WO2025169305A1 - Accident inference device, insurance fee examination device, on-vehicle equipment, terminal device, information processing system, accident inference method, accident inference program, ai model generation method, and ai generation program - Google Patents

Abstract

An accident inference device according to one embodiment of the present invention is provided with a controller. When a near-miss occurs, the controller infers an accident that is likely to happen in said near-miss situation, on the basis of the situation of the near-miss. The controller outputs accident information related to the inferred virtual accident.

Description

Accident estimation device, insurance premium review device, in-vehicle device, terminal device, information processing system, accident estimation method, accident estimation program, AI model generation method, and AI generation program

The disclosed embodiments relate to an accident estimation device, an insurance premium review device, an on-board device, a terminal device, an information processing system, an accident estimation method, an accident estimation program, an AI model generation method, and an AI generation program.

There is a danger point information display device that contributes to traffic safety by notifying or presenting vehicle drivers with accident information such as the location, details, and surrounding environment of past accidents, as well as information on risk factors at the location of near misses that could lead to accidents (see, for example, Patent Document 1).

JP 2007-51973 A

However, because near misses do not become accidents, information about near misses is somewhat less stimulating. Furthermore, information about past accidents and past accidents and near misses is something that does not concern anyone other than those involved in the accident. For these reasons, it is insufficient as teaching material to improve the effectiveness of traffic safety education.

One aspect of the embodiment has been made in consideration of the above, and aims to provide an accident estimation device, insurance premium review device, in-vehicle device, terminal device, information processing system, accident estimation method, accident estimation program, AI model generation method, and AI generation program that can contribute to improving the effectiveness of traffic safety education.

An accident estimation device according to one aspect of the embodiment includes a controller. When an attempted accident, known as a near miss ("something dangerous happened, but fortunately did not result in a disaster"), occurs, the controller estimates an accident that is likely to occur in the near miss situation based on the situation. The controller outputs accident information related to the accident.

The accident estimation device, insurance premium review device, in-vehicle device, terminal device, information processing system, accident estimation method, accident estimation program, AI model generation method, and AI generation program according to one aspect of the embodiment estimate and generate information on accidents that are feared to occur in response to near misses that involve the individual. This allows for the notification of information that is relevant to the individual, thereby contributing to improving the effectiveness of traffic safety education.

FIG. 1 is an explanatory diagram showing the configuration of an accident estimation device and a vehicle according to an embodiment. FIG. 2 is an explanatory diagram showing a method for generating an accident estimation AI model according to the embodiment. FIG. 3 is an explanatory diagram of a method for creating training data according to the embodiment. FIG. 4 is an explanatory diagram illustrating the configuration of the accident information generating unit according to the embodiment. FIG. 5 is an explanatory diagram illustrating an example of a table stored in the video DB according to the embodiment. FIG. 6 is an explanatory diagram illustrating an example of a table stored in the video DB according to the embodiment. FIG. 7 is an explanatory diagram illustrating an example of a table stored in the video DB according to the embodiment. FIG. 8 is an explanatory diagram illustrating an example of a table stored in the accident DB according to the embodiment. FIG. 9 is an explanatory diagram showing the configuration of an accident information generation unit equipped with AI according to the embodiment. FIG. 10 is an explanatory diagram illustrating an example of a training dataset for the video information generation AI model according to the embodiment. FIG. 11 is an explanatory diagram showing an example of a learning method for a video information generation AI model according to an embodiment. FIG. 12 is an explanatory diagram showing an example of the operation of the video information generation AI model according to the embodiment. FIG. 13 is an explanatory diagram illustrating an example of a learning dataset for the damage information generation AI model according to the embodiment. FIG. 14 is an explanatory diagram showing an example of a learning method for a damage information generation AI model according to an embodiment. FIG. 15 is an explanatory diagram illustrating an example of the operation of the damage information generation AI model according to the embodiment. FIG. 16 is a flowchart showing a process executed by the controller of the accident estimation device according to the embodiment. FIG. 17 is a flowchart showing a process executed by the controller of the accident estimation device according to the embodiment. FIG. 18 is an explanatory diagram of an information processing system according to an embodiment.

Below, with reference to the attached drawings, embodiments of the accident estimation device, insurance premium review device, in-vehicle device, terminal device, information processing system, accident estimation method, accident estimation program, AI model generation method, and AI generation program will be described in detail. Note that the present invention is not limited to the embodiments shown below.

[1. Overview of accident estimation method]
First, an outline of an accident estimation method performed by an accident estimation device according to an embodiment will be described. The accident estimation device includes a computer, and the computer executes an accident estimation program stored in a memory to realize the accident estimation method according to the embodiment.

The accident estimation device according to the embodiment is a device that, when a near-miss accident (hereinafter referred to as a "near miss") that is likely to develop into an accident occurs in a moving vehicle, estimates the accident that is likely to occur in the circumstances of the near miss and outputs accident information related to that hypothetical accident, i.e., hypothetical accident information based on the near miss. The accident estimation device estimates the accident that is likely to occur based on the circumstances of the near miss.

In this way, the accident prediction device can output and provide accident information about accidents that are likely to occur as a result of that near miss to drivers who have actually experienced a near miss, raising their awareness of the danger of accidents. Drivers are therefore provided with accident information that is relevant to them, and the accident prediction device can contribute to improving the effectiveness of traffic safety education.

[2. Accident estimation device]
An accident estimation device 1 according to an embodiment will be described with reference to Fig. 1. Fig. 1 is an explanatory diagram showing the configuration of the accident estimation device 1 and a vehicle 2 according to an embodiment. First, the configuration of the vehicle 2 according to the embodiment will be described.

As shown in FIG. 1, a drive recorder 20 having the ability to communicate with external devices is installed in the vehicle 2. The drive recorder 20 is connected to the accident estimation device 1 so that it can communicate information with the accident estimation device 1 via a communication network N such as the Internet. The drive recorder 20 is also connected to a steering angle sensor 22 and emotion sensor 23 installed in the vehicle 2 via an in-vehicle LAN (Local Area Network). The drive recorder 20 is further equipped with an acceleration sensor 21, an on-board camera 24, an on-board microphone 25, a flash memory 26, a controller 27, and a communication unit 28.

The controller 27 acquires vehicle steering angle information detected by the steering angle sensor 22 and driver emotion information detected by the emotion sensor 23. The controller 27 also acquires vehicle acceleration detected by the acceleration sensor 21, image information captured by the onboard camera 24, and audio information collected by the onboard microphone 25.

The controller 27 stores the image information captured by the onboard camera 24 and the audio information collected by the onboard microphone 25 in the flash memory 26. The controller 27 also reads out the image information and audio information stored in the flash memory 26, uses them for various processes, and transmits them to an external device such as the accident estimation device 1 via the communication unit 28.

The acceleration sensor 21 is a sensor that detects the acceleration of the vehicle. For example, it consists of an elastic body and a mass body, and detects acceleration by measuring the force applied to the mass body due to acceleration as the displacement of the elastic body. It can be configured as a frequency change type, piezoelectric type, piezo-resistive type, capacitance type, or other acceleration sensor. The steering angle sensor 22 is a sensor that detects the rotation state of the steering wheel of the vehicle 2.

