JP7462179B2 - Information processing device, information processing method, program, and trained model - Google Patents

Description

The present invention relates to an information processing device, an information processing method, a program, etc.

As an example of a device (system) that measures a certain measurement target, there is one that can measure driving information such as vehicle position information and acceleration information (for example, Patent Document 1).

JP 2022-58421 A

This is just one example, but when considering developing a new service (business) using measurement data obtained by a device that measures a specific measurement target as described above, it is normal to actually use the device to obtain a huge amount of measurement data and evaluate it, which incurs time and monetary costs.
In addition, currently, the locations where measurement devices can be installed are limited, and the installation of more measurement devices is desired. However, due to cost issues, it is necessary to install the devices in locations where data can be collected efficiently.

According to a first aspect of the present invention, an information processing device includes a data generation unit that generates estimated data of second measurement data measured by a second device that measures information about a moving body and is different from the first device, based on first measurement data measured by a first device that measures information about the moving body.
According to a second aspect of the present invention, an information processing method includes generating estimated data of second measurement data measured by a second device that measures information about a moving body and is different from the first device, based on first measurement data measured by a first device that measures information about a moving body.
According to a third aspect of the present invention, a program to be executed by a computer causes the computer to generate estimated data of second measurement data measured by a second device that measures information about a moving body and is different from the first device, based on first measurement data measured by a first device that measures information about a moving body.
According to a fourth aspect of the present invention, the information processing device includes a first determination unit that determines, based on first measurement data measured by a first device that measures information regarding a moving body, a measurement period of first measurement data necessary for evaluating driving based on second measurement data measured by a second device that measures information regarding a moving body and is different from the first device.
According to a fifth aspect of the present invention, the information processing method includes determining, based on first measurement data measured by a first device that measures information related to a moving body, a measurement period of first measurement data required for evaluating driving based on second measurement data measured by a second device that measures information related to a moving body and is different from the first device.
According to a sixth aspect of the present invention, a program to be executed by a computer causes the computer to determine, based on first measurement data measured by a first device that measures information regarding a moving body, a measurement period of first measurement data necessary for evaluating driving based on second measurement data measured by a second device that measures information regarding a moving body and is different from the first device.
According to a seventh aspect of the present invention, an information processing device includes a generation unit that generates pseudo-second measurement data for a specified period based on first measurement data measured by a first device that measures information about the moving body, the second measurement data being measured by a second device that measures information about the moving body and is different from the first device, an acquisition unit that acquires data loss ratio information in the generated pseudo-second measurement data, and a judgment unit that judges the validity of the pseudo-second measurement data for the specified period based on the data loss ratio information.
According to an eighth aspect of the present invention, an information processing method includes generating pseudo-second measurement data for a predetermined period based on first measurement data measured by a first device that measures information about a moving body, the second measurement data being measured by a second device that measures information about the moving body and is different from the first device, obtaining data loss ratio information in the generated pseudo-second measurement data, and determining validity of the pseudo-second measurement data for the predetermined period based on the data loss ratio information.
According to a ninth aspect of the present invention, a program to be executed by a computer causes the computer to generate pseudo-second measurement data for a specified period of time based on first measurement data measured by a first device that measures information about a moving body, the second measurement data being measured by a second device that measures information about a moving body and is different from the first device, obtain data loss rate information in the generated pseudo-second measurement data, and determine the validity of the pseudo-second measurement data for the specified period of time based on the data loss rate information.
According to a tenth aspect of the present invention, an information processing device is provided, which is a roadside unit that collects information regarding the traveling of a mobile body based on a method compatible with ETC 2.0, and includes an existing roadside unit position information acquisition unit that acquires existing roadside unit position information, which is position information of an existing roadside unit that has already been installed, a first installation candidate facility position information acquisition unit that acquires first installation candidate facility position information, which is position information of at least one facility where the roadside unit can be newly installed, a traveling information acquisition unit that acquires traveling information collected based on a method different from the method compatible with ETC 2.0, the traveling information including at least information regarding the position of the mobile body, and a calculation unit that calculates first collection degree information regarding the collection degree when it is assumed that the traveling information is collected by a roadside unit that corresponds to the existing roadside unit position information and the first installation candidate facility position information, based on the method compatible with ETC 2.0.
According to an eleventh aspect of the present invention, an information processing method includes acquiring existing roadside unit position information, which is position information of an existing roadside unit that has already been installed, which is a roadside unit that acquires driving data of a moving body based on a method compatible with ETC 2.0; acquiring first installation candidate facility position information, which is position information of at least one facility where the roadside unit can be newly installed; acquiring driving information measured based on a method different from the method compatible with ETC 2.0, which includes at least information regarding the position of the moving body; and calculating first collection degree information regarding the collection degree of the driving information when measurement is virtually performed based on a method compatible with ETC 2.0, based on the existing roadside unit position information and the first installation candidate facility position information.
According to a twelfth aspect of the present invention, a program to be executed by a computer causes the computer to execute the following steps: acquire existing roadside unit position information, which is position information of an existing roadside unit that has already been installed, which acquires driving data of a moving body based on a method compatible with ETC 2.0; acquire first installation candidate facility position information, which is position information of at least one facility where the roadside unit can be newly installed; acquire driving information measured based on a method different from the method compatible with ETC 2.0, which includes at least information regarding the position of the moving body; and calculate first collection degree information regarding the collection degree of the driving information when a measurement is virtually performed based on a method compatible with ETC 2.0, based on the existing roadside unit position information and the first installation candidate facility position information.
According to a tenth aspect of the present invention, an information processing device includes a mobile body position information acquisition unit that acquires mobile body position information from a first device provided in a mobile body based on a method different from a method compatible with ETC 2.0, a density information acquisition unit that acquires density information of the mobile body position information based on the acquired mobile body position information, and an identification unit that identifies a location for installing a new roadside unit that acquires driving information based on a method compatible with ETC 2.0 based on the acquired density information.
According to an eleventh aspect of the present invention, an information processing device includes a roadside unit position information acquisition unit that acquires existing roadside unit position information, which is position information of at least one existing roadside unit, an output unit that outputs the roadside unit position information, and an identification unit that identifies a position for installing a new roadside unit based on a user input performed in response to the output.
According to a twelfth aspect of the present invention, an information processing device includes an existing roadside unit position information acquisition unit that acquires existing roadside unit position information, which is position information of at least one existing roadside unit that acquires driving information based on a method compatible with ETC 2.0, and an identification unit that identifies a position for installing a new roadside unit based on the existing roadside unit position information.
According to a thirteenth aspect of the present invention, an information processing device includes a facility location information acquisition unit that acquires facility location information, which is location information of a facility where a new roadside unit that acquires driving information based on a method compatible with ETC 2.0 can be installed, and an identification unit that identifies a location for installing the new roadside unit based on the facility location information.
According to a fourteenth aspect of the present invention, an information processing method includes acquiring mobile body position information from a first device provided in a mobile body based on a method different from a method compatible with ETC 2.0, acquiring density information of the mobile body position information based on the acquired mobile body position information, and identifying a location for installing a new roadside unit that acquires driving information based on a method compatible with ETC 2.0 based on the acquired density information.
According to a fifteenth aspect of the present invention, an information processing method includes acquiring existing roadside unit position information which is position information of at least one existing roadside unit, outputting the roadside unit position information, and identifying a position for installing a new roadside unit based on user input performed in response to the output.
According to a sixteenth aspect of the present invention, an information processing method includes acquiring existing roadside unit position information, which is position information of at least one existing roadside unit that acquires driving information based on a method compatible with ETC 2.0, and identifying a position for installing a new roadside unit based on the existing roadside unit position information.
According to a seventeenth aspect of the present invention, an information processing method includes acquiring facility location information, which is location information of a facility where a roadside unit that acquires driving information based on a system compatible with ETC 2.0 can be newly installed, and identifying a location for installing the new roadside unit based on the facility location information.
According to an 18th aspect of the present invention, a program to be executed by a computer causes the computer to acquire mobile body position information from a first device provided on a mobile body based on a method different from a method compatible with ETC 2.0, acquire density information of the mobile body position information based on the acquired mobile body position information, and identify a location for installing a new roadside unit that acquires driving information based on a method compatible with ETC 2.0 based on the acquired density information.
According to a nineteenth aspect of the present invention, a program to be executed by a computer causes the computer to acquire existing roadside unit position information, which is position information of at least one existing roadside unit, output the roadside unit position information, and identify a position for installing a new roadside unit based on user input performed in response to the output.
According to a twentieth aspect of the present invention, a program to be executed by a computer causes the computer to acquire existing roadside unit position information, which is position information of at least one existing roadside unit that acquires driving information based on a method compatible with ETC 2.0, and to identify a position for installing a new roadside unit based on the existing roadside unit position information.
According to a twenty-first aspect of the present invention, a program to be executed by a computer causes the computer to acquire facility location information, which is location information of a facility where a new roadside unit that acquires driving information based on a method compatible with ETC 2.0 can be installed, and to identify a location for installing the new roadside unit based on the facility location information.

FIG. 2 is an explanatory diagram of the principle according to the embodiment. FIG. 2 is an explanatory diagram of the principle according to the embodiment. 1 is a flowchart showing an example of a processing flow according to an embodiment. FIG. 2 is a diagram showing an example of a functional configuration of a server according to the first embodiment; FIG. 2 is a diagram illustrating the principle according to the first embodiment. FIG. 2 is a diagram illustrating the principle according to the first embodiment. FIG. 2 is a diagram illustrating the principle according to the first embodiment. 1 is a flowchart showing an example of a processing flow according to a first embodiment. FIG. 11 is a diagram showing an example of information stored in a storage unit of a server according to a second embodiment. 10 is a flowchart showing an example of a process flow according to a second embodiment. FIG. 11 is a diagram showing an example of a histogram according to the second embodiment. FIG. 11 is a diagram showing an example of a histogram according to the second embodiment. FIG. 13 is a diagram showing an example of information stored in a storage unit of a server according to a third embodiment. FIG. 13 is a diagram illustrating the principle according to the third embodiment. FIG. 13 is a diagram illustrating the principle according to the third embodiment. FIG. 13 is a diagram illustrating the principle according to the third embodiment. 13 is a flowchart showing an example of a process flow according to the third embodiment. FIG. 13 is a diagram showing an example of information stored in a storage unit of a server according to a fourth embodiment; FIG. 13 is a diagram illustrating the principle according to the fourth embodiment. 13 is a flowchart showing an example of a process flow according to the fourth embodiment. FIG. 13 is a diagram illustrating the principle according to the fourth embodiment. FIG. 13 is a diagram illustrating the principle according to the fourth embodiment. FIG. 13 is a diagram illustrating the principle according to the fourth embodiment.

Hereinafter, an example of an embodiment of the present invention will be described with reference to the drawings.
In the description of the drawings, the same elements are denoted by the same reference numerals, and duplicate descriptions may be omitted.
However, the components described in this embodiment are merely examples and are not intended to limit the scope of the present invention.

<Embodiment>
An example of an embodiment for realizing the present invention will be described below.

<Principle>
1.1 Principle of Data Estimation Fig. 1 is a diagram showing a flow of processing performed by an information processing device according to an aspect of the present embodiment.
The information processing device has, as a functional unit, a second device measurement data estimation unit 11 that generates (estimates) data when a measurement object is measured by a second device different from the first device, based on data (hereinafter referred to as "first device measurement data") of the measurement object measured by the first device. Note that measurement may also be detection or measurement. The same can be applied below. Moreover, the data generated by the second device measurement data estimation unit 11 is referred to as "estimated second device measurement data."

The first device may be, for example, a device that measures the measurement object at a predetermined time interval (hereinafter referred to as the "predetermined time interval"). In this case, the first device measurement data is data obtained by measuring the measurement object by the first device at each predetermined time interval.

The second device may be, for example, a device whose sampling unit (which may also be called a unit of measurement) when measuring the measurement object is different from that of the first device. For example, if the first device is a device that measures the measurement object at predetermined time intervals using time as a sampling unit as described above, the second device may be a device that measures the measurement object using a sampling unit other than time.
In carrying out the learning and other processes described below, it is of course preferable to use measurement data from the second device as teacher data, but the present invention is characterized in that data generated from measurement data from the first device can be used. Transfer learning is an example of an approach that uses other data to process target data, but transfer learning is merely a method of using a model learned for a certain task as a starting point for a model that executes a similar task, and is fundamentally different from the present invention, which uses data generated from other data on a zero-base basis.

For example, if the measurement target is the position (position information) of a device, the first device can be a device that measures its position at a predetermined time interval.
In this case, the second device may be a device that measures position using, for example, a sampling unit of "distance" and that measures position at predetermined distance intervals (hereinafter referred to as "predetermined distance intervals").

In this example, if the specified time interval is "1 second interval" and the specified distance interval is "100 m", the second device measurement data estimation unit 11 can generate estimated second device measurement data, for example, by thinning (resampling) the position measurement data for each second included in the first device measurement data to every "100 m" based on the position information.

When a set of multiple first device measurement data is a "first device measurement data set," the second device measurement data estimation unit 11 can generate estimated second device measurement data corresponding to each of the first device measurement data by performing the above processing on each of the first device measurement data included in the first device measurement data set. The set of estimated second device measurement data generated in this manner may be referred to as an "estimated second device measurement data set."

In addition, at least one of the first device and the second device may be a device that measures an object based on multiple sampling units (e.g., distance and acceleration: a device that obtains distance when the acceleration value is equal to or greater than a threshold value (or exceeds the threshold value)).
Other devices may also be used.

1.2 Machine Learning Using Estimated Measurement Data Fig. 2 is a diagram for explaining the principle of learning and inference using estimated measurement data in this embodiment.

