23. If the error rate U2 is equal to or greater than the threshold Th23, the determination unit 14 makes a positive determination regarding the prediction result C22.

[0211] On the other hand, when the error rate U2 included in the determination information M2 is less than the threshold value Th23, the determination unit 14 makes a negative determination regarding the prediction result C22.

[0212] (b3) Determination of the Prediction Result of Whether or Not Handover H Will Occur For example, the determination unit 14 makes a determination of the prediction result C13 of whether or not handover H will occur, using determination information M3 included in the auxiliary information.

[0213] More specifically, for example, the determination unit 14 determines the reliability of the prediction result C13 using the accuracy rate U3 included in the determination information M3.

[0214] Specifically, the determination unit 14 compares the accuracy rate U3 included in the determination information M3 with a predetermined threshold value Th31. If the accuracy rate U3 is equal to or greater than the threshold value Th31, the determination unit 14 makes a positive determination regarding the prediction result C13.

[0215] On the other hand, if the accuracy rate U3 is less than the threshold value Th31, the determination unit 14 makes a negative determination regarding the prediction result C13.

[0216] (Correction Unit) For example, the correction unit 15 performs a correction process to correct the predicted result C21 of the communication speed Vc using parameters p1 and p2 included in the correction information.

[0217] When the correction unit 15 receives the prediction result information from the determination unit 14, the correction unit 15 corrects the prediction result C21 of the communication speed Vc included in the prediction result information using the correction information included in the prediction result information.

[0218] 10 is a diagram illustrating an example of a communication speed prediction result and correction information included in prediction result information received from a server by an in-vehicle device according to an embodiment of the present disclosure, in which the horizontal axis represents time and the vertical axis represents the communication speed Vc.

[0219] 10, the dashed line graph Ga indicates the change over time of the prediction result C21 of the communication speed Vc included in the prediction result information received from the server 301. The solid line graph Gb indicates the change over time of the communication speed Vc represented by the parameters p1 and p2 included in the correction information.

[0220] For example, the correction unit 15 compares the prediction result C21 shown in the graph Ga with the communication speed Vc shown in the graph Gb at each time included in the time slot Tp to be predicted.

[0221] If the communication speed Vc indicated by the graph Gb is smaller than the predicted result C21, the correction unit 15 corrects the predicted result C21 to the value of the communication speed Vc.

[0222] On the other hand, if the communication speed Vc indicated by the graph Gb is equal to or greater than the prediction result C21, the correction unit 15 does not correct the prediction result C21.

[0223] In the example shown in Figure 10, in the time period from the prediction start time ts to time t11, the communication speed Vc shown in graph Gb is smaller than the prediction result C21, and in the time period from time t11 to the prediction end time tf, the value of the communication speed Vc shown in graph Gb is greater than or equal to the prediction result C21.

[0224] 11 is a diagram illustrating an example of a predicted communication speed after correction processing by the in-vehicle device according to the embodiment of the present disclosure, in which the horizontal axis represents time and the vertical axis represents the communication speed Vc.

[0225] 11, in the prediction result C21 after the correction process by the correction unit 15, the time change of the corrected prediction result C21 of the communication speed Vc in the time period from the prediction start time ts to the time t11 is similar to the time change of the graph Gb shown in Fig. 10. Furthermore, in the time period from the time t11 to the prediction end time tf, the time change of the corrected prediction result C21 of the communication speed Vc is similar to the time change of the graph Ga shown in Fig. 10.

[0226] Upon completing the correction process, the correction unit 15 outputs correction completion information indicating the corrected prediction result C21 to the determination unit 14.

[0227] When the determination unit 14 receives the correction completion information from the correction unit 15, the determination unit 14 makes a determination on the prediction results C22 and C13, and also makes a determination on the corrected prediction result C21 indicated by the correction completion information.

[0228] If the judgment unit 14 makes a positive judgment for all of the prediction results C21, C22, and C13, it outputs judgment positive information to the service management unit 12 indicating that the judgment result for the communication quality prediction result is positive.

[0229] On the other hand, if the judgment unit 14 makes a negative judgment about at least one of the prediction results C21, C22, and C13, it discards the prediction result information received from the exterior communication unit 13 and outputs negative judgment information to the service management unit 12 indicating that the judgment result about the prediction result of the communication quality is negative.