The emotion sensor 23 is a sensor that detects biometric information to estimate the driver's emotions, such as a heart rate sensor or brain wave sensor, and emotions are estimated based on the brain waves and heart rate detected by these sensors. The heart rate sensor is provided on the steering wheel or other part of the vehicle 2. Note that a method of estimating emotions from the driver's facial information (facial expression) using artificial intelligence or the like can also be applied. In this case, the on-board camera 24 can be used as a sensor to acquire the driver's facial information.

The onboard camera 24 includes an interior imaging device that captures video from within the cabin of the vehicle 2, and an exterior imaging device that captures video from around the vehicle 2. The onboard microphone 25 is a sound collection device that collects audio from within the cabin of the vehicle 2. The flash memory 26 is a recording device that stores the video from within the cabin captured by the onboard camera 24 and the audio from within the cabin collected by the onboard microphone 25, as well as the video from around the vehicle 2 captured by the onboard camera 24 and the audio from around the vehicle 2 collected by the onboard microphone 25.

The controller 27 includes a microcomputer with a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and various other circuits. The controller 27 is equipped with a near-miss detection unit 29 that functions when the CPU executes a program stored in the ROM using the RAM as a working area.

The controller 27 may be partially or entirely composed of hardware such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). The communication unit 28 is a communication interface that communicates information with the accident estimation device 1 via the communication network N.

Next, we will explain the operation of each component of the vehicle 2. The acceleration sensor 21 detects acceleration in the front-rear, left-right, up-down directions of the vehicle 2, assuming that the forward direction of the vehicle 2 is the front-rear direction, and outputs the results to the controller 27.

The steering angle sensor 22 uses the state of the steering wheel when the vehicle 2 is traveling straight as a reference position, and detects the steering wheel's rotation angle, rotation speed, and rotation acceleration from the reference position, and outputs these to the controller 27. Note that the controller 27 may calculate the steering wheel's rotation speed and rotation acceleration based on the rotation angle through arithmetic processing (differential processing and second-order differential processing).

The emotion sensor 23 detects the driver's heartbeat from the pulse of the driver's palms gripping the steering wheel. When a near miss occurs, the driver's emotions change suddenly, and so do biosignals that correlate with emotions, such as heartbeat. Therefore, the drive recorder 20 uses the heartbeat to estimate emotions (emotional state at the time of the near miss). The emotion sensor 23 detects the driver's heartbeat and outputs it to the controller 27. Note that the emotion sensor 23 can also be a sensor that detects biosignals other than heartbeat that are correlated with emotions, such as brain waves.

The onboard camera 24 outputs images of the interior of the vehicle 2 and images of the surroundings of the vehicle 2 to the controller 27. The onboard microphone 25 picks up audio from within the interior of the vehicle 2 and outputs the audio to the controller 27.

The controller 27 stores the video input from the in-vehicle camera 24 and the audio input from the in-vehicle microphone 25 in the flash memory 26. The controller 27 associates the input video with the date and time it was captured, and the input audio with the date and time it was collected, and stores these in the flash memory 26.

The controller 27 may also store the steering wheel rotation angle, rotation speed, and rotation acceleration detected by the steering angle sensor 22, as well as emotion information (heart rate information) detected by the emotion sensor 23, in the flash memory 26, in association with the date and time.

The near-miss detection unit 29 detects near-misses that occur while the vehicle 2 is traveling. The near-miss detection unit 29 detects near-misses based on at least one of the acceleration of the vehicle 2 input from the acceleration sensor 21, the steering wheel rotation angle, rotation speed, and rotation acceleration input from the steering angle sensor 22, and the driver's heart rate.

The near-miss detection unit 29 determines that a near-miss has occurred when the acceleration of the vehicle 2 exceeds an acceleration threshold. In other words, when a near-miss occurs, sudden braking or abrupt steering is often performed, and some kind of impact is often applied even if it does not result in an accident, so the occurrence of a near-miss is detected based on the acceleration detection status.

The lower threshold of acceleration used to determine whether a near miss has occurred is the upper limit (with an appropriate offset) of acceleration that can occur when vehicle 2 is being driven safely, and the upper threshold is the lower limit (with an appropriate offset) of acceleration when an accident occurs, and is set based on experiments during design and development, for example. The deployment status of vehicle 2's airbags may also be used to determine whether a near miss has occurred (if the airbags are deployed, it is determined that an accident has occurred but not a near miss).

Furthermore, the near-miss detection unit 29 determines that a near-miss has occurred when the rotational speed and rotational acceleration of the steering wheel exceed the corresponding rotational speed threshold and rotational acceleration threshold, respectively. In other words, when a near-miss occurs, the driver often makes an abrupt turn in the steering wheel, so it determines that a near-miss has occurred when the rotational speed and rotational acceleration of the steering wheel are greater than the threshold. Note that the rotational speed threshold and rotational acceleration threshold are set to values that provide good judgment accuracy, for example, based on experiments during design and development.

The rotational speed threshold and rotational acceleration threshold are the upper limit (with an appropriate offset) of the rotational speed and rotational acceleration of the steering wheel operated by the driver when vehicle 2 is being driven safely, and are set to values that provide good judgment accuracy based on, for example, experiments conducted during design and development.

The near-miss detection unit 29 determines that a near-miss has occurred if the increase in the driver's heart rate within the determination time exceeds the heart rate threshold. The determination time here is, for example, 10 seconds. Note that the determination time here is not limited to 10 seconds. The heart rate threshold is the upper limit (with an appropriate offset) of the increase in the driver's heart rate within the determination time while the vehicle 2 is being driven safely. Note that the determination time and heart rate threshold are set to appropriate values that provide good determination accuracy, for example, based on experiments during design and development.

In addition, the near-miss occurrence determination may be made by a combination of the above-mentioned determination based on the acceleration state of the vehicle 2, determination based on the steering wheel rotation state, and determination based on the driver's heart rate (emotions) state (determination based on the results of a logical operation of these determination results).

In this way, if the near miss detection unit 29 can detect the occurrence of a near miss, information about the location where the near miss occurred and the circumstances of the near miss can be provided to, for example, an institution that provides traffic safety education. Then, the institution that provides traffic safety education can carry out traffic safety awareness activities, for example, by creating a near miss map that clearly shows the location where the near miss occurred and the circumstances of the near miss, and providing this to drivers, etc.

However, drivers do not make a strong impression when they see a near miss, partly because no accident actually occurs. For example, when a near miss occurs, they may not know what kind of accident it could have become, the extent of the damage, or the percentage of responsibility for the accident, so the situation when the near miss escalates into an accident can be unclear, which can make it seem a little unrealistic.

On the other hand, insurance that compensates for damages caused by accidents comes in two types: distance-linked insurance (PAYD), where the premium is determined according to the distance driven, and behavior-linked insurance (PHYD), where the premium is linked to driving operations. PHYD uses information from various sensors in the in-car device to determine whether the driver is driving safely based on the number and frequency of sudden braking, sudden acceleration, sudden steering, etc. Compared to PAYD, PHYD is said to be able to grasp the risk of driving accidents more accurately and reflect this in insurance premiums.