(1) During Learning As shown in FIG. 2(1), the information processing device has, for example, the above-mentioned second device measurement data estimation unit 11, first model processing unit 13, and second model generation unit 15 as functional units.

The first model processing unit 13 may be a functional unit that, for example, takes a predetermined model (existing model) as the first model, inputs certain measurement data to the first model, and calculates (calculates) predetermined quantities. Note that the quantities may be regarded as calculation targets. The same applies below.
In this embodiment, the first model processing unit 13 calculates various predetermined quantities using the above-mentioned first device measurement data as an input to the first model.

The second model generating unit 15 can be, for example, a functional unit that generates a second model capable of calculating the same quantities as the above-mentioned first model.
In this embodiment, the second model generation unit 15 generates the second model by machine learning using, as a learning data set, the estimated second device measurement data generated by the above-mentioned second device measurement data estimation unit 11 and the quantities calculated by the first model processing unit 13. The second model generated in this manner is referred to as a "learned second model."
The second model is not limited to a machine learning model. For example, the second model generating unit 15 may be a mathematical statistical model such as a linear or nonlinear regression model, or a connectionist model such as a deep neural network (DNN).
The quantities calculated by the first model processing unit 13 are calculated using the existing first model and can be considered desirable data (correct data), so they are referred to here as "teacher quantities" (teacher data).

(2) At the Time of Inference As shown in FIG. 2(2), the information processing device has, for example, a trained second model processing unit 17 as a functional unit.

The learned second model processing unit 17 can be, for example, a functional unit that calculates various specified quantities using data measured by a second device (hereinafter referred to as ``second device measurement data'') as input to the learned second model.
In other words, the learned second model processing unit 17 takes as input data obtained by actually measuring the object to be measured by the second device, infers certain quantities using the learned second model, and outputs the inference results (hereinafter referred to as ``quantity inference results'').

In addition to the above, for example, the estimated second device measurement data may be used as input data to the first model, and some previously held data corresponding to the first device measurement data accumulated in the past (e.g., data on elements that may affect various quantities, data on elements that are related to various quantities) may be used as training data to generate the second model by machine learning or the like.

<Processing>
FIG. 3 is a flowchart showing an example of the flow of the process shown in FIG.
First, the information processing device performs a learning dataset generation process.
In the learning dataset generation process, the information processing device performs processing of loop A (A1 to A7) on the first device measurement data of each processing target (in the example of Figure 2, each of the first device measurement data included in the first device measurement dataset).
The first device measurement data to be processed may be, for example, data having the same measurement period.

In the processing of loop A, the second device measurement data estimation unit 11 generates estimated second device measurement data based on the first device measurement data (A3).
Moreover, the first model processing unit 13 calculates various teaching quantities based on the first device measurement data and the first model (A5).
Then, the information processing device moves to processing the next first device measurement data.

For example, when processes A3 and A5 have been performed for all of the first device measurement data to be processed, the information processing device ends the processing of loop A (A7).

Then, the information processing device performs a learning process (A9). Specifically, the second model generation unit 15 generates a learned second model, for example, by supervised learning, based on the estimated second device measurement data generated in the processing of loop A and the teacher quantities generated in the processing of loop A.

Next, the information processing device performs an inference process (A11). Specifically, the trained second model processing unit 17 calculates various quantities based on the trained second model generated in A9 and the second device measurement data of each processing target (in the example of FIG. 2, each of the second device measurement data included in the second device measurement data set). The calculation results are then treated as various quantity inference results.
Then, the information processing device ends the process.

It is also possible to omit step A11 and perform the process up to generating the model.
Alternatively, only step A3 may be performed, and the process may include generating estimated second device measurement data based on the first device measurement data.

The principle has been explained above, but the measurement target can be, for example, information about a moving object.

The moving object may include, for example, at least one of the following:
・Four-wheeled vehicles ・Two-wheeled vehicles ・Personal mobility ・Ships ・Railways ・Air vehicles (at least one of drones and aircraft)

Furthermore, the information about the moving object may include, for example, at least one of the following:
・Location information of a moving object ・Speed information of a moving object ・Acceleration information of a moving object ・Orientation information of a moving object (directional information)
・Angular velocity information of a moving object ・Time information of a moving object (time information, etc.)
Note that information including at least one of these pieces of information may be used as the travel information of the mobile object.

Furthermore, the measurement target is not limited to information relating to a moving body (travel information, etc.), but may be other information.
For example, the measurement target may be "sound," the sampling unit of the first device may be "time," and the sampling unit of the second device may be "frequency," and similar processing may be performed.
In addition, for example, processing similar to the above may be performed by measuring information related to a living body (blood pressure, pulse rate, heart rate, respiratory rate, body temperature, brain waves, etc.). In this case, for example, the sampling unit of the first device may be set to "time" and the sampling unit of the second device may be set to "frequency", and the same processing as the above may be performed.

<Example>
Next, an embodiment of a server, which is an example of the information processing device, will be described.
In the following, an embodiment will be described in which a vehicle such as a four-wheeled automobile is used as the moving body.
The information processing device is not limited to a server, but may be, for example, an electronic device (which may be called a terminal) such as a personal computer, a mobile phone including a smartphone, a PDA, various tablet terminals, etc. Furthermore, these devices including the server may be called a computer device.
However, the embodiments to which the present invention can be applied are not limited to the embodiments described below.

First Example
Recently, ETC 2.0 is becoming more and more popular. For example, it is conceivable to develop telematics insurance products as a service (business) based on ETC 2.0.
However, this requires measurement data to be collected by having a driver actually drive a vehicle equipped with an ETC 2.0-compatible device (e.g., an ETC 2.0-compatible on-board unit or a DSRC (Dedicated Short Range Communications)-compatible on-board unit), as well as corresponding accident data, etc., which incurs time and monetary costs.

In this case, for example, by estimating (reproducing) data measured by a device compatible with ETC 2.0 from data measured (acquired data) by a device other than a device compatible with ETC 2.0, it is possible to develop insurance products, etc., without collecting data using a device compatible with ETC 2.0.

When developing a telematics insurance product, it may be possible to generate a model (hereinafter referred to as a "driving score model" (driving quality score model)) of a score (hereinafter referred to as a "driving score" (driving quality score)) regarding a driver's driving (driving quality).
Generally speaking, the driving score model can be constructed as a model in which, for example, at least one of the following factors (including factors standardized per unit time or unit distance, etc.) affects the score value:
Number of times steering suddenly (or frequency of steering suddenly or distribution of steering operation degree including steering suddenly)
Number of sudden decelerations (or frequency of sudden decelerations or distribution of decelerations including sudden decelerations)
- Number of sudden accelerations (or frequency of sudden accelerations or distribution of acceleration including sudden accelerations)
- Number of times speeding (or frequency of speeding or distribution of speeds including speeding)

For example, a driving score model can be constructed in which the greater the number of these elements (or the greater the frequency or degree of sudden operation or speeding), the higher the risk of an accident and the higher the driving score (or, conversely, the lower the score).
One of the uses (methods of utilizing) of this driving score is, for example, to reflect the driving score in telematics insurance premiums (determining premiums based on the driving score).

In this embodiment, although the above elements are not necessarily reflected precisely, a method is described in which estimated ETC 2.0 measurement data is generated that reflects at least the driver's general driving tendencies, and then a driving score model is generated based on this estimated ETC 2.0 measurement data.

The contents described in the first embodiment can also be applied to the other embodiments and other variations.

<Functional configuration>
FIG. 4 is a diagram illustrating an example of the functional configuration of the server 100 in this embodiment.
The server 100 can be a computer system that includes, for example, a processing unit (control unit) 110, an operation unit 120, a display unit 130, a sound output unit 140, a communication unit 150, a clock unit 160, and a memory unit 190, all of which are connected by a bus B.

The processing unit 110 is, for example, a processing device or control device that comprehensively controls each part of the server 100 and performs various processes according to various programs such as a system program stored in the memory unit 190, and is configured with processors such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a DSP (Digital Signal Processor), and integrated circuits such as an ASIC (Application Specific Integrated Circuit) and an FPGA (Field Programmable Gate Array).

The operation unit 120 is configured with input devices, such as operation buttons and operation switches, that allow the user to input various operations to the server 100. The operation unit 120 may also have a touch panel that is configured integrally with the display unit 130, and this touch panel may function as an input interface between the user and the server 100. From the operation unit 120, an operation signal according to the user's operation is output to the processing unit.

The display unit 130 is a display device including an LCD or the like, and performs various displays based on the display signal output from the processing unit 110. The display unit 130 may be integrated with a touch panel to form a touch screen.

The sound output unit 140 is a sound output device including a speaker, etc., and outputs various sounds based on the sound output signal output from the processing unit 110.

The communication unit 150 is a communication device for sending and receiving information used within the device to and from an external device. Various communication methods can be applied to the communication unit 150, such as a wired connection via a cable that complies with a specific communication standard, a connection via an intermediate device that also serves as a charger called a cradle, or a wireless connection using wireless communication.

The clock unit 160 is a built-in clock of the server 100, and outputs time information (timekeeping information). The clock unit 160 is configured to include, for example, a clock using a crystal oscillator.
The clock unit 160 may be configured to include a clock that complies with the Network Identity and Time Zone (NITZ) standard or the like.

The memory unit 190 is a storage device that includes volatile or non-volatile memory such as ROM, EEPROM, flash memory, RAM, etc., a hard disk device, etc.

In this embodiment, the memory unit 190 stores, for example, a driving score model processing program 191, a cigarette lighter type device measurement database 192, an ETC 2.0 measurement database 193, an estimated ETC 2.0 measurement database 194, a first driving score model data 195, and a learned second driving score model data 196.

The driving score model processing program 191 is a program that is read by the processing unit 110 and executed as the driving score model processing described below.

The cigarette lighter type device measurement database 192 is, for example, a database in which cigarette lighter type device measurement data including driving information measured by a cigarette lighter type device mounted on a vehicle is accumulated and stored.
The cigarette lighter type device is a type of first device, and the cigarette lighter type device measurement data can be a type of first measurement data (first device measurement data).

Note that the processing unit 110 may perform processing by adding, for example, at least one of the following pieces of information (attribute information) in addition to information about the vehicle (mobile body) as the first device measurement data.
・Vehicle information (model, year, type, etc.)
・Driver information (gender, age, address, purpose, etc.)

A cigarette lighter type device is a device that can be used by inserting it into a vehicle's accessory socket, and can measure vehicle driving information (information about a moving object) based on, for example, various built-in sensors and an inertial measurement unit (IMU (Inertial Measurement Unit)).

The first device is not limited to a cigarette lighter type device, and may be any device that can obtain information about a moving object and is directly or indirectly provided on the moving object. For example, the first device may be at least one of the following devices:
・Driving recorder ・Digital tachograph (digital tachograph)
・Any on-board device installed or built into a vehicle (including connected cars)
・In-vehicle devices with communication functions (e.g., Bluetooth (registered trademark) communication functions) that work in conjunction with terminals such as smartphones to collect data, and such terminals ・Standalone devices including terminals such as smartphones ・Vehicle-mounted and retrofitted in-vehicle devices including the above, or devices including terminals such as smartphones

The driving information may include, for example, information such as position, speed, acceleration, and angular velocity, and any sensor capable of measuring this information may be provided in the cigarette lighter type device. The device may be configured to be capable of measuring at least one of these pieces of information.

The driving information measured by the cigarette lighter type device can be transmitted to the server 100 or a cloud server, for example, at any time or at regular intervals.

The cigarette lighter device may also be provided with a satellite positioning unit (satellite positioning sensor), which is a unit (sensor) that detects position and speed using a satellite positioning system such as GPS (Global Positioning System), or a UWB positioning unit (UWB positioning sensor), which is a unit (sensor) that calculates position using UWB (Ultra Wide Band).

The satellite positioning unit may have, for example, an RF receiving circuit that converts RF (Radio Frequency) signals, including positioning satellite signals transmitted from positioning satellites and received by an antenna (not shown), into digital signals, and a baseband processing circuit that performs correlation calculation processing or the like on the digital signals output from the RF receiving circuit to capture the positioning satellite signals, and outputs information such as satellite orbit data and time data extracted from the positioning satellite signals as positioning information. Note that Doppler positioning or the like may be used to detect information such as speed in addition to position.

The inertial measurement unit (IMU) may have, for example, an inertial sensor that detects information necessary to calculate driving information by inertial navigation calculations. The inertial sensor may include, for example, a three-axis acceleration sensor and a three-axis gyro sensor, and may output the acceleration detected by the acceleration sensor and the angular velocity detected by the gyro sensor.

The UWB positioning unit may include, for example, an ultra-wideband RF receiving circuit that converts an ultra-wideband RF signal, including an ultra-wideband pulse signal for positioning transmitted from a positioning beacon and received by an antenna not shown, into a digital signal, and a relative position calculation processing circuit that calculates the relative position between the device and the positioning beacon based on the digital signal output from the ultra-wideband RF receiving circuit.
In addition, the UWB positioning unit may or may not function as a positioning beacon by transmitting an ultra-wideband RF signal including an ultra-wideband pulse signal for positioning from an antenna not shown.

For example, the cigarette lighter type device measurement database 192 can store the measurement data of a plurality of drivers for each measurement period. In other words, the measurement data of a plurality of drivers can be stored in association with the measurement period.
The measurement period may be, for example, a period in months (one month, two months, ..., six months, etc.) or other units of time. Furthermore, the measurement data may be stored by classifying it by road type, by region, etc.