[0230] In addition, when the judgment unit 14 makes a positive judgment about some of the prediction results C21, C22, and C13, it may output the prediction result information received from the exterior communication unit 13 to the correction unit 15, and may also output positive judgment information indicating that the judgment result about the prediction result of the communication quality is positive to the service management unit 12.

[0231] 12 is a diagram illustrating another example of a predicted communication speed after correction processing by the in-vehicle device according to the embodiment of the present disclosure. In FIG. 12, the horizontal axis represents time, the vertical axis on the left represents the communication speed Vc, and the vertical axis on the right represents the bit rate B of communication between the in-vehicle device 101 and the server 201 when the service S is being executed.

[0232] 12, a solid line graph Gc indicates the change over time of the prediction result C21 of the communication speed Vc included in the prediction result information received from the server 301. A dashed line graph Gd indicates the change over time of the bit rate B.

[0233] 12, in the time variation of the prediction result C21 shown in graph Gc, the communication speed Vc in the time period from time t31 to time t32 gradually decreases as time passes. In this case, the service management unit 12 sets the bit rate B in the time period from time t31 to time t32 to a value equal to or less than a predetermined value Na. In the example shown in FIG. 12, the predetermined value Na is greater than 0 Mbps and less than or equal to 5 Mbps.

[0234] 12, if the communication speed Vc in the time period from time t31 to time t32 is zero, the correction unit 15 determines that the vehicle 1 is traveling outside the communication range in the time period from time t31 to time t32. In this case, the correction unit 15 outputs a change notification requesting a change of the communication method Cm, the communication carrier, etc. to the service management unit 12 via the determination unit 14.

[0235] When the service management unit 12 receives a change notification from the correction unit 15 via the determination unit 14, it changes the communication method Cm, the communication carrier, and the like.

[0236] (Service Management Unit) Referring back to FIG. 2, when the service management unit 12 receives negative determination information from the determination unit 14, it determines that the service S should not be executed.

[0237] On the other hand, when the service management unit 12 receives the positive judgment information from the judgment unit 14, it performs a service judgment process to judge whether or not to execute the service S. For example, the service management unit 12 performs the service judgment process using the prediction results C21, C22, and C13 included in the prediction result information received from the exterior-vehicle communication unit 13. For example, when the service management unit 12 judges from the prediction results C21, C22, and C13 that a deterioration in communication quality will not occur on the planned driving route of the vehicle 1, it determines to execute the service S.

[0238] Then, if the service management unit 12 determines to execute the service S, it starts executing the service S when the scheduled departure time ta indicated in the route information received from the in-vehicle communication unit 11 arrives.

[0239] On the other hand, if the service management unit 12 determines from the prediction results C21, C22, and C13 that a deterioration in communication quality will occur on the planned driving route of the vehicle 1, it does not execute service S and discards the prediction result information received from the external vehicle communication unit 13.

[0240] While the service S is being executed, the service management unit 12 periodically or irregularly generates communication result information indicating an actual value J1 of the communication speed Vc, an actual value J2 of the communication delay time Td, and whether or not a handover H has occurred. Then, the service management unit 12 includes the vehicle ID in the generated communication result information and transmits it to the server 301.

[0241] (Update Unit) Referring back to FIG. 7, in the server 301, when the communication unit 21 receives the communication result information from the in-vehicle device 101, the communication unit 21 outputs the received communication result information to the update unit 24.

[0242] When the update unit 24 receives the communication result information from the communication unit 21, it updates the communication performance information stored in the storage unit 25 using the communication result information.

[0243] In addition, when the update unit 24 receives communication result information from the communication unit 21, it updates the error rates U1, U2 and accuracy rate U3 stored in the memory unit 25 using the communication result information, as well as the predicted value of the actual value J1, the predicted value of the actual value J2 and the prediction result C13 regarding whether or not handover H will occur, which are stored in the memory unit 25.

[0244] [Operation Flow] Next, the operation flow of each device in the communication system 501 according to the embodiment of the present disclosure will be described with reference to the drawings.

[0245] 13 and 14 are flowcharts defining an example of an operation procedure when the in-vehicle device according to the embodiment of the present disclosure makes a determination on the predicted result of communication quality.

[0246] 13 and 14, first, the in-vehicle device 101 waits for reception of route information from the navigation device 202C (NO in step ST101).