In this way, insurance companies use various methods to set appropriate insurance premiums. However, the reality is that the information used to set premiums is primarily data related to accidents that have occurred, and information on near misses that did not result in accidents is not used sufficiently.

For example, drivers who have a high frequency of near misses are predicted to have a high probability of causing an accident, and drivers who have a high frequency of near misses that lead to accidents causing major damage are predicted to have a high probability of causing a major accident, but the reality is that this near miss information is not used to calculate insurance premiums.

Therefore, when the near-miss detection unit 29 according to the embodiment determines that a near-miss has occurred, it transmits near-miss information relating to the near-miss to the accident estimation device 1. The accident estimation device 1 estimates a likely accident scene based on the near-miss information received from the near-miss detection unit 29 of the vehicle 2, and outputs accident information relating to that accident scene.

For example, the accident estimation device 1 outputs accident type information indicating the type of accident that is feared to occur as a result of a near miss, and hypothetical accident information indicating the accident event that is feared to occur as a result of a near miss.

This means that, for example, by checking estimated accident information, drivers can find out what kind of accident could have occurred when a near miss occurs, the extent of damage, and the percentage of responsibility for the accident, as well as the severity and impact of the accident if the near miss escalates into an accident.

On the other hand, by checking the estimated accident information, insurance companies can estimate the potential probability of an accident occurring and the scale of the accident (scale of damage), which is expected to enable them to calculate more appropriate insurance premiums.

Specifically, when the near miss detection unit 29 determines that a near miss has occurred, it reads from the flash memory 26 the date and time data and location data of the near miss, as well as video and audio recorded a predetermined time before and after the date and time of the near miss.

Furthermore, the near-miss detection unit 29 reads from the flash memory 26 the acceleration data detected by the acceleration sensor 21, the steering speed data detected by the steering angle sensor 22, and the driver's emotion data detected by the emotion sensor 23.

The specified time here is the length of time for the various pieces of information necessary to understand the characteristics of the near-miss occurrence situation, and is, for example, 5 seconds, but is not limited to 5 seconds and can be any appropriate time determined based on experiments, etc. during the design and development of the product system.

The near-miss detection unit 29 outputs these various data when a near-miss occurs to the communication unit 28. The communication unit 28 transmits these various data when a near-miss occurs, which are input from the controller 27, to the accident estimation device 1 as near-miss information.

Next, the configuration of the accident estimation device 1 will be described. The accident estimation device 1 includes a controller 10, a communication unit 11, and an accident database 12 (hereinafter referred to as "accident DB 12"). The communication unit 11 is a communication interface that communicates information with the vehicle 2 via the communication network N.

The controller 10 includes a microcomputer with a CPU, ROM, RAM, etc., and various circuits. It is equipped with an accident estimation model 13 that functions when the CPU executes a program stored in the ROM using the RAM as a work area, and an accident information generation unit 14.

The controller 10 may be configured in part or in whole using hardware such as an ASIC or FPGA. The accident DB 12 is an information storage device such as a data flash, and stores information about accidents that have occurred in the past. An example of the information stored in the accident DB 12 will be described later with reference to Figure 4.

Next, the operation of each component of the accident estimation device 1 will be described. When the communication unit 11 receives near-miss information from the vehicle 2, it outputs the received near-miss information to the accident estimation model 13 of the controller 10.

Accident inference model 13 is an inference model that, when near-miss information is input, outputs an accident scene that is likely to occur in the near-miss situation corresponding to the near-miss information. Accident inference model 13 can be realized by a matching process using a database (accident type database) made up of accident scene information (information that identifies the type of accident) that uses various information from the near-miss information as parameters, but because accident scenes come in a wide variety of patterns (types), it is more realistic to realize it using an AI (Artificial Intelligence) model.

This AI model is an AI model that has been machine-learned to estimate accident scenes based on input near-miss information. For ease of understanding, accident estimation model 13 using an AI model will be referred to as accident estimation AI model 13A.

The circumstances surrounding the accident (level of damage, etc.) are estimated based on the relative speed with respect to the object struck, the direction of impact (direction relative to the vehicle (e.g., the forward direction of the vehicle)), and the point of impact (location) immediately before the accident (collision, etc.). The circumstances surrounding the accident (level of damage, etc.) also change depending on the environment around the accident site; for example, accident footage can vary greatly depending on the surrounding scenery.

For this reason, increasing the number of types of parameters used in estimation enables more detailed estimation that takes into account the data for each item. By increasing the types of information input to the accident estimation AI model 13A and the level of detail of that information, and by training the accident estimation AI model 13A with a large amount of detailed data, the accident estimation AI model 13A can make more detailed estimations.

However, in practice, the data used for estimation must be of an appropriate type and with an appropriate level of accuracy, taking into account the performance of the device (CPU, etc.) and the AI learning time. For this reason, and to make the explanation easier to understand, this embodiment will explain using an example in which the data items used for estimation are limited to items with a large impact. However, even if other items are used, the accident situation can be estimated using similar processing.

The accident estimation AI model 13A is a model that estimates the accident situation based on near-miss information. It estimates the accident situation that would occur if appropriate evasive action were not taken in the event of a near-miss, based on the similarity between the situation immediately before the near-miss information was generated (before any action to avoid the accident was taken) and the situation immediately before the accident occurred (the amount of time before the action to avoid the accident was taken).

Therefore, when learning the accident estimation AI model 13A, the input data for the learning data (accident data) is "status data on vehicle driving, etc. immediately before the accident," and the correct answer data is "accident result (accident situation)." Therefore, the input data for the learning data is vehicle status data for an appropriate time period (an appropriate time is set based on experiments, etc.) ending an appropriate time before the accident occurs (an appropriate time is set based on experiments, etc.).

Hereinafter, this time period will be referred to as the estimated data time period. Furthermore, when the accident estimation model 13 is used (estimated operation), the input data is "operational state data before evasive action in the near miss," and in reality, just as during learning, the input data for the learning data is vehicle state data from the estimated data time period (an appropriate time period (an appropriate time is set based on experiments, etc.) ending an appropriate time before the near miss occurs (an appropriate time is set based on experiments, etc.)).

In this embodiment, the near-miss information includes the detection results of the acceleration sensor 21 (acceleration sensor information) during the estimated data time period when the near-miss occurred. The near-miss information also includes the detection results of the steering angle sensor 22 (steering angle sensor information) during the estimated data time period.

The near miss information includes the detection results (emotion sensor information) of the emotion sensor 23 during the estimated data time period when the near miss occurred. The near miss information includes images captured by the in-vehicle camera 24 during the estimated data time period (in-vehicle camera video). The near miss information includes audio picked up by the in-vehicle microphone 25 during the estimated data time period (in-vehicle microphone audio).

Here, a method for generating (learning) the accident estimation AI model 13A will be described with reference to Figure 2. Figure 2 is an explanatory diagram showing a method for generating the accident estimation AI model 13A according to an embodiment. As shown in Figure 2, the accident estimation AI model 13A includes a DNN (Deep Neural Network) 13B that performs predictive calculations from input values to derive output values.

The accident estimation AI model 13A is generated, for example, by supervised learning. In supervised learning, a learning data set is prepared that consists of a large amount of learning data in which input values are paired with correct values. In the case of the learning data for the accident estimation AI model 13A, predictive information, which is information similar to near-miss information immediately before an accident or near-miss occurs, is used as the input value, and the accident scene of the accident (or an accident predicted to occur from a near-miss) is used as the correct value.