The ETC 2.0 measurement database 193 is a database in which ETC 2.0 measurement data including driving information measured, for example, by an ETC 2.0 onboard unit (ETC 2.0 compatible onboard unit, DSRC compatible onboard unit) installed in a vehicle or an ETC 2.0 compatible car navigation device that can be linked to the same is stored.
The ETC 2.0 vehicle-mounted unit is a type of a first device, and the ETC 2.0 measurement data can be a type of a second measurement data (second device measurement data).

ETC 2.0 measurement data can be acquired by server 100 based on the probe information recorded in the ETC 2.0 vehicle-mounted unit being collected by devices such as ITS spots and route information collection devices (collectively referred to as "roadside units" below) installed (deployed) on roads.

Roadside units are installed at several thousand locations on expressways and national roads across the country.
The roadside unit and the ETC 2.0 vehicle-mounted unit are configured, for example, to be capable of two-way communication, and in addition to the roadside unit collecting probe information from the ETC vehicle-mounted unit, information to support safe driving, such as route guidance information based on road conditions, information on obstacles ahead, and information on accident-prone locations, can be provided from the roadside unit to the ETC 2.0 vehicle-mounted unit.

The ETC 2.0 vehicle-mounted device also has, for example, the various sensors and units described above, and can measure and record various types of driving information.
When a vehicle equipped with an ETC 2.0 on-board unit passes near the roadside unit, the probe information can be transmitted to the server 100 or a probe server via the roadside unit.
In addition, various sensors, units, etc. may be provided in an ETC 2.0 compatible car navigation device, etc., so that various driving information is recorded in the ETC 2.0 vehicle unit based on information output (transmitted) from the ETC 2.0 compatible car navigation device.

Here, the probe information may include, for example, the following information:
・Basic information ・Driving history information ・Behavior history information

The basic information may include, for example, information about the ETC 2.0 onboard unit, information about the ETC 2.0 compatible car navigation device, information about the vehicle, etc.

The driving history information may include, for example, time information, position information, speed information, and the like.
This driving history information may be recorded (accumulated) whenever it is detected that the vehicle has traveled 200 meters, for example. In addition, it may be recorded when the vehicle's heading changes by 45 degrees or more.
The value "200 m" is based on the ETC2.0 specifications, but is not limited to this and may be, for example, "100 m." Similarly, "45 degrees" may be, for example, "22.5 degrees." For convenience, distances such as "200 m" based on the ETC2.0 specifications are referred to as "unit distances."
As will be described later, it is also possible to change the unit distance depending on the road type or speed range.

The behavior history information may include, for example, time information, position information, orientation information, acceleration information (such as forward/backward acceleration and left/right acceleration), and angular velocity information (such as yaw angular velocity information).
This behavior history information may record (accumulate) peak values when any of the longitudinal acceleration, lateral acceleration, and yaw angular velocity exceeds a threshold value (for example, a longitudinal acceleration threshold value of -0.25 G, a lateral acceleration threshold value of ±0.25 G, and a yaw angular velocity threshold value of ±8.5 deg/sec. These threshold values are merely examples). For example, longitudinal acceleration of 0.25 G or more indicates a sudden behavior, and may indicate that an action such as crisis avoidance has been taken.

In addition, the driving history information and behavior history information may also include road type information (information such as expressway, urban expressway, general road, etc., speed range information, etc.).

In addition, among the above information, for example, at least one of the driving history information and the behavior history information may be used as driving information.

In addition, in ETC 2.0, from the standpoint of protecting personal information, it is possible to prevent the acquisition of driving information (e.g., at least driving history information) for the start point of the journey (e.g., within 500 m of the starting point of the journey) and the end point of the journey (e.g., within 500 m of the end point of the journey).
Here, "not being acquired" may mean not being recorded in the ETC 2.0 vehicle-mounted device, or not being transmitted to the roadside device.

The estimated ETC 2.0 measurement database 194 is a database in which estimated ETC 2.0 measurement data generated based on cigarette lighter type device measurement data is cumulatively stored.
The estimated ETC 2.0 measurement data can be a type of estimated data (estimated second device measurement data).

The first driving score model data 195 is data of the first driving score model that serves as the basis for the driving score used to determine vehicle insurance premiums, etc.

The learned second driving score model data 196 is data of a model of a driving score generated by learning using estimated ETC 2.0 measurement data, etc.

Here, the driving score model may be, for example, a mathematical model for calculating a driving score for telematics insurance.
With telematics insurance, for example, an insurance company obtains driving information such as driving speed and braking method measured by equipment installed in a vehicle (car, etc.), and uses that information to analyze the driver's accident risk and calculate insurance premiums.

The driving score model can be, for example, a model capable of estimating the risk of accidents (accident risk) of a driver, and can be a model in which the driving score is calculated as a value in a numerical range such as "0 to 1", "0 to 10", or "0 to 100". For example, the model can be such that the higher the driving score, the higher the accident risk is estimated (the lower the driving score, the lower the accident risk is estimated). In this case, the insurance company can set the insurance premium higher for a higher driving score and lower the insurance premium lower for a lower driving score.
Conversely, a model may be used in which a higher driving score indicates a lower accident risk.

In addition, the input to the driving score model can be measurement data of the driving information, and the driving score can be calculated based on the measurement data of the driving information.
In this embodiment, for example, an existing first driving score model is used as a base, and a learned second driving score model is generated by the method described below.
Then, by inputting the driver's actual ETC 2.0 measurement data into the learned second driving score model, the driver's driving score can be calculated.

In addition, the server 100 may be a cloud server.
Furthermore, when the server 100 is separate from the cloud server, the server 100 may acquire the cigarette lighter type device measurement data from the cloud server.
The server performing the processing described below may acquire and store data similar to the data stored in the storage unit 190 of the server 100 .

Similarly, the probe server may be the server 100 .
Furthermore, when the server 100 is separate from the probe server, the server 100 may acquire ETC 2.0 measurement data from the probe server.
The server performing the processing described below may acquire and store data similar to the data stored in the storage unit 190 of the server 100 .

<Principle>
FIG. 5 is a diagram for explaining the principle of learning and inference using estimated ETC 2.0 measurement data in this embodiment, and corresponds to FIG.

(1) During Learning As shown in FIG. 5 (1), the processing unit 110 of the server 100 has, for example, an ETC 2.0 measurement data estimation unit 111, a first driving score model processing unit 113, and a second driving score model generation unit 115 as functional units.
The ETC 2.0 measurement data estimating unit 111 can be a type of the second device measurement data estimating unit 11 in FIG. 2(1).
The first driving score model processing unit 113 may be a type of the first model processing unit 13 in FIG. 2(1).
The second driving score model generation unit 115 may be a type of the second model generation unit 15 in FIG.

6 and 7 are diagrams showing an example of a method for generating estimated ETC 2.0 measurement data.
6, the left side of the drawing shows the change over time of the measurement position of the vehicle measured by the cigarette lighter device every second (hereinafter referred to as the "cigarette lighter device measurement position"). Note that this shows the change over time of the measurement position included in one cigarette lighter device measurement data.
Also, on the right side of the drawing, the change over time of the estimated ETC 2.0 measurement positions selected from the cigarette lighter type device measurement positions and serving as data elements of the estimated ETC 2.0 measurement data is shown. Each hatched rectangle is an estimated ETC 2.0 measurement position.
As an example, a method of generating estimated ETC 2.0 measurement data based on position information included in the travel information is illustrated here.

In addition, this cigarette lighter type device measurement data is represented as "D1", and the cigarette lighter type device measurement position included in the cigarette lighter type device measurement data "D1" is represented as "D1(p(t))" (t: time).
The estimated ETC 2.0 measurement data is represented as "D2", and the estimated ETC 2.0 measurement position included in this estimated ETC 2.0 measurement data is represented as "D2(p(t))" (t: time).

Here, as an example, the measurement position is shown as two-dimensional position information excluding the altitude direction (height direction) (a bird's-eye view of the measurement position). Note that the same can be applied to the case of three-dimensional position information.
Moreover, the measurement positions shown in the figures are merely for the convenience of explanation and are not necessarily accurate.

From the viewpoint of protecting personal information in ETC 2.0 as mentioned above, the processing unit 110 excludes from the processing target, for example, driving information within a range of "500 m" from the start point of driving and driving information within a range of "500 m" from the end point of driving, among the cigarette lighter type device measurement positions. Alternatively, if driving information within these ranges is included in the measurement data, it is deleted.

Next, the processing unit 110 selects the first cigarette lighter type device measurement position D1 (p(1)), for example, outside a range of 500 m from the starting point of the journey, and sets this as the first estimated ETC 2.0 measurement position D2 (p(1)). In addition, for example, information such as the speed, acceleration, and angular velocity of the cigarette lighter type device measurement data corresponding to that time is obtained.

Next, the processing unit 110 selects the cigarette lighter device measurement position D1 (p(q)) that is the first to deviate on the time axis from a circle with a radius of 200 m based on the unit distance (hereinafter, for convenience, referred to as the "unit circle") centered on the cigarette lighter device measurement position P11, and sets this as the second estimated ETC2.0 measurement position D2 (p(2)). In addition, for example, information such as the speed, acceleration, and angular velocity of the cigarette lighter device measurement data corresponding to that time is obtained.

Thereafter, the processing unit 110 repeats this process, and by selecting the cigarette lighter type device measurement positions [D1(p(1)), D1(p(q)), D1(p(r)), D1(p(s)), ...], the estimated ETC 2.0 measurement positions [D2(p(1)), D2(p(2)), D2(p(3)), D2(p(4)), ...] are obtained. Information such as speed, acceleration, and angular velocity can also be obtained in a similar manner.
Then, data in which this driving information is associated with time can be used as estimated ETC 2.0 measurement data.

In this way, basically, based on the ETC 2.0 specifications, it is possible to obtain driving information that is used as estimated ETC 2.0 measurement data from cigarette lighter type device measurement data based on unit distance.

Note that, although an example has been shown in which the processing unit 110 processes the unit distance as "200 m," the processing may be performed with the unit distance set to "100 m" or the like as described above, depending on the ETC 2.0 specifications.

Furthermore, the processing unit 110 may perform processing by changing the unit distance depending on the road type and speed range. For example, if the road type is "expressway", the unit distance may be longer than that of the road type "general road" since there is a possibility that the driving will be relatively stable despite the high speed driving. Conversely, the unit distance may be shorter than that of the road type "general road" in consideration of the risk of high speed driving.
Similarly, for speed zones, the unit distance may be set to be longer (or shorter) as the speed zone increases.
The road type and speed range can be specified, for example, from road type information.

Further, in this embodiment, the estimated ETC 2.0 measurement data is generated by selecting the driving information included in the cigarette lighter type device measurement data based on the distance (unit distance), but this is not limited to this.
Alternatively or in addition, the estimated ETC 2.0 measurement data may be generated by selecting the driving information included in the cigarette lighter type device measurement data based on the time. Specifically, for example, the time required to travel a unit distance may be calculated based on the speed information included in the cigarette lighter type device measurement data, and the estimated ETC 2.0 measurement data may be generated by selecting the driving information at the corresponding time based on the calculated time.
Similarly, estimated ETC 2.0 measurement data may be generated by assuming that the vehicle travels at a constant speed (for example, 60 km/h) and calculating the time required to travel a unit distance.

Here, as described above, ETC 2.0 has a specification that, for example, when the vehicle's heading direction changes by 45 degrees or more, the driving information is recorded. Therefore, at least one of the following conditions may be additionally applied to acquire the driving information.
However, it is not essential that these conditions apply.

FIG. 7A shows, as condition A, a condition related to a change in the traveling direction (traveling orientation) of the vehicle.
In this example, if the cigarette lighter device measurement position D1(p(u)) has changed by more than a threshold angle (or more than the threshold angle) relative to the most recent cigarette lighter device measurement position or the average position of multiple most recent cigarette lighter device measurement positions, the cigarette lighter device measurement position D1(p(u)) is selected as the estimated ETC 2.0 measurement position. This corresponds to a case where the vehicle's traveling direction has changed significantly. The threshold angle can be set to a value such as "45 degrees."
That is, a measurement position in the vicinity of a position where a change in the traveling direction has occurred that is equal to or exceeds a threshold angle may be appropriately selected.

Note that, although an example has been shown in which the processing unit 110 performs processing with the threshold angle set to "45 degrees," processing may also be performed with the threshold angle set to the aforementioned "22.5 degrees" or the like, depending on the ETC 2.0 specifications.

FIG. 7B shows a condition B based on the travel distance of the vehicle.
In this example, if the cigarette lighter device measurement position D1(p(n)) that first deviates from the unit circle is at least a threshold distance (or more than the threshold distance) away from the cigarette lighter device measurement position D1(p(m)) previously selected as the estimated ETC 2.0 measurement position, the cigarette lighter device measurement position D1(p(n)) is not selected. The threshold distance can be set to a value that is difficult to imagine as the distance the vehicle will travel from the previously selected point, based on the vehicle speed calculated based on the cigarette lighter device measurement data or a predetermined vehicle speed (e.g., 60 km/h).

In other words, a distance margin can be set, and if the distance falls within the distance margin, the driving information can be selected, but if the distance does not fall within the distance margin, the driving information cannot be selected.

In this case, for example, the measurement position D1(p(n-1)) immediately preceding the measurement position D1(p(n)) may be selected as the estimated ETC2.0 measurement position.
Also, for example, the measurement position D1(p(n+1)) next to the measurement position D1(p(n)) may be selected as the estimated ETC2.0 measurement position.
That is, a measurement position near an end of a unit circle may be appropriately selected. Also, the range used as the reference for selection is not limited to the unit circle, and may be any range based on the position of the reference data (e.g., an ellipse, a rectangle, etc.).