[0247] Then, when the in-vehicle device 101 receives the route information from the navigation device 202C (YES in step ST101), it creates vehicle information. For example, as described above, the in-vehicle device 101 creates vehicle information indicating the prediction start time ts, the latest position of the vehicle 1, the geohash value corresponding to the position, the latest traveling direction of the vehicle 1, and the latest vehicle speed of the vehicle 1 (step ST102).

[0248] Next, the in-vehicle device 101 generates communication information indicating the carrier ID, the cell ID, the bandwidth used in communication between the vehicle 1 and the server 201, and the measured RSSI, SINR, RSRP, and RSRQ (step ST103).

[0249] Next, the in-vehicle device 101 generates prediction target information indicating the type of communication quality to be predicted by the server 301 (step ST104). Note that steps ST102, ST103, and ST104 may be executed in reverse order or in parallel.

[0250] Next, the in-vehicle device 101 transmits the created request information including the vehicle information, communication information, and prediction target information to the server 301 (step ST105).

[0251] Next, the in-vehicle device 101 waits for reception of prediction result information from the server 301 (NO in step ST106).

[0252] Then, when the in-vehicle device 101 receives the prediction result information from the server 301 (YES in step ST106), the in-vehicle device 101 checks whether or not the received prediction result information includes correction information (step ST107).

[0253] Next, if the received prediction result information includes correction information (YES in step ST107), the in-vehicle device 101 performs a correction process using the correction information to correct the prediction result C21 of the communication speed Vc included in the prediction result information (step ST108).

[0254] Next, the in-vehicle device 101 makes a judgment on the corrected prediction result C21 using the judgment information M1 included in the received prediction result information (step ST109).

[0255] Next, if the judgment result for the prediction result C21 is positive (YES in step ST110), the in-vehicle device 101 uses the judgment information M2 contained in the received prediction result information to make a judgment on the prediction result C22 of the communication delay time Td contained in the prediction result information (step ST111).

[0256] Next, if the judgment result regarding the prediction result C22 is positive (YES in step ST112), the in-vehicle device 101 uses the judgment information M3 contained in the received prediction result information to make a judgment regarding the prediction result C13 regarding whether or not handover H will occur, which is contained in the prediction result information (step ST113).

[0257] Next, if the judgment result for the prediction result C13 is positive (YES in step ST114), the in-vehicle device 101 performs a service judgment process using the prediction results C21, C22, and C13 to determine whether or not to execute service S (step ST115).

[0258] When the in-vehicle device 101 determines to execute the service S (YES in step ST116), the in-vehicle device 101 executes the service S by communicating with the server 201 (step ST117).

[0259] Next, the in-vehicle device 101 transmits to the server 301 communication result information indicating the actual value J1 of the communication speed Vc, the actual value J2 of the communication delay time Td, and whether or not a handover H occurred during the execution of the service S (step ST118).

[0260] On the other hand, if the received prediction result information does not include correction information (NO in step ST107), the in-vehicle device 101 uses the judgment information M1 included in the received prediction result information to make a judgment on the prediction result C21 of the communication speed Vc (step ST109).

[0261] In addition, if the judgment result regarding the prediction result C21, the judgment result regarding the prediction result C22, or the judgment result regarding the prediction result C13 is negative (NO in step ST110, NO in step ST112, or NO in step ST114), the in-vehicle device 101 determines not to execute service S (step ST119).

[0262] Next, the in-vehicle device 101 discards the prediction result information received from the server 301 (step ST120), and waits for the reception of new route information (NO in step ST101).

[0263] Furthermore, if the in-vehicle device 101 determines not to execute service S (NO in step ST115), it discards the prediction result information received from the server 301 (step ST120) and waits to receive new route information (NO in step ST101).

[0264] FIG. 15 is a flowchart illustrating an example of an operation procedure when a server according to an embodiment of the present disclosure performs a prediction process.

[0265] Referring to FIG. 15, first, server 301 waits for reception of request information from in-vehicle device 101 (NO in step ST201).

[0266] Then, when the server 301 receives the request information from the in-vehicle device 101 (YES in step ST201), it performs a prediction process R1 to predict the communication quality required for the service S using the trained models E1, E2, and E3 stored in the memory unit 25. For example, as described above, the server 301 uses the trained models E1, E2, and E3 to predict the communication speed Vc, the communication delay time Td, and whether or not a handover H will occur (step ST202), respectively.