As mentioned above, the input predictive information includes acceleration sensor information, steering angle sensor information, emotion sensor information, in-vehicle camera footage, and in-vehicle microphone audio from the estimated data time period at the time of the accident. Accident scenes that serve as correct answers are associated with identification information (accident scene ID) that identifies each accident scene.

Next, we will explain how to create this supervised learning data, using three examples.

First learning data creation method: Near miss precursor states are extracted from the near miss occurrence process based on near miss information, and accident information on accidents with accident precursor states similar to the near miss precursor states is extracted. Then, the precursor information corresponding to the near miss precursor states is used as input data, and learning data is created in which the corresponding accident information is used as correct answer data.

Specifically, as shown in Figure 3, the correct data is accident information (accident scene) that includes predictive information similar to that of a near miss (information similar to the information immediately before the accident occurs in the event information) during an appropriate period before the accident avoidance action (sudden deceleration, sudden steering, avoidance action by the other vehicle (determined from images, etc.)) in the process of the near miss occurring. Note that the time of the accident avoidance action can be determined based on the vehicle's driving (operation) status (speed sensor information, steering angle sensor information, etc.), and the data collection period can be set at the time of the accident avoidance action, but as explained above, the estimated data time period when the near miss occurs (time of the accident avoidance action) can also be determined through experiments, etc.

In other words, since it is believed that an accident will occur if no action (maneuver) is taken to avoid an accident in the event of a near miss, accident information corresponding to the near miss information is extracted by comparing the similarity of data in which no action is taken to avoid the accident. This learning data creation method can be done manually (by looking at near miss information and accident information and creating it based on the above-mentioned perspective) or by computer processing (by comparing the similarity of near miss information and accident information based on the above-mentioned perspective).

Second learning data creation method: The second learning data creation method extracts accident precursor states (information) during the accident occurrence process based on accident information from actual accidents, and extracts near-miss information for near-misses that have near-miss precursor states (information) similar to the accident precursor states in question. The extracted near-miss information is then used as input data to create learning data with the corresponding accident information as correct answer data.

Specifically, as shown in Figure 3, accident information is used as the correct answer data, and the input data is near-miss information for near-misses that includes information similar to accident precursor information from the accident precursor period (a non-avoidance state period before the accident (a period during which the accident could have been prevented if evasive action had been taken, determined, for example, by an incident)) that ends an appropriate time before the accident occurred for the accident corresponding to the accident information. In other words, since it is considered that an accident can be limited to a near-miss if accident avoidance action (operation) is taken in the event of an accident, near-miss information corresponding to the accident information is extracted by comparing the similarity of data from before the non-avoidance state.

This learning data creation method can also be done manually (by looking at near-miss information and accident information and creating it based on the above-mentioned perspectives) or by computer processing (by comparing the similarities between near-miss information and accident information based on the above-mentioned perspectives).

Third learning data creation method: Based on near-miss information, near-miss information before evasive action is extracted during the near-miss occurrence process. Then, the vehicle's behavior if the behavior before the evasive action is maintained is predicted (if an other vehicle is present, the behavior of the other vehicle is also predicted), and accident information is predicted and generated. Note that data for predictive generation of accident information (various data for predictive generation of accident images and vehicle damage conditions) is generated in advance based on past accident information. For example, accident images are created using computer graphics technology using image components of vehicle images and landscape images, as well as various mechanical formulas and coefficients related to object movement.

In this embodiment, an identification code is assigned to each accident and this identification code is used as the correct answer data. Various information about each accident is managed and stored in a database in which this information is stored in data records that use the identification code as key data. In this case, the correct answer data in the training data is the identification code for each accident.

Furthermore, in the case of an accident estimation model 13 that uses an accident type database, for example, a group of information that associates input data in the above-mentioned learning data with corresponding correct answer data becomes the accident type database.

For example, a certain accident scene ID is associated with an accident scene (various data in the accident scene) called "a collision caused by an oncoming vehicle making an unreasonable right turn." Other accident scenes include, for example, "a collision with an oncoming vehicle going straight due to an unreasonable right turn," "a rear-end collision with a vehicle in front due to the vehicle in front suddenly braking," "a rear-end collision with a vehicle in front due to inattention to the road ahead," and "a collision caused by unreasonable overtaking."

DNN13B sequentially derives estimated values of likely accident scenes from the near-miss information (predictive information) of each piece of learning data in the sequentially input learning data set, and outputs the estimated values. Controller 10 causes accident estimation AI model 13A to perform machine learning by updating the weighting coefficients of each layer in DNN13B using processing such as backpropagation, so as to reduce the error between the estimated accident scene results output from DNN13B and the correct values of the learning data corresponding to the input values.

In other words, by executing the AI generation program, the controller 10 performs machine learning on the accident estimation AI model 13A so that the accident scene estimated by the accident estimation AI model 13A from the information about the near-miss situation inputted becomes an accident scene (correct data) that occurred in the input near-miss situation.

This allows the controller 10 to generate an accident estimation model 13 that can estimate, from the input near-miss information, the accident scene that is likely to occur in the near-miss situation corresponding to the near-miss information.

Returning to Figure 1, the controller 10 outputs near-miss information that has occurred in the vehicle 2, input via the communication unit 11, to the accident estimation model 13, which is configured using the accident estimation AI model 13A generated in this way. The controller 10 then uses the estimation process of the accident estimation model 13 to estimate an accident scene that is likely to occur in the near-miss situation.

In other words, when near-miss information is input, the accident inference model 13 infers an accident scene that is likely to occur in the near-miss situation corresponding to the near-miss information, and outputs the inference result to the accident information generation unit 14. At this time, the accident inference model 13 outputs the near-miss information used to infer the accident to the accident information generation unit 14, along with the accident scene ID of the inferred accident scene.

The accident information generation unit 14 generates accident information about the accident scene estimated by the accident estimation model 13 and outputs it to the information output device 15. The information output device 15 is a display capable of displaying an image of the accident information. The information output device 15 also includes a speaker capable of outputting the audio of the accident information. Furthermore, the information output device 15 includes a camera that captures images of viewers watching the accident information.

The information output device 15 outputs a captured image of the viewer viewing the accident information to the controller 10 of the accident estimation device 1. The information output device 15 may be configured to be connected to a heart rate sensor that measures the heart rate of the viewer viewing the accident information. In this case, the information output device 15 outputs information indicating the viewer's heart rate to the controller 10.

Next, the configuration and operation of the accident information generation unit 14 will be described with reference to Figures 4 to 8. As shown in Figure 4, the accident information generation unit 14 includes a video information generation unit 31 and a damage information generation unit 32. The video information generation unit 31 includes a video database (DB) 33, a video generation model 34, and an tolerance determination unit 35.

Next, a method for generating an accident video will be described.
First accident video generation method: The video DB 33 stores live-action videos, virtual reality (VR) videos, and computer graphics (CG) videos of various accident scenes. For example, the video DB 33 stores data such as that shown in FIG. 5, in which accident scene IDs are associated with accident videos. The data for the video DB 33 is collected and created in advance by design developers and the like, and stored in the video DB 33. In addition to the accident video, data such as accident audio may also be stored, so that the accident audio is played back synchronously when the accident image is played back.