For example, the average (which may be an arithmetic average or a weighted average) of the measurement position D1(p(n)) and the measurement position D1(p(m)) may be calculated, and the calculated position may be used as the estimated ETC2.0 measurement position. In addition, it is not limited to calculating the average, and for example, the position obtained by dividing the distance between the measurement position D1(p(n)) and the measurement position D1(p(m)) at any ratio (for example, 2:1) may be used as the estimated ETC2.0 measurement position, and there is no particular limitation on the content as long as the value between the measurement position D1(p(n)) and the measurement position D1(p(m)) is obtained.

Furthermore, with respect to the conditions in FIG. 7 (2), similar processing may be performed by providing a time margin instead of a distance margin.

Also, as mentioned above, in the ETC 2.0 specifications, a peak value is recorded when any of the longitudinal acceleration, lateral acceleration, and yaw angular velocity exceeds a threshold value. Conditions for reproducing this can be set, and driving information can be selected based on the acceleration information and angular velocity information contained in the cigarette lighter device measurement data.

In addition, the above describes an example of a method for generating estimated ETC 2.0 measurement data based on location information, but instead of or in addition to this, estimated ETC 2.0 measurement data may be similarly generated based on, for example, speed information and time information.

Mathematically, it can be said that the estimated ETC 2.0 measurement data generated as described above is in a relationship that is encompassed by the cigarette lighter type device measurement data.
By generating a driving score model using the estimated ETC 2.0 measurement data generated in this manner, a driving score model with high validity can be generated.

Returning to FIG. 5, the first driving score model processing unit 113, for example, calculates the driving score by inputting the cigarette lighter type device measurement data into the first driving score model.

The second driving score model generation unit 115 generates a second driving score model, for example, by performing learning (e.g., supervised learning) on the first driving score model using, as a learning data set, the estimated ETC 2.0 measurement data estimated by the ETC 2.0 measurement data estimation unit 111 and the driving score (teacher driving score) calculated by the first driving score model processing unit 113. The model for which learning is complete becomes the learned second driving score model.

(2) At the Time of Inference As shown in FIG. 5 (2), the processing unit 110 of the server 100 has, for example, a learned second driving score model processing unit 117 as a functional unit.
The learned second driving score model processing unit 117 may be a type of the learned second model processing unit 17 in FIG. 2 (2).

The learned second driving score model processing unit 117 calculates a driving score, for example, by inputting the ETC 2.0 measurement data stored in the ECT 2.0 measurement database 194 into the learned second driving score model.
In other words, the learned second driving score model processing unit 117 takes as input ETC 2.0 measurement data measured by an ETC 2.0 onboard unit actually installed in the vehicle, infers a driving score using the learned second driving score model, and outputs the inference result, which is the driving score inference result.

The first driving score model may be, for example, an existing driving score model.
Here, the cigarette lighter type device measurement data input to the first driving score model processing unit 113 includes information obtained by measuring driving information such as the vehicle position, acceleration (speed when integrated over time), and angular velocity (orientation when integrated over time). Therefore, the cigarette lighter type device measurement data reflects the driving conditions of the driver, and the driving score calculated by the first driving score model processing unit 113 may be a value that reflects the driving conditions of the driver. This can be used as teacher data for learning.

As described in the principle of the embodiment, for example, estimated ETC 2.0 measurement data may be used as input data to the first driving score model, and some previously held data corresponding to first device measurement data measured by various first devices including cigarette lighter type devices, which are accumulated in the past, may be used as training data to generate the second driving score model by machine learning or the like.
The teacher data in this case may be, by way of example only, data relating to vehicle accidents (data on the presence or absence of accidents, the number of accidents (frequency, rate), etc.).

<Processing>
FIG. 8 is a flowchart showing an example of the flow of the driving score model process executed by the processing unit 110 of the server 100 in this embodiment.
First, the processing unit 110 of the server 100 performs a learning dataset generation process.
In the learning data set generation process, the processing unit 110 performs the process of loop B on the cigarette lighter type device measurement data of each processing target (S1 to S7).
The cigarette lighter type device measurement data to be processed may be, for example, cigarette lighter type device measurement data stored in the cigarette lighter type device measurement database 192 that is measured under similar conditions (for example, the same measurement period, the same road type, the same area, etc.). However, the present invention is not limited to this.

In the process of loop B, the processing unit 110 generates estimated ETC 2.0 measurement data based on the cigarette lighter type device measurement data (S3). Then, the processing unit 110 associates the generated estimated ETC 2.0 measurement data with the cigarette lighter type device measurement data (e.g., associates the generated estimated ETC 2.0 measurement data with the identification information (e.g., data number) of the cigarette lighter type device measurement data) and stores the data in the estimated ETC 2.0 measurement database 194.
The processing unit 110 also calculates a teacher driving score based on the cigarette lighter device measurement data and the first driving score model (S5). The processing unit 110 then stores the calculated teacher driving score in the storage unit 190 in association with the cigarette lighter device measurement data.
Then, the processing unit 110 moves on to processing the next cigarette lighter type device measurement data.

When steps S3 and S5 have been performed for all cigarette lighter device measurement data to be processed, the processing unit 110 ends the processing of loop B (S7).

Then, the processing unit 110 performs a learning process (S9). Specifically, a model is trained based on each estimated ETC 2.0 measurement data stored in the estimated ETC 2.0 measurement database 194 and the corresponding teacher driving score (having the same identification information as the cigarette lighter type device measurement data) to generate a trained second driving score model.

Next, the processing unit 110 performs an inference process (S11). Specifically, the processing unit 110 calculates a second driving score based on the learned second driving score model generated in S9 and each of the ETC 2.0 measurement data of the processing target (for example, each of the ETC 2.0 measurement data stored in the ETC 2.0 measurement data). Then, the calculation result is set as the various quantity inference result.
Then, the processing unit 110 ends the process.

It is to be noted that step S11 may be omitted, and the process may be limited to generating the second driving score model.
Also, the process may be limited to step S3, and may be limited to generating estimated ETC 2.0 measurement data based on the cigarette lighter type device measurement data.

Moreover, the processing unit 110 may cause the display unit 130 to display the calculated value of the second driving score (inference result).
Further, instead of or in addition to displaying the calculated value of the second driving score itself on the display unit 130, the processing unit 110 may display information such as the rank of the driver (information having an order relationship) classified based on the calculated value of the second driving score on the display unit 130. Specifically, for example, the following rank information may be displayed on the display unit 130.
- From highest to lowest rank: 5 stars, 4 stars, 3 stars, 2 stars, 1 star
・The rankings are listed in descending order: ◎, 〇, △, ×
・The rankings are "A (A rating), B (B rating), C (C rating)" in descending order.
・The highest rank is "Gold, Silver, Bronze"

In addition, for example, when an insurance company that has obtained this information from server 100 presents the calculated insurance premium to a customer, the insurance company may also present information on the customer's second driving score value and rank.

In addition, the processing unit 110 may estimate, based on the cigarette lighter device measurement data, the amount of G-forces being applied in which direction due to the driver's acceleration/deceleration and left/right handling operations, and may display this in a graphical form on the display unit 130.

Effects of the First Embodiment
In this embodiment, as described above, the estimated ETC 2.0 measurement data is generated by selecting driving information from the cigarette lighter type device measurement data based on, for example, unit distance or unit time.
In this case, it is considered that information such as the sudden deceleration or sudden acceleration described above is not necessarily accurately reflected in the estimated ETC 2.0 measurement data (so-called near misses, etc. are not necessarily reflected).
However, the inventors of the present application have discovered that by using the estimated ETC 2.0 measurement data generated as described above, it is possible to obtain estimated ETC 2.0 measurement data that reflects the driver's general driving tendencies (e.g., tendency to drive safely, tendency to drive dangerously, etc.), and as a result, it is possible to generate a driving score model with high validity.

In this embodiment, the server 100 (an example of an information processing device) is equipped with a processing unit 110 that generates estimated ETC 2.0 measurement data, which is estimated data of ETC 2.0 measurement data (an example of second measurement data) measured by an apparatus such as an ETC 2.0 vehicle-mounted unit (an example of a second apparatus different from the first apparatus), based on cigarette lighter type device measurement data (an example of first measurement data) including driving information measured by an apparatus such as a cigarette lighter type device (an example of a first apparatus).
This makes it possible to generate estimated data of second measurement data measured by a second device that measures information about a moving body and is different from the first device, based on first measurement data measured by a first device that measures information about a moving body.

In this case, the sampling unit for measuring data may be different between a device such as a cigarette lighter type device (an example of a first device) and an ETC 2.0 vehicle-mounted device (an example of a second device).
This makes it possible to generate estimated data of second measurement data measured by a second device that measures information about a moving body based on first measurement data in which information about the moving body is measured by a first device of two devices with different sampling units.

In this case, the sampling unit of a device such as a cigarette lighter type device (an example of a first device) may be time, and the sampling unit of an ETC 2.0 vehicle unit (an example of a second device) may be distance.
This makes it possible to generate estimated data of second measurement data measured by a second device whose sampling unit for measuring information about a moving body is distance, based on first measurement data in which information about a moving body is measured by a first device whose sampling unit is time.

In this case, a device such as a cigarette lighter type device (an example of a first device) may measure at a predetermined time interval, and an ETC 2.0 vehicle unit (an example of a second device) may measure at a predetermined distance interval.
This makes it possible to generate estimated data of second measurement data measured by a second device that measures information about a moving body at a predetermined distance interval based on first measurement data measured by a first device that measures information about a moving body at a predetermined time interval.

In addition, a device such as a cigarette lighter type device (an example of a first device) may measure at a predetermined time interval, and an ETC 2.0 vehicle-mounted device (an example of a second device) may measure at a predetermined distance interval, and the processing unit 110 of the server 100 may generate estimated ETC 2.0 measurement data (an example of estimated data) based on the cigarette lighter type device measurement data (an example of the first measurement data), the predetermined distance intervals measured by the ETC 2.0 vehicle-mounted device, and predetermined conditions.
This makes it possible to appropriately generate the estimated data based on the first measurement data, the predetermined distance interval, and the predetermined conditions.

In this case, the predetermined condition may include at least a condition regarding a change in the direction of the moving object based on the position included in the cigarette lighter type device measurement data (an example of the first measurement data).
This makes it possible to generate estimated data that takes into account (reflects) changes in the direction of the moving object based on the position included in the first measurement data.

Furthermore, the processing unit 110 of the server 100 may generate a model for calculating various quantities based on estimated ETC 2.0 measurement data (an example of estimated data).
This makes it possible to appropriately generate a model for calculating various quantities based on the estimated data.

In this case, the processing unit 110 of the server 100 may generate a model such as a driving score model by machine learning using at least estimated ETC 2.0 measurement data (an example of estimated data) as learning data.
This makes it possible to generate a model capable of appropriately calculating various quantities based on the estimated data.

In this case, the processing unit 110 of the server 100 may calculate various quantities such as the driving score based on the ETC 2.0 measurement data and a model such as the generated driving score model.
This makes it possible to appropriately calculate various quantities based on the second measurement data and the generated model.

In this case, the cigarette lighter type device (an example of a first device) and the ETC 2.0 vehicle unit (an example of a second device) measure driving information, the various quantities are driving scores related to insurance of the mobile body whose driving information is measured, and the model may be a second driving score model for calculating a driving score based on the ETC 2.0 measurement data.
This makes it possible to generate a model for calculating the driving score based on the second measurement data.

In this case, the processing unit 110 of the server 100 may generate a second driving score model by machine learning using estimated ETC 2.0 measurement data (an example of estimated data) and a driving score (an example of information based on the first measurement data) based on cigarette lighter type device measurement data (an example of the first measurement data) as learning data.
This makes it possible to generate a highly valid model for calculating the driving score based on the second measurement data.

<Modification (1)>
In the above embodiment, the first and second devices may be reversed to perform the same processing.

Specifically, for example,
First device: ETC 2.0 vehicle-mounted device Second device: cigarette lighter type device It is also possible to estimate cigarette lighter type device measurement data based on ETC 2.0 measurement data.

In other words, including the above embodiment, the sampling unit of one of the first device and the second device may be time, and the sampling unit of the other may be distance. Also, one of the first device and the second device may measure at a predetermined time interval, and the other may measure at a predetermined distance interval.

<Modification (2)>
In the above embodiment, the estimated ETC 2.0 measurement data is generated based on the driving information of the discrete cigarette lighter type device measurement data in a time series, but the present invention is not limited to this.
A continuous function that approximates the driving information included in the cigarette lighter type device measurement data may be calculated, and estimated ETC 2.0 measurement data may be generated based on the calculated function.

For example, based on the two-dimensional time-series position information contained in the cigarette lighter device measurement data, a function of two-dimensional position (X, Y) "(X, Y) = f(t)" with time "t" as a parameter is calculated.
For example, the time required to travel a unit distance may be calculated from the speed information included in the cigarette lighter type device measurement data, and based on the calculated time, the position information for the corresponding time "t" may be calculated from the function "(X, Y) = f(t)," thereby generating estimated ETC 2.0 measurement data.
The same can be applied to the case where the information is three-dimensional position information.

In addition, by assuming that the vehicle travels at a constant speed (e.g., 60 km/h) and calculating the time required to travel a unit distance, estimated ETC 2.0 measurement data can be generated similarly using the function "(X, Y) = f(t)."

<Modification (3)>
In the above embodiment, the second driving score model generation unit 115 may generate the driving score model by learning using at least the estimated ETC 2.0 measurement data and the ETC 2.0 measurement data as learning data.

Specifically, for example, a driving score model may be generated by performing supervised learning in which the teacher driving score and ETC 2.0 measurement data are used as teacher data, and the estimated ETC 2.0 measurement data is also used as a learning dataset.