[0267] Next, the server 301 performs a prediction process R2 to predict the communication quality required for the service S, using the communication performance information stored in the storage unit 25. For example, as described above, the server 301 predicts the communication speed Vc and the communication delay time Td using the communication performance information (step ST203). Note that the order of steps ST202 and ST203 may be reversed, or they may be performed in parallel.

[0268] Next, server 301 creates judgment information M1, judgment information M2, and judgment information M3 that can respectively determine the reliability of the predicted result of communication speed Vc, the reliability of the predicted result of communication delay time Td, and the reliability of the predicted result regarding whether or not handover H will occur (step ST204).

[0269] Next, the server 301 creates correction information used to correct the predicted result of the communication speed Vc (step ST205).

[0270] Next, the server 301 transmits prediction result information indicating the results of the prediction processes R1 and R2, as well as the auxiliary information including the judgment information M1, M2, and M3 and correction information that it has created, to the in-vehicle device 101 (step ST206), and waits to receive new request information (NO in step ST201).

[0271] FIG. 16 is a diagram illustrating an example of a sequence of processes performed by the navigation device, the in-vehicle device, and the server in the communication system according to the embodiment of the present disclosure.

[0272] Referring to FIG. 16, navigation device 202C transmits route information to in-vehicle device 101 (step ST301).

[0273] Next, when the in-vehicle device 101 receives the route information from the navigation device 202C, it creates vehicle information, communication information, and prediction target information (steps ST302 to ST304).

[0274] Next, the in-vehicle device 101 transmits the created request information including the vehicle information, communication information, and prediction target information to the server 301 (step ST305).

[0275] Next, when the server 301 receives request information from the in-vehicle device 101, it performs a prediction process R1 to predict the communication quality required for service S using the learned models E1, E2, and E3 stored in the memory unit 25 (step ST306).

[0276] Next, the server 301 performs a prediction process R2 for predicting the communication quality required for the service S using the communication performance information stored in the storage unit 25 (step ST307). Note that the order of steps ST306 and ST307 may be reversed or they may be performed in parallel.

[0277] Next, the server 301 creates judgment information M1, judgment information M2, and judgment information M3 that can respectively determine the reliability of the prediction result C21 of the communication speed Vc, the reliability of the prediction result C22 of the communication delay time Td, and the reliability of the prediction result C13 regarding whether or not handover H will occur (step ST308).

[0278] Next, the server 301 creates correction information used to correct the predicted result of the communication speed Vc (step ST309).

[0279] Next, the server 301 transmits prediction result information indicating the results of the prediction processes R1 and R2, as well as the created auxiliary information including the judgment information M1, M2, and M3 and correction information to the in-vehicle device 101 (step ST310).

[0280] Next, when the in-vehicle device 101 receives the prediction result information from the server 301, it performs a correction process to correct the prediction result C21 of the communication speed Vc using the correction information included in the received prediction result information (step ST311).

[0281] Next, the on-board device 101 makes a judgment on the prediction results of the communication quality required for the service S. Here, the on-board device 101 makes a judgment on the corrected prediction result C21, the prediction result C22 of the communication delay time Td, and the prediction result C13 regarding whether or not handover H will occur, and it is assumed that the on-board device 101 makes a positive judgment on all of the corrected prediction result C21, the prediction result C22, and the prediction result C13 (step ST312).

[0282] Next, the in-vehicle device 101 performs a service determination process using the corrected prediction results C21, C22, and C13 to determine whether or not to execute service S. Here, it is assumed that the in-vehicle device 101 determines to execute service S (step ST313).

[0283] Next, the in-vehicle device 101 executes the service S by communicating with the server 201 (step ST314).

[0284] Next, the in-vehicle device 101 transmits to the server 301 communication result information indicating the actual value J1 of the communication speed Vc, the actual value J2 of the communication delay time Td, and whether or not a handover H occurred during the execution of the service S (step ST315).

[0285] Next, when the server 301 receives the communication result information from the in-vehicle device 101, the server 301 uses the received communication result information to update the communication performance information stored in the storage unit 25. The server 301 also uses the communication performance information, as well as the predicted value of the performance value J1, the predicted value of the performance value J2, and the prediction result regarding whether or not handover H will occur, which are stored in the storage unit 25, to update the error rates U1, U2, and the accuracy rate U3 stored in the storage unit 25 (step ST316).

[0286] In the communication system 501 according to the embodiment of the present disclosure, the auxiliary information includes determination information that can determine the reliability of the prediction result of the communication quality required for the service S and correction information that is used to correct the prediction result of the communication quality, but this is not limited to this. The auxiliary information may include one of the determination information and the correction information, but not the other.