The video generation model 34 searches the data in the video DB 33 using the accident scene ID input from the accident estimation model 13, and extracts the accident video corresponding to the near-miss information. The video generation model 34 uses this extracted accident video as a simulation video (image) of an accident scene that is likely to occur in the near-miss situation encountered by vehicle 2. Note that the video of the near-miss encountered by vehicle 2 may be joined with the simulation video of this accident scene to create a simulation video of the accident scene. In this case, the image contains real images (video), which increases the sense of realism and makes the image more effective.

Second accident video generation method: The first accident video generation method stores accident video for each type of accident in the video DB 33, taking into account the surrounding environment such as the background of the accident location, but this has the drawback of resulting in a large amount of data. The second accident video generation method generates accident video by combining video of the accident itself, that is, video of the accident vehicle and accident objects (people, objects, etc.), which are the objects involved in the accident, and background video, in this case a database of unchanging video (unchanging video over the duration of the accident, for example, images of still objects such as buildings) and changing video (people, animals, displays and traffic lights (changes due to displayed images or flashing)).

For example, the video generation model 34 performs image recognition on the background image of stationary objects in the near-miss information captured by the vehicle-mounted camera 24, such as information about the intersection shape, and based on the recognized information, selects the near-miss location and background image, such as a background image with a similar intersection shape, from the video DB 33. Information about the intersection shape here includes, for example, information about the road configuration, such as a T-junction, the number of lanes, and information indicating whether or not there are traffic lights. Note that a stationary object can be determined, for example, by comparing the moving speed of the object in the video with the moving speed of the vehicle filming it (if the moving speeds of both are the same, it is a stationary object).

In addition, the video generation model 34 performs image recognition on the video from the in-vehicle camera 24 to determine the presence or absence of people believed to have been involved in the near miss, as well as party information relating to their location, and, based on the recognized party information, selects accident video (video of the parties involved excluding the background, etc.) from the video DB 33 that contains party information similar to the situation of the parties involved in the near miss.

The video generation model 34 also performs image recognition on the video from the in-vehicle camera 24 to determine whether or not there are any vehicles believed to have been involved in the near-miss, as well as information regarding their locations. Based on the recognized involved vehicle information, the video generation model 34 selects accident video (video of the involved vehicle not including the background, etc.) from the video DB 33 that has information similar to that of the involved vehicle in the near-miss situation. The video generation model 34 then outputs a video that combines the selected background video, video of the parties involved, and video of the involved vehicle to the information output device 15, for viewing by, for example, the driver of the party involved in the near-miss.

In the case of this second accident video generation method, the video DB 33 will include a database for selecting background images (a database of information associating accident background image elements (intersection information, etc.) with background images for generating simulation images), a database for selecting images of parties involved in the accident (a database of information associating accident party elements (location information of each party, etc.) with images of parties involved for generating simulation images), and a database for selecting related vehicle images (a database of information associating accident-related vehicle elements (location information of each related vehicle, etc.) with related vehicle images for generating simulation images).

In the case of video, the video generation model 34 may output to the information output device 15 a video generated by stitching a near-miss image with a simulated accident image (splicing the near-miss image with the simulation image just before the avoidance action is taken in the near-miss image).

In this way, the accident estimation device 1 can help drivers understand the horror of accidents that can occur from near-miss situations they have experienced themselves, thereby contributing to improving the effectiveness of traffic safety education.

The video generation model 34 may also be configured to output video to the information output device 15 by superimposing a real-life scenery image (a real-life photographed image or a real-life scenery image of the accident location included in the map data) at the time of the near-miss onto the accident object (images of the parties involved and related vehicle images (no background image)) in the accident video selected from the video DB 33. This allows the scenery and other aspects of the simulation video to appear as real-life images, thereby increasing the impact on the viewer.

Third accident video generation method: The second accident video generation method requires less data than the first accident video generation method, but because video of the parties involved in the accident and accident-related vehicles is stored in the video DB 33, there is the issue of a large data volume. The third accident video generation method generates images of the accident subject objects (parties involved and related vehicles) based on physical characteristics estimated by image recognition processing of images taken at the time of the near miss. The generated images of the accident vehicles and accident objects (people, objects, etc.) are then combined with background video to generate an accident image.

Specifically, the position, speed of movement, and weight of the parties involved in the near miss and related vehicles are estimated based on images taken at the time of the near miss. Note that position and speed of movement are estimated based on the detection of each object using image recognition, and the detected position and position change. Weight is estimated by detecting the type and size of each object using image recognition, and estimating the weight of each object based on the detected type and size. Note that the weight of each object can be estimated, for example, using a database in which weight is associated with the type and size of the object.

Then, based on the estimated position, movement speed, and weight of each object, the laws of physics (mainly the laws of mechanics) are used to estimate the collision situation and post-collision damage of each object, and computer graphics technology is used to generate images of potential accidents that could occur from near misses.

Furthermore, the video DB 33 may store tables such as those shown in Figures 6 and 7, and accident images may be generated using the data in these tables. The table shown in Figure 6 associates accident scene IDs with various physical parameters in an accident. The various physical parameters in an accident are, for example, parameters indicating the vehicle involved, the relative speed of the vehicle at the time of the accident, the collision direction of the vehicle at the time of the accident, and the collision position of the vehicle at the time of the accident. The table shown in Figure 7 associates vehicle types with vehicle images.

When the tables shown in Figures 6 and 7 are stored in the video DB 33, the video generation model 34 generates a simulation video from near-miss information using various physical parameters of an accident similar to the video captured by the in-vehicle camera 24.

Specifically, the image generation model 34 estimates the movement and damage state of the target vehicle using various physical parameters in the accident, and generates an image of the accident target vehicle using CG. The image generation model 34 then superimposes this accident image on the scenery at the near miss (actually photographed image or scenery image included in map data) to generate a simulation image.

In this case, the image generation model 34 may be configured to read image data of the vehicle model of the target vehicle from the table shown in Figure 7 and use it for CG. The color of the vehicle may be included in the vehicle model, or may be colored into the vehicle image.

The image generation model 34 may also be configured with an image information generation AI model 34A (see Figure 9) that generates the above-mentioned simulation image from a near-miss image. The image information generation AI model 34A will be described later with reference to Figures 9 to 15.

In addition, the video generation model 34 (video information generation AI model 34A) generates and outputs a simulated video of the accident scene that includes at least one of the following information: the driver's own injury situation in the accident, images of personal injury, property damage, and images of property damage. This allows the accident estimation device 1 to help drivers more realistically understand the horror of accidents that can occur from near-miss situations, thereby contributing to improving the effectiveness of traffic safety education.

However, depending on the content of the accident scene simulation video that drivers are shown, it may not be possible to fully improve the effectiveness of traffic safety education. For example, if the content of the simulation video is too extreme, drivers with a low tolerance for extreme content may turn away from the simulation video and fail to understand its importance.

The video generation model 34 therefore processes the content of the simulation video to be viewed next by the same viewer, depending on the viewer's reaction to the accident scene simulation video. Alternatively, the content of the simulation video can be processed by providing a separate image processing unit downstream of the video information generation unit 31 that processes images according to the viewer's reaction (tolerance, described below).