In this modified example, the processing unit 110 of the server 100 generates a model by machine learning using at least estimated ETC 2.0 measurement data (e.g., an example of estimated data) and ETC 2.0 measurement data (e.g., an example of second measurement data) as learning data.
This makes it possible to generate a more accurate model.

<Modification (4)>
The validity of the estimation of the ETC 2.0 measurement data described in the above embodiment may be verified.
Specifically, for example, under similar conditions, ETC 2.0 measurement data obtained by having a driver drive a vehicle equipped with an ETC 2.0 on-board unit and estimated ETC 2.0 measurement data generated as in the above embodiment are obtained. Then, the processing unit 110 of the server 100 calculates a value (correlation value) indicating the correlation between these measurement data, and when it is determined that the correlation value is equal to or greater than a predetermined threshold value (or exceeds the threshold value) (when it is determined that there is a correlation between the two), it may be determined that the estimation of the ETC 2.0 measurement data is valid.
In this case, by carrying out the above process on a certain amount of data, the validity can be verified more appropriately.

In addition, the validity of the driving score model generated as in the above embodiment may be verified.
Specifically, as described above, for example, under similar conditions, a driver drives a vehicle equipped with an ETC 2.0 on-board unit to obtain ETC 2.0 measurement data and estimated ETC 2.0 measurement data generated as in the above embodiment. Then, the processing unit 110 of the server 100 compares the driving score obtained by inputting the ETC 2.0 measurement data into the driving score model with the driving score obtained by inputting the estimated ETC 2.0 measurement data into the driving score model, and if the difference between these driving scores is less than a predetermined threshold (or equal to or less than the threshold), the driving score model may be determined to be valid.
In this case, by carrying out the above process on a certain amount of data, the validity can be verified more appropriately.

<Modification (5)>
Although a user may have signed a contract for automobile insurance with an insurance company, there may be cases where, during the contract period, the device installed in the automobile is changed (switched) from a first device (e.g., a cigarette lighter type device) to a second device (e.g., an ETC 2.0 vehicle unit).

Therefore, for example, when the device (used to measure measurement data) installed in the vehicle is changed from a first device to a second device during the contract period of the automobile insurance (the period from the insurance start date to the end date (maturity date)), the processing unit 110 of the server 100 may calculate the driving score in the case of renewing the automobile insurance based on estimated data based on the measurement data of the first device during at least the first period from the start of the contract period to the device change (e.g., the first device measurement data measured from the insurance start date to the device change date) and the measurement data of the second device during the second period from the device change to the end of the contract period (e.g., the second device measurement data measured from the device change date to the driving score calculation date (renewal procedure date)).

In this case, the driving score may be calculated by, for example, any of the following methods.
- The first score value calculated by inputting estimated data (e.g., estimated ETC 2.0 measurement data) into the learned second driving score model and the second score value calculated by inputting second device measurement data (e.g., ETC 2.0 measurement data) into the learned second driving score model are calculated by averaging the first score value and the second score value calculated by inputting second device measurement data (e.g., ETC 2.0 measurement data) into the learned second driving score model.
- Data obtained by integrating estimated data (e.g., estimated ETC 2.0 measurement data) and second device measurement data (e.g., ETC 2.0 measurement data) is input into a learned second driving score model and calculated.

In addition, instead of the estimated data based on the first device measurement data in the first period from the start of the contract period to the device change, the first device measurement data in the first period from the start of the contract period to the device change may be used. In this case, the driving score may be calculated, for example, by the following method.
- Calculated by averaging a first score value calculated by inputting a first device measurement data (e.g., cigarette lighter type device measurement data) into a first driving score model and a second score value calculated by inputting a second device measurement data (e.g., ETC 2.0 measurement data) into a learned second driving score model.

Second Example
In the first embodiment, an example of an insurance product such as telematics insurance has been described. However, when developing an ETC 2.0-based service including such a telematics insurance product service, it is necessary to evaluate how many people can use the service, the size of the market, whether the service can be provided fairly throughout the country, etc. The second embodiment is an embodiment related to this evaluation.
The contents described in the second embodiment are applicable to the other embodiments and other modified examples.

<Data structure>
FIG. 9 is a diagram showing an example of information stored in the storage unit 190 of the server 100 in this embodiment.
The memory unit 190 stores, for example, a first ETC 2.0 service evaluation support processing program 197 that is read by the processing unit 110 and executed as an ETC 2.0 service evaluation support processing, the aforementioned cigarette lighter type device measurement database 192, a roadside unit information database 198, and a created histogram database 199.

As described above, the cigarette lighter type device measurement database 192 can store the measurement data of multiple drivers for each measurement period. The measurement period can be, for example, a period in months (one month, two months, ..., six months, etc.), or can be a period in other units. The data can also be stored by classifying it, for example, by road type or by region.

The roadside unit information database 198 is a database that stores information including, for example, location information of roadside units nationwide.

The created histogram database 199 is a database that stores histogram data created by the processing unit 110 for purposes such as evaluation of ETC 2.0-based services, which will be described later.

The aforementioned estimated ETC 2.0 measurement database 194, etc. may be stored in the memory unit 190.

<Specifications of ETC 2.0 in-vehicle unit>
In ETC 2.0, as described above, for example, a roadside unit collects probe information including driving information from an ETC 2.0 vehicle-mounted unit.
On the other hand, an ETC 2.0 on-board unit may only be able to record driving information for a specified distance, such as a maximum of 80 km, and any driving information exceeding the specified distance may be discarded without being recorded. For this reason, if a vehicle equipped with an ETC 2.0 on-board unit does not pass near a roadside unit for a long time, the recorded driving information may not be collected by the roadside unit.

The ETC 2.0 specifications may be updated in the future.
For example, the aforementioned unit distance (e.g., 200 m) may be changed, and in some cases, it may become necessary to increase the memory of the ETC 2.0 vehicle-mounted unit.
However, the technique of the present invention can also accommodate such changes in the ETC 2.0 specifications, and it is only necessary to adjust the values of various parameters.

<Processing>
FIG. 10 is a diagram showing an example of the flow of the first ETC2.0 service evaluation support process performed by the processing unit 110 of the server 100 in this embodiment.
The processing unit 110 performs the following processing on each piece of cigarette lighter type device measurement data included in the cigarette lighter type device measurement database 192, for example.

First, the processing unit 110 performs the following process on the driving information included in the cigarette lighter type device measurement data, for example, in chronological order starting from the oldest driving information.
The processing unit 110 determines whether or not the measured position of the vehicle included in the travel information is included in the area of the roadside unit (new roadside unit) (S21).

Here, the area near the roadside unit is referred to as a “roadside unit area.” The roadside unit area can be set, for example, as an area within a radius of 5 m or within a radius of 10 m from the position of the roadside unit.
Specifically, based on the position information of the roadside device stored in the roadside device information database 198, the processing unit 110 determines whether or not the measured position of the vehicle included in the driving information is included in the roadside device area of the roadside device.

If it is determined that the measured position of the vehicle is included (S21: YES), the processing unit 110 determines whether the measured position of the vehicle is more than 80 km from the position of the previous roadside unit (S23).

If it is determined that the limit has been exceeded (S23), the processing unit 110 sets the transmission flag (collection flag) to "ON" for the set of driving information included in the cigarette lighter type device measurement data that is within "80 km" going back from the location of the roadside unit (S25).

The transmission flag may be a flag that simulates the transmission of travel information to a roadside unit when a vehicle equipped with an ETC 2.0 onboard unit passes near the roadside unit.
In this embodiment, in order to evaluate ETC 2.0-based services, data measured by a cigarette lighter type device (already acquired cigarette lighter type device measurement data) is used to virtually reproduce (simulate) the transmission of driving information (collection of driving information by a roadside unit) when a vehicle equipped with an ETC 2.0 onboard unit passes near a roadside unit.

For example, if the measured position of the vehicle included in the driving information is 100 km away from the position of the previous roadside unit, then according to the ETC2.0 specifications mentioned above, the driving information exceeding 20 km from the position of the previous roadside unit will be discarded from the ETC2.0 in-vehicle unit. To reproduce this, in step S25, the transmission flag is set to "ON" for the set of driving information within 80 km of the roadside unit's position, excluding the portion exceeding 80 km.

On the other hand, if it is determined that the distance from the previous roadside unit position is not more than 80 km (S23: NO), the processing unit 110 sets the transmission flag to "ON" for the set of driving information from the previous roadside unit position among the driving information included in the cigarette lighter type device measurement data (S27).

After S25 or S27, the processing section 110 judges whether or not to end the processing (S29), and if it is judged that the processing should be continued (S29: NO), the processing section 110 returns to S1.
On the other hand, if it is determined that the process is to be ended (S29: YES), the processing unit 110 ends the first ETC2.0 service evaluation support process.

<Setting the measurement period>
FIG. 11 is a diagram showing an example of a histogram for ETC2.0 service evaluation generated by the processing unit 110 of the server 100 based on the processing result of the first ETC2.0 service evaluation support process in this embodiment.

Based on the results of the above processing, the processing unit 110 creates a histogram for each measurement period of the cigarette lighter type device measurement data, with the horizontal axis (class) representing the estimated collection time (h) at the roadside unit and the vertical axis (frequency) representing the number of vehicles (units).
The estimated collection time at the roadside device can be derived from, for example, the travel information (set of travel information) with the transmission flag set to "ON," and therefore can be said to be essentially synonymous with the estimated number of pieces of data that could be collected by the roadside device. For this reason, the horizontal axis may represent the estimated number of pieces of collected data.
The vertical axis may represent the number of runs (times).

This figure shows an example of a histogram for a measurement period of "1 month." As an example, the horizontal axis shows sections divided into "1-hour" intervals of the expected collection time at the roadside unit as classes. The vertical axis shows the number of vehicles contained in each section as a frequency. The total number of vehicles is the sum of the numbers of vehicles in all sections. Note that the values on the horizontal and vertical axes are merely examples and are not limited to these.

Figure 12 shows an example of a histogram created in a similar manner for the measurement period "2 months." The chart can be interpreted in the same way as Figure 11.

For example, the processing unit 110 sets the threshold value of the estimated collection time to "3 hours" and calculates the ratio of the number of vehicles to the total number of vehicles whose estimated collection time exceeds the threshold value of "3 hours" (hereinafter referred to as the "vehicle number ratio").
Suppose that the vehicle number ratio calculated based on the histogram for the measurement period "1 month" in FIG. 11 is, for example, "60%," and that the vehicle number ratio calculated based on the histogram for the measurement period "2 months" in FIG. 12 is, for example, "80%."

For example, if the target value for the vehicle number ratio is set to "75%" when the threshold value for the expected collection time is "3 hours," the vehicle number ratio does not reach the target value during the measurement period of "1 month" in Figure 11, but reaches the target value during the measurement period of "2 months" in Figure 12.
Therefore, in this example, it can be determined that there is no problem in reproducing the ETC 2.0 measurement data if the cigarette lighter type device measurement data, which has a measurement period of "two months", is used.

In actual processing, the processing unit 110 of the server 100 creates a histogram by updating the measurement periods, for example, in ascending order of measurement periods, until the vehicle number ratio reaches the target value. Then, when the vehicle number ratio reaches the target value, the measurement period is set as the measurement period to be operated.
In this case, for example, among the sets of cigarette lighter type device measurement data for each measurement period stored in the cigarette lighter type device measurement database 192 of the memory unit 190, only the set of cigarette lighter type device measurement data for the set measurement period may be retained, and the sets of cigarette lighter type device measurement data for the other measurement periods may be deleted.

Then, the processing unit 110 can generate estimated ETC 2.0 measurement data described in the first embodiment using, for example, a set of cigarette lighter type device measurement data for the set measurement period, and generate a driving score model.
In this way, it is possible to create telematics insurance products with guaranteed reliability.

The following processing may be performed.
For example, when the processing unit 110 sets a period of "two months" as the measurement period of the cigarette lighter type device measurement data, the processing unit 110 generates a driving score model using the cigarette lighter type device measurement data of the "two months" measurement period. Similarly, the processing unit 110 generates a driving score model using the estimated ETC 2.0 measurement data of the "two months" measurement period.
Then, the processing unit 110 determines whether similar results are obtained when specified measurement data is input into these generated driving score models (for example, whether the difference between the models is below a specified threshold (or less than the threshold)), and if similar results are obtained, the measurement period for the ETC 2.0 measurement data used to calculate the driving score (used to evaluate driving) is set to "two months."

On the other hand, if similar results are not obtained, the measurement period of the ETC 2.0 measurement data used to calculate the driving score (used to evaluate driving) may not be "two months," so the processing unit 110 may set the measurement period of the ETC 2.0 measurement data used to calculate the driving score to a measurement period longer than "two months" (for example, any period between "three months" and "six months").

This is because the method of the present invention generates estimated ETC 2.0 measurement data by thinning out (resampling) the driving information of the cigarette lighter type device measurement data, and therefore it may not necessarily be appropriate to directly apply the measurement period of the cigarette lighter type device measurement data to the ETC 2.0 measurement data.
Therefore, as described above, the two driving score models may be compared and verified, and the measurement period of the ETC 2.0 measurement data used to calculate the driving score may be set (determined) based on the results of the comparison and verification.

<Insurance product business evaluation>
The histogram created by the server 100 as described above can also be used to evaluate the commercial success of insurance products based on ETC 2.0.