[0287] Furthermore, in the communication system 501 according to an embodiment of the present disclosure, the determination information M1 capable of determining the reliability of the prediction result C21 of the communication speed Vc includes the size La of the prediction target region of the communication speed Vc, the number of actual values ​​J1, the ratio Ya of the number of actual values ​​J1 in the prediction target time period Tp, and the error rate U1 of the past prediction result C21. However, this is not limited to this. The determination information M1 may also include part of the size La, the number of actual values ​​J1, the ratio Ya, and the error rate U1. Furthermore, the determination information M1 may include other information capable of determining the reliability of the prediction result C21 of the communication speed Vc instead of part or all of the size La, the number of actual values ​​J1, the ratio Ya, and the error rate U1.

[0288] Furthermore, in the communication system 501 according to an embodiment of the present disclosure, the determination information M2 capable of determining the reliability of the prediction result C22 of the communication delay time Td is configured to include the size Lb of the prediction target region of the communication delay time Td, the number of actual values ​​J2, the ratio Yb of the number of actual values ​​J2 in the prediction target time period Tp, and the error rate U2 of the past prediction result C22, but this is not limited to this. The determination information M2 may also be configured to include part of the size Lb, the number of actual values ​​J2, the ratio Yb, and the error rate U2. Furthermore, the determination information M2 may be configured to include other information capable of determining the reliability of the prediction result C22 of the communication delay time Td instead of part or all of the size Lb, the number of actual values ​​J2, the ratio Yb, and the error rate U2.

[0289] Furthermore, in the communication system 501 according to the embodiment of the present disclosure, the server 301 is configured to perform a prediction process R1 that predicts communication quality using a trained model and a prediction process R2 that predicts communication quality using communication performance information that indicates past communication performance in the vehicle 1, but this is not limited to this. The server 301 may be configured to perform one of the prediction process R1 and the prediction process R2, but not the other.

[0290] In addition, in the communication system 501 according to the embodiment of the present disclosure, the in-vehicle device 101 is configured to communicate with the server 201 in accordance with TCP, but this is not limiting. The in-vehicle device 101 may be configured to communicate with the server 201 in accordance with a communication standard other than TCP.

[0291] In the communication system 501 according to the embodiment of the present disclosure, the correction information includes parameters p1 and p2 used to correct the prediction result C21 for the communication speed Vc, but this is not limiting. The correction information may include parameters used to correct a prediction result for a type of communication quality different from the communication speed Vc.

[0292] Furthermore, in the communication system 501 according to the embodiment of the present disclosure, the in-vehicle device 101 is configured to acquire a plurality of prediction results corresponding to a plurality of times included in the time slot Tp to be predicted and auxiliary information for each time, but this is not limited to this. The in-vehicle device 101 may be configured to acquire a prediction result corresponding to one time included in the time slot Tp to be predicted and auxiliary information at that time.

[0293] Furthermore, some or all of the functions of the server 301 according to the embodiment of the present disclosure may be provided by cloud computing, i.e., the server 301 according to the embodiment of the present disclosure may be a cloud server configured by a plurality of servers.

[0294] The above-described embodiments should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the above description, and is intended to include all modifications within the meaning and scope of the claims.

[0295] Each process (each function) in the above-described embodiments is realized by a processing circuit including one or more processors. The processing circuit may be configured as an integrated circuit or the like that combines one or more memories, various analog circuits, and various digital circuits in addition to the one or more processors. The one or more memories store programs (instructions) that cause the one or more processors to execute each of the processes. The one or more processors may execute each of the processes according to the program read from the one or more memories, or may execute each of the processes according to a logic circuit designed in advance to execute each of the processes. The processor may be any of various processors suitable for computer control, such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit). Note that the physically separated processors may cooperate with each other to execute the processes. For example, the processors installed in the physically separated computers may cooperate with each other via a network such as a LAN (Local Area Network), a WAN (Wide Area Network), or the Internet to execute the processes. The program may be installed into the memory from an external server device or the like via the network, or may be distributed in a state stored on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), or a semiconductor memory, and then installed into the memory from the recording medium.