When a driver is viewing a video for the first time, the video information generation unit 31 causes the driver to view a simulation video with the degree of extremity set to a default value. In this case, the video information generation unit 31 generates and causes the driver to view a simulation video that is, for example, CG-only video rather than live-action video, and cuts out scenes of contact with people. The video information generation unit 31 also generates and causes the driver to view a simulation video in which the model and color of the vehicle 2 that appears are different from those of the vehicle that appeared in the near-miss situation. In other words, the video information generation unit 31 adjusts the degree of reproduction (similarity) of the generated image to the actual image, thereby adjusting the intensity of the stimulation of the generated image (the extremity of the image).

The video information generation unit 31 then acquires images of the viewer who has watched the simulation video from the information output device 15, and determines the viewer's tolerance for extreme content using the tolerance determination unit 35. The video information generation unit 31 may also acquire the viewer's heart rate from the information output device 15 and determine tolerance based on the heart rate as well.

If the tolerance is below the threshold, for example, if it recognizes from an image that the viewer is not paying attention to the simulation video, the video information generation unit 31 will reduce the level of extremism in the next simulation video to be viewed. If the tolerance is above the threshold, for example, if it recognizes from an image that the viewer is paying attention to the simulation video, the video information generation AI model 34A will increase the level of extremism in the next simulation video to be viewed.

When reducing the level of extremism, the video information generation unit 31 performs image processing such as gradually reducing the resolution of the video or applying a blurring process. When increasing the level of extremism, the video information generation unit 31 performs image processing such as increasing the resolution of the video. Other methods that can be applied include adjusting the ratio of real images in the image (when reducing the level of extremism, reduce the ratio of real images and increase the ratio of CG), or adjusting the ratio of video to still images (when reducing the level of extremity, reduce the ratio of video and increase the ratio of still images).

Furthermore, when increasing the level of extremeness, the video information generation AI model 34A creates simulation images (video or still images) that appear more realistically, such as by allowing viewers to watch extreme scenes, such as scenes involving contact with people, without cutting them out, or by making the model and color of the vehicle 2 that appears the same as the vehicle that appeared in the near-miss situation.

As a result, the accident estimation device 1 can contribute to improving traffic safety education by providing simulated images of accident scenes that can be reliably viewed by various drivers with different tolerances for extreme situations, according to their tolerance level.

Next, the operation of the damage information generation unit 32 will be described. Based on the accident scene ID of the accident scene input from the accident estimation model 13 and the accident DB 12, the damage information generation unit 32 generates and outputs damage information regarding damages caused by an accident when an actual accident occurs from a near miss situation.

As a result, the accident estimation device 1 can help drivers involved in near-misses recognize the damage they would suffer if an actual accident were to occur from a near-miss situation, encouraging them to drive safely and thereby contributing to improving the effectiveness of traffic safety education.

As shown in Figure 8, the accident DB 12 stores a table that associates accident scene IDs with damage (personal injury and property damage compensation, repairs, and insurance) information. Damage (personal injury and property damage compensation, repairs, and insurance) information includes information such as the liability ratio in the event of the accident, the amount of damage suffered by the driver, and changes in insurance grade and insurance premiums after insurance is applied to the accident. In other words, this damage (repair and insurance) information can be used to calculate insurance premiums, and can be considered insurance premium calculation information for calculating insurance premiums.

For example, when an accident scene ID is input from the accident estimation model 13, the damage information generation unit 32 searches and reads out the damage information associated with that accident scene ID from the accident DB 12, and outputs the read damage information as accident information.

By providing such damage information to drivers, the accident estimation device 1 helps drivers recognize the damage situation after an accident, such as increased vehicle repair costs and insurance costs, and encourages safe driving, thereby contributing to improving the effectiveness of traffic safety education.

Furthermore, the information stored in the accident DB 12 shown in Figure 8 is generated based on cases of accidents that have actually occurred in the past. In other words, the damage information generation unit 32 generates accident information based on cases that have occurred in the past. This allows the damage information generation unit 32 to generate more realistic accident information for accidents that arise from near-miss situations and provide it to the driver. Note that while the table shown in Figure 8 stores damage information in association with accident scene IDs, it is also possible to store damage information in association with data that characterizes the accident scene, for example, physical characteristic data of the parties involved and related vehicles.

The damage information generation unit 32 may also be configured with a damage information generation AI model 32A (see Figure 9) that generates the above-mentioned damage information from near-miss images. Next, with reference to Figures 9 to 15, we will explain the accident information generation unit 14, which uses AI to generate simulation footage of an accident that occurs from a near-miss situation and the above-mentioned damage information.

As shown in FIG. 9, the accident information generation unit 14 includes a video information generation unit 31 and a damage information generation unit 32. The video information generation unit 31 includes a video information generation AI model 34A. The damage information generation unit 32 includes a damage information generation AI model 32A.

The video information generation AI model 34A is an AI model that has been trained to estimate the accident scene that will occur from the circumstances of the near-miss when near-miss information is input. For example, a training dataset such as that shown in Figure 10 is used for the machine learning of the video information generation AI model 34A.

As shown in Figure 10, the learning dataset for the video information generation AI model 34A is a dataset in which a plurality of near-miss images serving as input data are each associated one-to-one with a plurality of accident images serving as ground truth data.

As shown in Figure 11, the video information generation AI model 34A sequentially estimates simulation videos of potential accident scenes from the near-miss information input data that is sequentially input, and outputs them as estimation results. The video information generation unit 31 causes the video information generation AI model 34A to perform machine learning by updating the weighting coefficients of each layer in the DNN using processing such as backpropagation, so as to reduce the error between the estimated results of the simulation videos output from the video information generation AI model 34A and the correct values of the learning data corresponding to the input values.

As a result, the video information generation unit 31 can generate a video information generation AI model 34A that can estimate, from the input near-miss information, a simulation video of an accident scene that is likely to occur in the near-miss situation corresponding to the near-miss information.

As shown in Figure 12, when an actual near-miss image is input, the machine-learned video information generation AI model 34A estimates and outputs a simulation video of an accident that occurs from the near-miss situation.

Furthermore, the video information generation AI model 34A may be configured to add image processing parameters corresponding to the viewer's reaction (the aforementioned tolerance) to the data for generating a simulation image, and perform image processing according to those parameters. In this case, for example, viewer reaction data may be added to the input data of the learning data, and the correct answer data may be image data that also corresponds to the viewer's reaction.

As a result, when near-miss information and viewer images are input, the video information generation AI model 34A can process the content of the simulation video to be viewed next by the same viewer based on the viewer's reaction to the simulation video of the accident scene. It is also possible to implement this configuration by providing a separate image processing unit downstream of the video information generation AI model 34A that processes images based on the viewer's reaction (the aforementioned tolerance).

Furthermore, the damage information generation AI model 32A is an AI model that has been machine-trained to estimate damage information relating to damage caused by an accident when near-miss information is input and an actual accident occurs based on the circumstances of the near-miss. For example, a learning dataset such as that shown in FIG. 13 is used for the machine learning of the damage information generation AI model 32A.