When considering creating an insurance product based on ETC 2.0 driving information, it is necessary to have the user actually drive a vehicle equipped with an ETC 2.0 onboard unit and collect the measurement data. As mentioned above, this has the problem of costs, and if it turns out that the ETC 2.0 system is not capable of collecting sufficient driving information, the insurance product may not be viable as a business in the first place. In other words, as mentioned above, it is expected that data loss will occur if a vehicle equipped with an ETC 2.0 onboard unit drives, for example, more than 80 km without passing near a roadside unit. This will not be known until a demonstration experiment is conducted using an actual vehicle equipped with an ETC 2.0 onboard unit.

Therefore, for example, an estimated collection time is calculated for each vehicle based on data collected by other means than ETC 2.0 (e.g., cigarette lighter type device measurement data) and location information of roadside units included in the area covered by the telematics insurance (e.g., the Tokyo metropolitan area, etc.). Then, the amount of driving information virtually collected by the roadside unit (or, conversely, how much driving information is discarded) for all target vehicles can be evaluated using the histograms described above, etc.

Specifically, the processing unit 110 of the server 100 creates a histogram, for example, of how much driving information is collected per month based on the cigarette lighter type device measurement data during a specified measurement period, in a manner similar to that shown in Figures 11 and 12.
The processing unit 110 of the server 100 can then set a threshold value (e.g., 7 hours or 10 hours) for the expected collection time at the roadside unit, and if the number of vehicles (or the proportion of vehicles) whose expected collection time is equal to or greater than the threshold value (or exceeds the threshold value) is equal to or greater than the set value (or exceeds the set value), it can determine that the business is viable as a general insurance product.

In this case, for example, a period of "one month" may be used as the predetermined measurement period, or a period of "two months" or "three months" may be used as the predetermined measurement period. In this case, the estimated collection time in the predetermined measurement period may be used as is, or the estimated collection time per month may be used.
Alternatively, a histogram may be created from cigarette lighter device measurement data for a number of different measurement periods, and the estimated collection time for each measurement period may be averaged to use the calculated time.

Moreover, instead of the estimated collection time, the aforementioned estimated number of pieces of collected data may be used.
Similar processing may also be performed based on the number of data (estimated number of discarded data) or time (estimated discard time) indicating how much driving information was virtually discarded without being collected by the roadside device. This can also be considered as information (data loss ratio information (data loss degree information)) indicating the ratio (degree) of data loss in estimated ETC 2.0 measurement data (pseudo ETC 2.0 measurement data generated based on cigarette lighter type device measurement data) generated based on cigarette lighter type device measurement data.

In addition, determining the business success of the above insurance product can also be said to be determining the validity of the estimated ETC 2.0 measurement data (pseudo ETC 2.0 measurement data generated based on cigarette lighter type device measurement data) generated based on the cigarette lighter type device measurement data, based on the above data loss rate information.

In this case, the server 100 may or may not actually generate estimated ETC 2.0 measurement data. In other words, the server 100 may generate estimated ETC 2.0 measurement data based on cigarette lighter type device measurement data and then determine the validity of the generated estimated ETC 2.0 measurement data using the above method, or, in the case where estimated ETC 2.0 measurement data is generated based on cigarette lighter type device measurement data without generating estimated ETC 2.0 measurement data, the validity of the generated estimated ETC 2.0 measurement data may be determined using the above method.

If it is determined that the business is viable as a general insurance product, the processing unit 110 of the server 100 may set a measurement period based on the method described above in <Setting the measurement period>, and then use the set of cigarette lighter type device measurement data for the set measurement period to generate estimated ETC 2.0 measurement data as described in the first embodiment, and generate a driving score model.

Conversely, if the processing unit 110 of the server 100 determines that the insurance product is not viable as a business, it may take one of the following steps.
· Do not generate estimated ETC 2.0 measurement data · Generate estimated ETC 2.0 measurement data but do not use it

In addition, in the above-mentioned <Setting the measurement period> and <Evaluating the insurance product business>, the processing unit 110 of the server 100 may perform a similar evaluation using estimated ETC 2.0 measurement data instead of cigarette lighter type device measurement data.

<Effects of the Second Example>
In this embodiment, the server 100 determines whether or not to generate estimated ETC 2.0 measurement data based on the ETC 2.0 measurement data (an example of the second measurement data) transmitted from an ETC 2.0 vehicle-mounted device (an example of the second device) to a roadside unit (an example of the third device) installed at a predetermined location, and based on the cigarette lighter type device measurement data (an example of the first measurement data) and the location information of the roadside unit.
This makes it possible to appropriately determine whether or not to generate estimated data of the second measurement data, based on the first measurement data and position information of the third device installed at a predetermined position.

In addition, in this embodiment, the server 100 determines whether or not to use the generated estimated ETC 2.0 measurement data (an example of the second measurement data) that is transmitted from an ETC 2.0 vehicle-mounted device (an example of the second device) to a roadside unit (an example of the third device) installed at a predetermined location, based on the cigarette lighter type device measurement data (an example of the first measurement data) and the location information of the roadside unit.
This makes it possible to appropriately determine whether or not to use estimated data of the second measurement data, based on the first measurement data and position information of the third device installed at a predetermined position.

In addition, in this embodiment, the server 100 determines the measurement period of the cigarette lighter type device measurement data used to generate the ETC 2.0 measurement data, which is an example of the second measurement data, transmitted from an ETC 2.0 vehicle-mounted unit (an example of the second device) to a roadside unit (an example of the third device) installed at a predetermined location, based on the cigarette lighter type device measurement data (an example of the first measurement data) and the location information of the roadside unit.
This makes it possible to appropriately determine the measurement period of the first measurement data used to generate estimated data of the second measurement data, based on the first measurement data and position information of the third device installed at a predetermined position.

In addition, in this embodiment, the server 100 is equipped with a processing unit 110 that determines the measurement period of the cigarette lighter type device measurement data required for evaluating driving based on the ETC 2.0 measurement data measured by the ETC 2.0 vehicle-mounted unit, based on the cigarette lighter type device measurement data measured by the cigarette lighter type device.
This makes it possible to appropriately determine the measurement period of the first measurement data required for evaluating driving based on the second measurement data measured by a second device that measures information about a moving body and is different from the first device, based on the first measurement data measured by a first device that measures information about a moving body.

In this case, the server 100 determines the measurement period of ETC 2.0 measurement data required for evaluating driving based on ETC 2.0 measurement data, based on the cigarette lighter type device measurement data for the measurement period determined as described above.
This makes it possible to appropriately determine the measurement period of the second measurement data, which is necessary for evaluating driving based on the second measurement data, based on the first measurement data for the measurement period determined as described above.

The server 100 also generates pseudo ETC 2.0 measurement data for a predetermined period based on the cigarette lighter device measurement data, and acquires data loss rate information in the pseudo ETC 2.0 measurement data. The server 100 then determines the validity of the pseudo ETC 2.0 measurement data based on the acquired data loss rate information.
This makes it possible to appropriately determine the validity of the pseudo second measurement data for a predetermined period, which is generated based on the first measurement data, based on the data loss rate information.

<Modification>
In the above embodiment, the server 100 may determine whether or not a business for insurance products for leisure (leisure use) is viable, using a method similar to that described in the above embodiment, based on cigarette lighter type device data and estimated ETC 2.0 measurement data measured with a leisure destination as the destination (for example, when the vehicle is located within the leisure area for a predetermined period of time). If it is determined that the business is viable, the insurance company may provide a special insurance product for leisure.

In addition, in this case, as described in the above embodiment, if the server 100 determines that the insurance product is valid as a general insurance product based on an evaluation based on cigarette lighter type device measurement data and estimated ETC 2.0 measurement data in the metropolitan area, etc., the insurance company will uniformly provide the general insurance product that is determined to be valid, regardless of whether it is for leisure use or not.

In contrast, if the server 100 determines that the insurance product is not viable as a general insurance product based on the evaluation based on cigarette lighter type device measurement data and estimated ETC 2.0 measurement data in the metropolitan area and the like, the server 100 may determine whether or not a business for insurance products for leisure use is viable, as described above. If it is determined that the business is viable, the insurance company may provide a special insurance product for leisure use.

<Third Example>
In collecting driving information under a system based on ETC 2.0, it is expected that more driving information can be collected by newly installing a roadside unit (ITS spot) that acquires driving information from an ETC 2.0-compatible device installed in a vehicle. Therefore, a third embodiment will be described below in which a degree of collection of driving information collected under a system based on ETC 2.0 is measured in a case where a new roadside unit is installed in addition to the existing roadside units, and the feasibility of a service related to ETC 2.0 is determined based on the measured degree of collection.
The contents described in the third embodiment are applicable to the other embodiments and other modified examples.

<Data structure>
FIG. 13 is a diagram showing an example of information stored in the storage unit 190 of the server 100 in this embodiment.
The memory unit 190 stores, for example, a second ETC 2.0 service evaluation support processing program 201 that is read by the processing unit 110 and executed as an ETC 2.0 service evaluation support processing, map information 202, a roadside unit information database 198, a cigarette lighter type device measurement database 192, and a facility information database 203 where a new roadside unit can be installed.

The map information database 201 is a database that stores information including, for example, nationwide road information, topography, etc.

As mentioned above, the roadside unit information database 198 is a database that stores information including, for example, location information of roadside units nationwide.

As described above, the cigarette lighter type device measurement database 192 is a database that accumulates and stores cigarette lighter type device measurement data including driving information measured by a cigarette lighter type device mounted on a vehicle.
In this embodiment, the travel information includes at least location information.

In the database 203 of information on facilities where new roadside units can be installed, for example, location information of facilities where new roadside units can be installed is stored. Preferably, each piece of location information is linked to property information of the facility (such as the name of a chain store, e.g., shopping mall name, gas station name, convenience store name, etc.), and based on the linked property information, it is possible to extract a set of a certain business type (gas station, etc.) operated by the same company.

14 is a diagram showing an example of the distribution of already installed ETC 2.0 roadside units on a map, which is generated by referring to the information stored in the map information database 201 and the roadside unit database 198. In this embodiment, area information is also displayed for each roadside unit, which is a guide to the acquisition location range of the travel information that is likely to be collected. It is preferable that such area information is stored in advance, or is calculated based on the position information of each roadside unit and surrounding road information, etc., in the new roadside unit installation location derivation process described later. Details of the area information will be described later.
As shown in Fig. 14, the location of the roadside unit is shown by a diamond in a manner superimposed on the map information. As described above, when a vehicle equipped with a device compatible with ETC 2.0 is located nearby, the roadside unit acquires travel information stored in the device. Also, with reference to the area information, a circle is depicted for each roadside unit, with the location of the roadside unit as the center and a radius of a predetermined length r.
The predetermined length r, which is the radius of the reference range, may be 80 km, which is the maximum recordable amount of driving information in an ETC 2.0 on-board device, or may be a circle with a radius of less than 80 km from the roadside unit (for example, 50 km) in consideration of driving a distance of 80 km including curves, etc. Furthermore, it is also possible to show multiple reference ranges, such as showing two circles, one with a radius of 80 km and the other with a radius of less than 80 km. In addition, the reference range may be any range based on the roadside unit position, rather than a circle with the roadside unit position as the center. For example, it is also possible to show a range calculated to reflect the actual road distance in consideration of the curves of the road, etc., as a distance of 80 km from the roadside unit position.

Ultimately, if roadside units were installed everywhere, it would be possible to collect driving information without exception; however, due to the cost involved, the number of roadside units that can be installed is limited.
In addition, when installing new roadside units, it is possible to cooperate with private companies, for example, to have them install the units at private business facilities. For example, cooperation based on an agreement in which the private companies pay for the installation at their own business facilities, and in return, the companies pay a fee for the data provided by the installed roadside units.

Consider a case in which, based on such cooperation, new roadside units are installed in a number of business facilities operated by the private company.
Fig. 15 shows information on candidate facilities superimposed on the diagram shown in Fig. 14. Specifically, in Fig. 15, for the situation in Fig. 14, by referring to the roadside unit new installation facility information database 203, for example, shopping mall X, gas station Y, and convenience store Z are extracted and shown as candidate facilities. Of course, the types of candidate facilities are not limited to these, and any type can be set.

The example shown in Figure 16 shows a situation in which a new roadside unit is installed at gas station Y, one of these candidate facilities. As shown in Figure 16, hollow diamonds are placed at each gas station Y to indicate the location where a new roadside unit will be installed, and the approximate range for obtaining driving information based on the location of each gas station Y is shown by dotted lines.

For the situation shown in Figure 16, it is possible to virtually obtain the collection status of driving information that is assumed when providing services related to ETC 2.0 based on driving information collected using a method other than that compatible with ETC 2.0, and by evaluating this, it is possible to judge the feasibility, effectiveness, etc. of services related to ETC 2.0. These processes will be explained in detail below.

<Processing>
Below, we will explain the process of acquiring the degree of collection of driving information under assumed conditions when a new roadside unit is installed, and the process of evaluating the degree of collection to determine the feasibility of services related to ETC 2.0.
17 is a diagram showing an example of the flow of the second ETC2.0 service evaluation support process performed by the processing unit 110 of the server 100 in this embodiment. Note that the second ETC2.0 service evaluation support process is partially similar to the first ETC2.0 service evaluation support process, and therefore a description of such similar content will be omitted.
The processing unit 110 performs the following processing, for example, on each cigarette lighter type device measurement data included in the cigarette lighter type device measurement database 192, in the same manner as the first ETC2.0 service evaluation support processing.

First, the processing unit 110 performs the following processing on the driving information included in the cigarette lighter type device measurement data, for example, in chronological order starting from the oldest driving information (of course, it may start from the newest driving information, and the order does not matter).
The processing unit 110 judges whether any of the driving information for a predetermined distance (e.g., 80 km) after the relevant driving information is included in the roadside unit area of the existing roadside unit or the newly installed roadside unit (S31). That is, as described above, the ETC 2.0 vehicle-mounted unit may only be able to record driving information for a predetermined distance, such as a maximum of "80 km", and driving information exceeding the predetermined distance is discarded without being recorded. In this context, the processing unit 110 judges whether such driving information is ultimately acquired by the roadside unit or discarded without being recorded.