[0296] The above description includes the following additional features: [Additional Note 1] An in-vehicle device mounted in a vehicle, the in-vehicle device comprising: a processing circuit, the processing circuit acquiring, from a providing device, a prediction result of communication quality required for a service executed in the vehicle, and auxiliary information used for determining the prediction result.

[0297] [Supplementary Note 2] A providing device comprising a processing circuit, wherein the processing circuit predicts communication quality required for a service executed in a vehicle, creates auxiliary information to be used for determining the prediction result of the prediction unit, and transmits the prediction result of the prediction unit and the auxiliary information created by the creation unit.

[0298] [Supplementary Note 3] A communication prediction method in a providing device, comprising: a step of predicting a communication quality required for a service executed in a vehicle; a step of creating auxiliary information used to judge the prediction result of the prediction unit; and a step of transmitting the prediction result of the prediction unit and the created auxiliary information.

[0299] REFERENCE SIGNS LIST 1 Vehicle 10, 10A, 10B, 10C Communication port 11 In-vehicle communication unit 12 Service management unit 13 Out-vehicle communication unit 14 Communication management unit 15 Correction unit 16, 25 Storage unit 21 Communication unit 22 Prediction unit 23 Creation unit 24 Update unit 51, 51A, 51B, 51C, 51D Ethernet cable 101 In-vehicle device 202, 202A, 202B, 202C, 202D In-vehicle equipment 151 External network 161 Wireless base station device 301 Server 401 In-vehicle network 501 Communication system

Claims

1. An on-board device mounted in a vehicle, the on-board device comprising an acquisition unit that acquires, from a providing device, a predicted result regarding the communication quality required for a service executed in the vehicle and auxiliary information used to judge the predicted result.

2. The in-vehicle device according to claim 1, wherein the auxiliary information includes information that allows the reliability of the prediction result to be determined.

3. The in-vehicle device according to claim 1 or 2, wherein the auxiliary information includes information used to correct the prediction result.

4. The in-vehicle device according to any one of claims 1 to 3, wherein the auxiliary information includes the size of a region for which the communication quality is to be predicted.

5. The in-vehicle device according to any one of claims 1 to 4, wherein the auxiliary information includes the number of pieces of data used in predicting the communication quality.

6. An in-vehicle device according to any one of claims 1 to 5, wherein the auxiliary information includes a ratio of the number of data in the time period for which the communication quality is to be predicted to the number of data used to predict the communication quality in other time periods different from the time period for which the communication quality is to be predicted.

7. The in-vehicle device according to any one of claims 1 to 6, wherein the auxiliary information includes information indicating a judgment result regarding the past prediction result.

8. An in-vehicle device according to any one of claims 1 to 7, wherein the prediction result is an output value of the trained model when the providing device predicts the communication quality using the trained model.

9. The in-vehicle device according to any one of claims 1 to 8, wherein the service is executed by communication conforming to TCP, and the auxiliary information includes a parameter relating to a limit on the communication speed at the start of execution of the service.

10. An in-vehicle device according to any one of claims 1 to 9, wherein the acquisition unit acquires a plurality of prediction results corresponding to a plurality of times included in the time period for which the communication quality is to be predicted, and the auxiliary information for each of the times.

11. A providing device comprising: a prediction unit that predicts the communication quality required for a service executed in a vehicle; a creation unit that creates auxiliary information used to judge the prediction result of the prediction unit; and a transmission unit that transmits the prediction result of the prediction unit and the auxiliary information created by the creation unit.

12. A communication system comprising an on-board device mounted in a vehicle and a providing device, wherein the providing device transmits to the on-board device a predicted result of communication quality required for a service executed in the vehicle and auxiliary information used to make a judgment about the predicted result.

13. A vehicle assistance method in an on-board device installed in a vehicle, the vehicle assistance method including a step of obtaining, from a providing device, a prediction result regarding the communication quality required for a service executed in the vehicle and auxiliary information used to make a judgment regarding the prediction result.

14. A vehicle assistance program used in an on-board device installed in a vehicle, which causes a computer to function as an acquisition unit that acquires, from a providing device, predicted results regarding the communication quality required for a service executed in the vehicle and auxiliary information used to make judgments about the predicted results.

15. A communication prediction program used in a providing device, which causes a computer to function as: a prediction unit that predicts the communication quality required for a service executed in a vehicle; a creation unit that creates auxiliary information used to judge the prediction result of the prediction unit; and a transmission unit that transmits the prediction result of the prediction unit and the auxiliary information created by the creation unit.

Images