As shown in Figure 13, the learning dataset for the damage information generation AI model 32A is a dataset in which a plurality of near-miss images serving as input data are each associated one-to-one with a plurality of pieces of damage information serving as correct answer data.

As shown in Figure 14, the damage information generation AI model 32A sequentially estimates damage information related to damage caused by an accident when an actual accident occurs, from the near-miss information of the input data that is sequentially input, and outputs the estimated results. The damage information generation unit 32 causes the damage information generation AI model 32A to perform machine learning by updating the weighting coefficients of each layer in the DNN using processing such as backpropagation so as to reduce the error between the estimated damage information output from the damage information generation AI model 32A and the correct value of the learning data corresponding to the input value.

As a result, the damage information generation unit 32 can generate a damage information generation AI model 32A that can estimate damage information related to damage caused by an accident when an actual accident occurs, from the input near-miss information.

As shown in Figure 15, when an actual near-miss image is input, the machine-learned damage information generation AI model 32A estimates and outputs damage information regarding the damage that will result from an accident if an actual accident occurs based on the near-miss situation.

[3. Processing Executed by the Controller of the Accident Estimation Device]
Next, a process executed by the controller 10 of the accident estimation device 1 will be described with reference to Fig. 16 and Fig. 17. Fig. 16 and Fig. 17 are flowcharts showing an example of a process executed by the controller 10 of the accident estimation device 1 according to the embodiment. This process is repeatedly executed at appropriate intervals after the accident estimation device 1 is started up (the power switch is turned on).

As shown in FIG. 16, the controller 10 determines whether or not near-miss information has been received from the vehicle 2 (step S101). If the controller 10 determines that near-miss information has not been received (step S101, No), it repeats the determination process of step S101 until near-miss information is received.

If the controller 10 determines that near-miss information has been received (step S101, Yes), it uses the accident estimation model 13 to estimate an accident scene that is likely to occur in the near-miss situation (step S102).

Next, the controller 10 executes an accident information generation process (step S103). Through this accident information generation process, the controller 10 generates damage information and a simulation video for the estimated accident scene.

Then, the controller 10 determines whether it is time to output accident information (step S104). If the accident estimation device 1 is installed in the vehicle 2, the timing to output the accident information is, for example, when the vehicle 2 stops, or when the vehicle 2 completes one trip of travel (for example, when it stops after arriving at the destination). In this case, the accident estimation device 1 can determine whether the vehicle 2 has stopped or completed one trip of travel based on vehicle speed data from the vehicle (determined by a vehicle speed of 0 for a longer period than the stop determination threshold), a destination arrival signal from the navigation device (determined by a match between the destination data and the current location), etc.

Furthermore, if the accident estimation device 1 is installed in a location other than the vehicle 2, the timing for outputting the accident information is the timing when the user of the accident estimation device 1 performs an operation to request the output of the accident information. In this case, the accident estimation device 1 is configured to acquire, from an operation device that accepts user operations, information indicating that an operation to request the output of the accident information has been performed.

If the controller 10 determines that it is not time to output the accident information (step S104, No), it repeats the determination process of step S104 until it is time to output the accident information. Then, if the controller 10 determines that it is time to output the accident information (step S104, Yes), it outputs the accident information including damage information and a simulation video of the accident (step S105).

Then, the controller 10 determines whether or not the viewer's video of the simulation video has been input (step S106). If the controller 10 determines that the viewer's video has not been input (step S106, No), it repeats the determination process of step S106 until the viewer's video is input.

If the controller 10 determines that the viewer's video has been input (step S106, Yes), it determines the viewer's tolerance for the extremeness of the simulation video from the input video (step S107).

If heart rate information is used to determine the viewer's tolerance, step S106 determines the viewer's heart rate input, and step S107 determines the tolerance based on the viewer's heart rate.

Then, the controller 10 changes the level of extremism of the accident simulation video to be output next, based on the determined tolerance (step S108), and ends the process. Note that if the tolerance determined in step S107 is the same as the tolerance for extremism of the previous simulation video, the controller 10 does not change the level of extremism.

Next, the details of the accident information generation process in step S103 will be explained using Figure 17. When the accident information generation process starts, the controller 10 generates damage information based on the accident scene estimated based on the near-miss information (step S201).

The controller 10 generates information indicating the estimated percentage of fault between the driver and the other party in the accident scene, information indicating at least one of the personal injury status and property damage status, and information indicating future changes (trends) in insurance premiums due to the occurrence of the accident. In addition, damage information may be generated from images of personal injury status and property damage status, which can be generated based on the accident simulation image described below (by cutting out images of people or damaged parts of the vehicle, for example).

Furthermore, the controller 10 reads the viewer's tolerance for the level of extremeness of the simulation video previously viewed, and generates a simulation image (video or still image) of an accident of an extreme level according to the tolerance level using the video generation model 34 (step S202), and then ends the process. In other words, the accident information generation process of step S103 ends. Step S104 shown in Figure 16 is then executed.

[4. Application examples of accident estimation device]
The accident estimation device 1 according to the embodiment can be applied to various devices. Application examples of the accident estimation device will be described below.

[4-1. Example of application to insurance premium review device]
The insurance premium review device is used to create insurance premium calculation standards and calculate insurance premiums for each individual. Currently, insurance premiums are calculated based on data related to accidents that have occurred. Therefore, insurance premiums are calculated based only on actual data, and do not include information on potential accidents. On the other hand, since near misses are potential accidents, when applied to an insurance premium review device, the accident estimation device 1 can estimate the damage from potential accidents based on the occurrence of near misses and the magnitude (damage) of the accidents that lead to them, and reflect this in the insurance premiums.

In other words, for example, an insurance premium calculation standard (for example, near-miss occurrence status data for an appropriate period is added to the parameters) would be created based on the total amount of damages caused by actual accidents and the amount of damages caused by accidents resulting from near-misses (multiplied by an appropriate coefficient (big data is required (data collected over an appropriate period))), and the insurance premium for each individual would be calculated based on each individual's data (near-miss occurrence status data for an appropriate period is used to calculate the insurance premium).

[4-2. Application example to in-vehicle equipment]
The accident estimation device 1 may be mounted on an on-board device such as a car navigation device or a drive recorder. In this case, when a near-miss occurs, the on-board device estimates an accident that is likely to occur in the near-miss situation based on the near-miss situation, and notifies the vehicle occupants of accident information related to the accident.

For example, the in-vehicle device notifies the occupants of accident information when the vehicle comes to a stop or when one trip has ended. This allows the in-vehicle device to notify the occupants of accident information about an accident that is likely to occur as a result of a near miss when the vehicle comes to a stop or when one trip has ended, in the event of a near miss occurring, without much time having passed since the near miss. Therefore, by providing the occupants involved with accident information about an accident that is likely to occur as a result of a near miss that they recently experienced, the in-vehicle device can encourage safe driving and contribute to improving the effectiveness of traffic safety education.

[4-3. Example of application to terminal device]
The accident estimation device 1 may be mounted on a terminal device installed in an institution that provides traffic safety education, such as a driving school or a driver's license testing center. In this case, the terminal device estimates an accident that is likely to occur in a near-miss situation based on the near-miss situation, and outputs traffic safety education teaching material information corresponding to the accident.