If it is determined that the roadside unit area is included (S31: YES), the processing unit 110 sets the transmission flag (collection flag) to "ON" for the driving information that is determined to be included in the roadside unit area (S32).

On the other hand, if it is determined that the vehicle will not pass through the roadside unit area (S31: NO), no flag is set in the driving information (the transmission flag is "OFF").

After S31 or S32, the processing unit 110 judges whether or not to end the process (S33). That is, it judges whether or not there is any driving information to be processed again, and if there is, the processing continues, and if there is no information, the processing ends. (S33: NO), and the processing returns to S31.
On the other hand, if it is determined that the process is to be ended (S33: YES), the processing unit 110 ends the second ETC2.0 service evaluation support process.

In this embodiment, the criterion for each piece of travel information is whether or not the travel information for a predetermined distance passes through the roadside unit area, but the criterion is not limited to this, and any criterion based on the position of the roadside unit can be applied. For example, the criterion for each piece of travel information may be whether it is included in an area that is an indication of the acquisition position range of the travel information described above. Also, for example, the criterion may be whether it is included in a range that reflects the actual road distance taking into account road curves, etc., as a range of a predetermined distance (e.g., 80 km) from the roadside unit position.

In the third embodiment, as in the second embodiment, information indicating the likely degree of collection (hereinafter referred to as "collection degree information") is calculated based on the processing results of the second ETC 2.0 service evaluation support process. For example, a histogram similar to that shown in Figures 11 and 12 in the second embodiment is created. Then, the processing unit 110, for example, sets an appropriate threshold value for the expected collection time (for example, "3 hours") and calculates the percentage of vehicles whose expected collection time is equal to or exceeds this threshold value.

The method of evaluating the collection degree information is not limited to the above-mentioned method based on the threshold of the expected collection time, and any method can be applied. For example, the collection degree information may be obtained for each region, and an evaluation may be performed as to whether the region with the collection degree equal to or greater than a predetermined threshold is equal to or greater than a predetermined value.
Furthermore, the collection degree information may be evaluated by visually displaying the positions where the travel information was acquired on map information, and the user may visually check the distribution of these positions. This allows the user to evaluate whether travel information is collected in a specific area, whether travel information is collected uniformly in a specific area (or the entire area), etc.

<Determination of the viability of a specified business>
Then, by using the collection degree information such as the histogram created by the server 100 as described above, it is possible to judge the feasibility of a given business based on ETC 2.0, as in the second embodiment.

In determining the viability of a given business, the following determination method can be applied, by way of example and not by way of limitation.
The processing unit 110 sets, for example, a period of "two months" as the measurement period for cigarette lighter type device measurement data, and if the collection level information calculated based on the measurement data from such measurement period satisfies some condition for realizing a specified business (for example, the threshold expected collection time is set to "3 hours", the proportion of the number of vehicles out of the total number of vehicles for which the expected collection time exceeds "3 hours" is "75%," and exceeds the threshold of "70%" for the proportion of the number of vehicles necessary to carry out the specified business), it determines that the specified business can be established.
On the other hand, if the conditions are not met, it is determined that it will be difficult to establish the specified business.

For example, by preparing multiple installation patterns for new roadside units (e.g., pattern 1: install at "gas station Y", pattern 2: install at "shopping mall X" and "gas station Y", pattern 3: install at "shopping mall X", "gas station Y" and "convenience store Z"), it is possible to determine whether a specified business can be conducted for each of them. In other words, for example, even if it is determined that it is difficult to establish a specified business in pattern 1, if it is determined that the specified business can be established in pattern 2 or pattern 3, a decision can be made to install a new roadside unit based on pattern 2 or pattern 3.

<Modification>
In the above embodiment, the collection degree information calculated under the assumed situation in which a roadside unit is newly installed in a first pattern, which is a combination of roadside units, may be compared with the collection degree information calculated under the assumed situation in which a roadside unit is newly installed in a second pattern, which is a combination of roadside units different from the first pattern, to obtain a difference therebetween. For example, in a situation in which a measurement period of "two months" is set, if the ratio of the number of vehicles with an assumed collection time of more than "3 hours" to the total number of vehicles in the former is "75%" and the ratio of the number of vehicles with an assumed collection time of more than "3 hours" to the total number of vehicles in the latter is "70%, the difference is 5 points. This difference of 5 points can be said to be the difference in effect when a roadside unit is newly installed in the first pattern and the second pattern.

Furthermore, in the above embodiment, the collection degree information calculated under the assumed situation in which a new roadside unit has been installed may be compared with the collection degree information under the situation in which a new roadside unit has not been installed, and the difference between them may be calculated. For example, in a situation in which a measurement period of "two months" is set, if the ratio of the number of vehicles with an assumed collection time of more than "3 hours" to the total number of vehicles in the former is "75%" and the ratio of the number of vehicles with an assumed collection time of more than "3 hours" to the total number of vehicles in the latter is "65%, " then it can be seen that the difference is 10 points. This difference of 10 points can be said to be the effect of installing a new roadside unit.

Furthermore, for example, a combination of new roadside units to be installed may be identified using an optimization method that identifies new roadside units from among candidate facilities where new roadside units can be installed, which will provide as high a collection rate as possible while installing as few new roadside units as possible. That is, based on the location information of existing roadside units, the location information of candidate facilities where new roadside units can be installed, and driving information, an optimization method, such as the steepest descent method as an example and not a limitation, is used to identify the optimal combination of candidate facilities where new roadside units can be installed.

Furthermore, for example, when selecting a new roadside unit to be processed, the new roadside units to be installed may be differentiated according to the type of business, region, etc., as described above. Specifically, for example, "in region A, limit to gas station Y and shopping mall X, and in region B, extract all candidates of gas station Y, shopping mall X, and convenience store Z." Furthermore, any one (or any multiple) may be specified as the target for installing a new roadside unit. Note that the calculation of the collection degree information may be performed for a combination pattern of multiple candidates, and a combination of new roadside unit candidates that provides the optimal collection degree information may be determined.

Furthermore, there may be cases where the location where a new roadside unit can be installed is almost the same as an existing roadside unit, and little additional effect can be expected from the installation. In consideration of such cases, the collection degree information may be calculated in a manner that does not include any new roadside unit. That is, based on the location information of existing roadside units, user input, etc., it is possible to specify what will not be included in the roadside unit to be newly installed (for example, a set of "gas stations Y" excluding some "gas stations Y" specified based on some criteria from the set of "gas stations Y"), and calculate the collection degree information in a manner that does not include such roadside units. In addition, the collection degree information may be calculated for multiple patterns, including patterns that do not include such specific roadside units, and a pattern that provides the optimal collection degree information may be obtained.

<Effects of the Third Embodiment>
In this embodiment, the server 100 calculates the collection degree of cigarette lighter type device measurement data based on the positions of existing roadside units and the positions of facilities where new roadside units can be installed, imitating the manner in which the measurement data is collected by each roadside unit based on a method compatible with ETC 2.0. Also, based on the collection degree, the feasibility of a predetermined service related to ETC 2.0 is determined. Also, the difference in the collection degree when a new roadside unit is installed in a different pattern, and the difference in the collection degree between when a new roadside unit is installed and when not, are calculated, and the effect is calculated. Furthermore, a combination of candidate facilities for installing a new roadside unit is calculated by an optimization method, taking into account the installation cost of the roadside unit and the level of the collection degree.
This makes it possible to perform evaluations of the degree of collection under the assumption that a new roadside unit will be installed, without having to install a new roadside unit.

<Fourth Example>
When collecting driving information under a system based on ETC 2.0, it is expected that more driving information can be collected by newly installing roadside units (ITS spots) that acquire driving information from devices compatible with ETC 2.0 installed in vehicles. Therefore, as a fourth embodiment, an example of deriving the installation positions of new roadside units that can effectively collect driving information under a system based on ETC 2.0 will be described.
The contents described in the fourth embodiment are applicable to the other embodiments and other modified examples.

<Data structure>
FIG. 18 is a diagram showing an example of information stored in the storage unit 190 of the server 100 in this embodiment.
The memory unit 190 stores, for example, a new roadside unit installation position derivation processing program 204 which is read by the processing unit 110 and executed as a process for deriving a new roadside unit installation position, a map information database 202, a roadside unit information database 198, a cigarette lighter type device measurement database 192, and a new roadside unit installation possible facility information database 203.
These data configurations are the same as those in the data configuration in the storage unit 190 in the third embodiment, except that the second ETC2.0 service evaluation support processing program 201 has been replaced by a new roadside unit installation position derivation processing program 204, and therefore detailed explanations are omitted.

FIG. 14 is a diagram showing an example of the distribution of already installed ETC 2.0 roadside units on a map, which is generated by referring to the information stored in the map information database 201 and the roadside unit database 198, as described above.
Ultimately, if roadside units were installed everywhere, it would be possible to collect driving information without exception, but this would be difficult in terms of cost, and the number of roadside units that can be installed is limited. Therefore, it is necessary to install roadside units in locations where collection efficiency is as high as possible.
In the example shown in Figure 14, four roadside units are installed, but there are areas near the center of the map that are outside the approximate range for obtaining driving information by each roadside unit, indicating that there is a possibility that there will be many gaps in the acquisition of driving information in such areas.

Figure 19 is a diagram showing the situation in Figure 14 when a fifth roadside unit is newly installed. As shown in Figure 19, the position of the newly installed roadside unit is indicated by a hollow diamond, and a circle indicating the driving information position that the new roadside unit will cover is indicated by a dotted line. In this way, by installing a new roadside unit in a manner that fills in the area that cannot be covered by the driving information acquisition by the existing roadside units, it is expected that more effective collection of driving information will be possible.

<Process for deriving new roadside unit installation position>
The following describes a process for deriving an appropriate position for a new roadside unit when the new roadside unit is installed.
FIG. 20 is a flowchart showing an example of the flow of a new roadside unit installation position derivation process executed by the processing unit 110 of the server 100 in this embodiment.

First, the processing unit 110 of the server 100 identifies an area where a new roadside unit is to be installed (S41). The area is identified, for example, by receiving a user-input designation of the area via the operation unit 120 or the communication unit 150. The area may be identified by a user-input designation of a range by a mouse operation or a pinch operation, or may be specified by entering characters to designate an address, such as "3-chome, C-cho, B-city, A-ken."
The method for identifying the area where the roadside unit is to be newly installed in step S41 is not limited to the above-mentioned method, and any method may be applied. For example, based on the position information of the existing roadside units, an area where a large number of areas that cannot be covered by the acquisition of driving information by the existing roadside units are extracted, and the area may be one of the extracted areas.

Next, by referring to the map information database 202 and the roadside device information database 198, information corresponding to the area identified in S31 as shown in FIG. 19 is displayed on the display unit 130 (S42). At this time, it is preferable to also display information on the approximate range for which driving information can be obtained by each roadside device.

Next, the position of the roadside unit to be newly installed is specified (S43). When specifying the position of the roadside unit to be newly installed, for example, the user may specify the position by looking at the content presented via the display unit 130 or the like (for example, manually specify a position that does not overlap as much as possible with the target coverage area of the existing roadside units), or multiple recommended positions may be specified based on the position information of the existing roadside units, and the user may specify one of the multiple recommended positions.
The latter case will be explained below with several examples.

<Identifying the location of a new roadside unit based on the location information of existing roadside units>
One example of a method for identifying the installation location of a new roadside unit is to use location information of existing roadside units in the vicinity.

As a specific method, by way of example and not limitation, as shown in Fig. 21, within a specified area, a position farthest from each existing roadside unit may be identified as the installation position of the new roadside unit. Alternatively, a score may be set that decreases as the distance from each existing roadside unit increases (e.g., inversely proportional to the distance), and the installation position of the new roadside unit may be identified based on the total value of the scores (e.g., the score with the lowest total value) may be identified. Furthermore, a plurality of candidate positions may be derived by such a method, and the plurality of candidate positions may be presented to the user via the display unit 130 or the like, and one of them may be selected by the user to identify it as the installation position of the new roadside unit.
Also, based on the estimated ranges for acquiring travel information by each roadside device, the location of the new roadside device may be determined to be the furthest from those ranges.Furthermore, the estimated ranges for acquiring travel information by each roadside device may be compared with the estimated range for acquiring travel information by the new roadside device (e.g., a circular area of radius s; s=r or s≠r) and the location of the new roadside device may be determined taking into consideration the smallness of the overlapping area.

<Identifying the location of a new roadside unit based on the density of location information of multiple moving objects acquired separately>
Another example of a method for identifying the installation location of a new roadside unit is to perform the identification based on the density of position information of a plurality of moving objects obtained separately.

22 is a diagram showing the situation in FIG. 14 superimposed with a display based on the density of location information included in driving information measured by a cigarette lighter device, with reference to the cigarette lighter device measurement database 192. A circle indicates a location where the location information was observed at a predetermined density (e.g., 20 times/day) or more, and the larger the circle, the greater the density.
Based on the density of the location information as shown in Figure 22, for example, the location with the highest density may be identified as the installation location of the new roadside unit, or the roadside unit location with the highest density within the estimated range for obtaining driving information by the new roadside unit may be identified as the installation location of the new roadside unit.
Furthermore, similarly to the above-mentioned method, a plurality of candidate positions may be derived by the above-mentioned method, and the plurality of candidate positions may be presented to the user via the display unit 130 or the like, and one of them may be selected by the user to be specified as the installation position of the new roadside unit. Also, the installation position of the new roadside unit may be specified by comparing the estimated range of travel information acquisition by each existing roadside unit with the estimated range of travel information acquisition by a new roadside unit, taking into consideration the smallness of the overlapping area.

Note that, although the location information used in this embodiment is included in the driving information stored in the cigarette lighter type device measurement database 192, it is not limited to this. For example, location information collected by another means may be used. This other means may be provided by a device compatible with ETC 2.0.

<Identifying facilities where new roadside units can be installed based on their locations>
As yet another example of a method for identifying the installation location of a new roadside unit, facilities where new roadside units can be installed may be extracted in advance as candidate installation sites, and the installation location of the new roadside unit may be identified from these candidate installation sites.

Fig. 23 shows installation candidate sites superimposed near the center of the map shown in Fig. 14. In Fig. 23, for the situation in Fig. 14, for example, shopping malls, gas stations, and convenience stores are extracted and shown as installation candidate sites by referring to the roadside unit new installation facility information database 203. Of course, the types of installation candidate sites are not limited to these, and any type can be set.
In the example shown in Figure 23, of these candidate locations, a shopping mall has been identified as the location for a new roadside unit. A hollow diamond is placed in the shopping mall to indicate that this is a new installation location, and a dotted line is used to indicate the approximate range for obtaining driving information based on the location of the shopping mall.

That is, as shown in FIG. 23 , map information of the area where a new roadside unit is to be installed may be presented to the user via the display unit 130 or the like, with existing roadside units and the approximate range of driving information positions covered by the existing roadside units, as well as candidate installation sites for the roadside unit, superimposed thereon, and the user may be allowed to select one of the candidate installation sites, thereby identifying it as the installation location for the new roadside unit.
Furthermore, similar to the above-mentioned method, the approximate range of driving information positions covered by each existing roadside unit may be compared with the approximate range of driving information positions expected to be covered by a new roadside unit, and the installation location of the new roadside unit may be identified taking into consideration the smallness of overlapping areas.

Although the method of identifying the installation location of a new roadside unit has been described above, the method is not limited to these, and any method may be used. For example, a method may be a combination of two or more of the three methods described above. That is, for example, in a state in which density information of the location where driving information is acquired and candidate installation site information of facilities where roadside units can be installed are known in advance and displayed superimposed on map information, multiple candidate installation locations for a new roadside unit are derived based on the density information and candidate installation site information and displayed to the user via the display unit 130 or the like, and the user can select one of them to identify the installation location of the new roadside unit.

In the above explanation, it is assumed that a certain number of roadside units have already been installed, but this is not limited to the above, and it is also possible to identify new installation positions for roadside units in areas where no roadside units are installed.
Also, instead of identifying a single new roadside unit position, multiple new roadside unit positions may be identified. That is, after identifying one new roadside unit position, the next new roadside unit position may be identified assuming that the new roadside unit has already been installed, or multiple new roadside unit positions may be identified simultaneously based on the above-mentioned factors.

As in the second embodiment, it is possible to compare the expected degree of acquisition of driving information before and after the installation of a new roadside unit to determine how much the amount of acquired driving information is expected to increase, and further to compare the amount of acquired driving information after installation with a threshold value, thereby making it possible to evaluate whether the acquisition status of driving information is sufficient, etc. In other words, it is possible to determine whether a business such as an insurance product is viable, for example, based on the amount of acquired driving information expected by installing a new roadside unit.

<Effects of the Fourth Embodiment>
In this embodiment, the server 100 identifies a location for installing a new roadside unit based on user input, location information of existing roadside units, the density of location information obtained from multiple mobile objects, location information of facilities where a roadside unit can be installed, etc.
This makes it possible to know in advance the most effective locations for installing roadside units that acquire driving information using a method compatible with ETC 2.0, without having to install new roadside units.

＜Other＞
The server 100 in the above embodiment may be configured as multiple physically separated servers, with a first server performing part of the processing described in the above embodiment and a second server performing other processing. A server system may be considered to be configured by multiple servers.

Movements are not limited to four-wheeled vehicles, but may be two-wheeled vehicles, personal mobility vehicles, ships, trains, aircraft, etc., as described above. In this case, the various measuring devices can be the first device and the second device mounted on or built into the moving object, and the same processing as in the above embodiment can be performed.

As described above, for example, the measurement object may be "sound," the sampling unit of the first device may be "time," and the sampling unit of the second device may be "frequency," and similar processing may be performed.
In addition, for example, processing similar to the above may be performed by measuring information related to a living body (blood pressure, pulse rate, heart rate, respiratory rate, body temperature, brain waves, etc.). In this case, for example, the sampling unit of the first device may be set to "time" and the sampling unit of the second device may be set to "frequency", and the same processing as the above may be performed.

As an example of this application, when the insured user is driving a vehicle, biometric information of the user is measured by a first device, and the server 100 may generate estimated second device measurement data in the case where the biometric information of the user is measured by a second device, based on first device measurement data obtained by measuring the biometric information by the first device.
In this case, the driving information of the vehicle driven by the user may be measured by a cigarette lighter type device or the like, and the server 100 may generate the above-mentioned estimated ETC 2.0 measurement data. In other words, the insured user and the vehicle may be linked, and the server 100 may generate estimated data of information regarding a living body in addition to generating the estimated ETC 2.0 measurement data. The user's health condition may be evaluated in addition to evaluating the user's driving (driving quality).
In addition, estimated data of information regarding the user's biological information may be generated not only while driving, but also while not driving.

Also, the cigarette lighter type device measurement data and the ETC 2.0 measurement data are acquired from a vehicle equipped with the cigarette lighter type device measurement data and the ETC 2.0 on-board device. Then, the proportion of data loss in the estimated ETC 2.0 measurement data generated based on the cigarette lighter type device measurement data with respect to the actual ETC 2.0 measurement data may be calculated.
In addition, the score calculated by inputting estimated ETC 2.0 measurement data into the second driving score model may be compared with the score calculated by inputting actual ETC 2.0 measurement data into the second driving score model, and if the difference is below a threshold value (or less than the threshold value), the estimated ETC 2.0 measurement data may be determined to be valid.

In the above embodiments, various programs and data relating to various processes are stored in the storage unit, and the processing unit reads and executes these programs to realize the processes in the above embodiments. In this case, the storage unit of each device may have internal storage devices such as ROM, EEPROM, flash memory, hard disk, and RAM, as well as recording media (recording media, external storage devices, storage media) such as memory cards (SD cards), CompactFlash (registered trademark) cards, memory sticks, USB memory, CD-RW (optical disks), and MO (magneto-optical disks), and the above various programs and data may be stored in these recording media.

Reference Signs List 100 Server 110 Processing section 120 Operation section 130 Display section 140 Sound output section 150 Communication section 160 Clock section 190 Storage section

Claims (14)

An information processing device,
a data generating unit that generates, based on first measurement data measured by a first device that measures information about the moving object, estimated data of second measurement data measured by a second device that measures information about the moving object and is different from the first device;
a model generation unit that generates a model for calculating various quantities related to the moving object in a driving evaluation using the second device or in telematics insurance based on the estimated data;
Equipped with
The data generation unit generates at least a portion of the estimated data of the second measurement data based on the following formula:
D2(1)=D1(1)
D2(k) = D1g(D2(k-1))
here,
k is a natural number equal to or greater than 2,
D1 and D2 indicate data in an argument that is a sequence value from a reference data when the first measurement data and the estimated data of the second measurement data are arranged in a time series, respectively, and
The function D1g outputs data that satisfies a predetermined distance relationship or a predetermined time relationship with the data corresponding to the argument in the first measurement data.
Information processing device. An information processing device,
a function generating unit that generates a function that approximates traveling information included in first measurement data based on first measurement data measured by a first device that measures information about a moving object;
A calculation unit that sequentially calculates time information that satisfies a predetermined time condition;
a data generating unit that generates, based on the generated function and the calculated time information, estimated data of second measurement data measured by a second device that measures information about a moving object and is different from the first device;
a model generation unit that generates a model for calculating various quantities related to the moving object in a driving evaluation using the second device or in telematics insurance based on the estimated data;
An information processing device comprising: 1. An information processing method, comprising:
generating, based on first measurement data measured by a first device that measures information about the moving object, estimated data of second measurement data measured by a second device that measures information about the moving object and is different from the first device;
generating a model for calculating various quantities related to the moving object in a driving evaluation using the second device or in telematics insurance based on the estimated data;
Including,
At least a portion of the estimated data is generated based on the following formula:
D2(1)=D1(1)
D2(k) = D1g(D2(k-1))
here,
k is a natural number equal to or greater than 2,
D1 and D2 indicate position information of data in an argument, which is an order value from a reference data when the first measurement data and the estimated data of the second measurement data are arranged based on a time series, respectively, and
The function D1g outputs data that satisfies a predetermined distance relationship or a predetermined time relationship with the data corresponding to the argument in the first measurement data.
Information processing methods. 1. An information processing method, comprising:
generating a function that approximates travel information included in first measurement data measured by a first device that measures information about a moving object;
Sequentially calculating time information when a predetermined time condition is satisfied;
generating, based on the generated function and the calculated time information, estimated data of second measurement data measured by a second device that measures information about a moving object and is different from the first device;
generating a model for calculating various quantities related to the moving object in a driving evaluation using the second device or in telematics insurance based on the estimated data;
An information processing method comprising: On the computer,
generating, based on first measurement data measured by a first device that measures information about the moving object, estimated data of second measurement data measured by a second device that measures information about the moving object and is different from the first device;
generating a model for calculating various quantities related to the moving object in a driving evaluation using the second device or in telematics insurance based on the estimated data;
A program for executing
At least a portion of the estimated data is generated based on the following formula:
D2(1)=D1(1)
D2(k) = D1g(D2(k-1))
here,
k is a natural number equal to or greater than 2,
D1 and D2 indicate position information of data in an argument, which is an order value from a reference data when the first measurement data and the estimated data of the second measurement data are arranged based on a time series, respectively, and
The function D1g outputs data that satisfies a predetermined distance relationship or a predetermined time relationship with the data corresponding to the argument in the first measurement data.
program. On the computer,
generating a function that approximates travel information included in first measurement data measured by a first device that measures information about a moving object;
Sequentially calculating time information when a predetermined time condition is satisfied;
generating, based on the generated function and the calculated time information, estimated data of second measurement data measured by a second device that measures information about a moving object and is different from the first device;
generating a model for calculating various quantities related to the moving object in a driving evaluation using the second device or in telematics insurance based on the estimated data;
A program to execute. A trained model for causing a computer to function to output various quantities related to a moving object in a driving evaluation using a second device or a telematics insurance based on information about the moving object acquired by the second device,
learning is performed based on estimated data of second measurement data measured by the second device, the estimated data being generated based on first measurement data measured by a first device different from the second device;
At least a portion of the estimated data of the second metrology data is generated based on the following formula:
D2(1)=D1(1)
D2(k) = D1g(D2(k-1))
here,
k is a natural number equal to or greater than 2,
D1 and D2 indicate position information of data in an argument, which is an order value from a reference data when the first measurement data and the estimated data of the second measurement data are arranged based on a time series, respectively, and
The function D1g outputs data that satisfies a predetermined distance relationship or a predetermined traveling direction relationship with the data corresponding to the argument in the first measurement data.
Trained model . A trained model for causing a computer to function to output various quantities related to a moving object in a driving evaluation using a second device or a telematics insurance based on information about the moving object acquired by the second device,
learning is performed based on estimated data of second measurement data measured by the second device, the estimated data being generated based on first measurement data measured by a first device different from the second device;
At least a portion of the estimated data of the second metrology data is generated based on the following formula:
D2(1)=D1(1)
D2(k) = D1g(D2(k-1))
here,
k is a natural number equal to or greater than 2,
D1 and D2 indicate position information of data in an argument, which is an order value from a reference data when the first measurement data and the estimated data of the second measurement data are arranged based on a time series, respectively, and
The function D1g outputs data that satisfies a predetermined time relationship with the data corresponding to the argument in the first measurement data.
Trained model . A trained model for causing a computer to function to output various quantities related to a moving object in a driving evaluation using a second device or a telematics insurance based on information about the moving object acquired by the second device,
learning is performed based on estimated data of second measurement data measured by the second device, the estimated data being generated based on a function approximating driving information included in the first measurement data, the function being generated based on first measurement data measured by a first device different from the second device, and time information being sequentially calculated based on the first measurement data and a predetermined distance condition or a predetermined traveling direction condition;
Trained model . A trained model for causing a computer to function to output various quantities related to a moving object in a driving evaluation using a second device or a telematics insurance based on information about the moving object acquired by the second device,
The learning is performed based on estimated data of the second measurement data measured by the second device, the estimated data being generated based on a function approximating driving information included in the first measurement data measured by a first device different from the second device, and time information being sequentially calculated based on a predetermined time condition.
Trained model . A trained model according to any one of claims 7 to 10,
The first device directly transmits the acquired first measurement data to a server;
The second device transmits the acquired second measurement data to the server via a roadside device.
Trained model . A trained model according to any one of claims 7 to 10,
the first device transmits the acquired first measurement data to a server without substantially storing the data internally;
the second device stores the acquired second measurement data therein, and then transmits the stored second measurement data to the server;
Trained model . The trained model according to claim 12,
In the second device, when an amount of the second measurement data stored therein reaches a predetermined amount, the oldest data of the stored second measurement data is discarded in response to new data being acquired as the second measurement data.
Trained model . A trained model according to any one of claims 7 to 10,
In the first measurement data, data within a first predetermined range from a traveling start time point or data within a second predetermined range from a traveling end time point is excluded from the learning target.
Pre-trained models.

Images