This allows the program to provide a wide range of accident-related information in the form of highly stimulating actual accidents, including not only actual accidents but also accidents that can occur in various situations, such as accidents that follow on from near misses. This means that the information provided will have a strong impact on participants, contributing to improving the effectiveness of traffic safety education.

For example, the terminal device can estimate the type of accident that is likely to occur based on the near miss that the participant has actually experienced, and provide a simulation video showing the process from the near miss to the actual accident as educational material for traffic safety education.

In this way, the terminal device can help drivers understand the horror of accidents that can occur from near-miss situations experienced by the trainees themselves, thereby contributing to improving the effectiveness of traffic safety education.

[5. Information Processing System]
Next, an information processing system according to an embodiment will be described. Fig. 18 is an explanatory diagram of an information processing system 100 according to an embodiment. As shown in Fig. 18, the information processing system 100 includes an accident estimation device 1, an insurance premium review device 101, an on-board device 102, a terminal device 103, and a vehicle 2.

The accident estimation device 1, insurance premium review device 101, on-board device 102, terminal device 103, and vehicle 2 are connected via a communications network N so that information can be communicated. The insurance premium review device 101, on-board device 102, and terminal device 103 are each equipped with the accident estimation device 1.

As a result, the information processing system 100 can provide the above-mentioned accident information at any of the locations where the accident estimation device 1 is installed, the location where the insurance premium review device 101 is installed, inside the vehicle where the onboard device 102 is installed, and the location where the terminal device 103 is installed. Therefore, by providing accident information to users in various locations, the information processing system 100 can contribute to improving the effectiveness of traffic safety education.

[6. Notes]
As an appendix, the features of the present invention are as follows.
(1)
Obtain near miss information about the near miss that occurred,
outputting accident type information for the near miss that has occurred by referring to an accident type database that associates the near miss information with an accident type relating to an accident that is feared to occur from the near miss;
An accident estimation device including a controller.
(2)
The controller
By referring to an accident database in which the output accident type information is associated with virtual accident information indicating an accident event that may occur due to the accident, the virtual accident information for the near miss that has occurred is output.
The accident estimation device according to (1) above.
(3)
The controller
Instead of the accident type database, the near-miss information is used as input information, and an accident type model trained with learning data in which the accident type information is correct answer data is used to output the accident type information for the near-miss that has occurred.
The accident estimation device according to (1) above.
(4)
The controller
Instead of the accident type database, the near-miss information is used as input information, and an accident type model trained with learning data in which the accident type information is correct answer data is used to output the accident type information for the near-miss that has occurred.
The accident estimation device according to (2) above.
(5)
The virtual accident information
Includes simulated images of virtual accidents,
The accident estimation device according to (2) or (4).
(6)
The virtual accident information
Contains hypothetical damage information regarding damages resulting from hypothetical accidents;
The accident estimation device according to (2) or (4).
(7)
The virtual damage information
Including information showing the percentage of fault between us and the other party in the accident,
The accident estimation device according to (6) above.
(8)
The virtual damage information
information indicating at least one of a personal injury situation, an image of the personal injury situation, a property damage situation, and an image of the property damage situation in the virtual accident;
The accident estimation device according to (6) or (7) above.
(9)
The virtual damage information
Contains information showing the change in premiums resulting from the occurrence of hypothetical accidents;
The accident estimation device according to (6), (7) or (8).
(10)
Obtain near miss information about the near miss that occurred,
The acquired near-miss information is input into a virtual accident model trained with learning data in which near-miss information on the occurrence status of the near-miss is used as input information and virtual accident information on accidents that are feared to occur as a result of the near-miss is used as correct answer information, thereby outputting the virtual accident information.
Accident estimation device.
(11)
Obtain near miss information about the near miss that occurred,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
Calculating insurance premium calculation information related to the calculation of insurance premiums based on the hypothetical accident information;
outputting the calculated insurance premium calculation information;
Insurance premium review device.
(12)
Obtain near miss information about the near miss that occurred,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
notifying a vehicle occupant of the virtual accident information;
Onboard machine.
(13)
Obtain near miss information about the near miss that occurred,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
Outputting information on teaching materials for traffic safety education including virtual accident information.
Terminal device.
(14)
The virtual accident information
This is a simulation video showing the situation of an accident that may occur due to a near miss.
The terminal device according to (13) above.
(15)
An information processing system having an accident estimation device and an on-board device mounted on a vehicle,
The accident estimation device
Acquire near-miss information regarding the near-miss that has occurred from the vehicle,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
transmitting the calculated hypothetical accident information to the in-vehicle device;
The in-vehicle device
When a near miss occurs, near miss information is collected from the vehicle's sensors,
The collected near-miss information is transmitted to the accident estimation device,
receiving the hypothetical accident information transmitted from the accident estimation device;
notifying a vehicle occupant of the received virtual accident information;
Information processing system.
(16)
An information processing system having an accident estimation device, an in-vehicle device mounted on a vehicle, and an insurance premium review device,
The in-vehicle device
When a near miss occurs, near miss information is collected from the vehicle's sensors,
The collected near-miss information is transmitted to the accident estimation device,
The accident estimation device
Acquire near-miss information regarding the near-miss that has occurred from the vehicle,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
The calculated hypothetical accident information is transmitted to the insurance premium examination device,
The insurance premium review device
receiving the hypothetical accident information transmitted from the accident estimation device;
Calculating insurance premium calculation information related to the calculation of insurance premiums based on the received hypothetical accident information;
outputting the calculated insurance premium calculation information;
Information processing system.
(17)
The computer
Obtain near miss information about the near miss that occurred,
outputting accident type information for the near miss that has occurred by referring to an accident type database that associates the near miss information with an accident type relating to an accident that is feared to occur from the near miss;
Accident estimation method.
(18)
Obtain near miss information about the near miss that occurred,
a computer is caused to execute a procedure of outputting accident type information for the near miss that has occurred by referring to an accident type database that associates the near miss information with an accident type relating to an accident that is feared to occur from the near miss, using the acquired near miss information;
Accident estimation program.
(19)
The computer
Near miss information on the occurrence of near misses is used as input information, and learning information is used to train an AI (Artificial Intelligence) model, with the correct answer being the accident type of the accident that is feared to occur from the near miss.
AI model generation method.
(20)
A computer is caused to execute a procedure for making an AI (Artificial Intelligence) model learn using learning information in which near-miss information relating to the occurrence status of near-misses is used as input information and the accident types of accidents that are feared to occur as a result of the near-misses are used as correct answer information.
AI generation program.

Further advantages and modifications may readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and equivalents thereof.

REFERENCE SIGNS LIST 1 Accident estimation device 2 Vehicle 10 Controller 11 Communication unit 12 Accident DB
13 Accident estimation model 13A Accident estimation AI model 14 Accident information generation unit 15 Information output device 20 Drive recorder 21 Acceleration sensor 22 Steering angle sensor 23 Emotion sensor 24 In-vehicle camera 25 In-vehicle microphone 26 Flash memory 27 Controller 28 Communication unit 29 Near-miss detection unit 31 Video information generation unit 32 Damage information generation unit 32A Damage information generation AI model 33 Video DB
34 Image generation model 34A Image information generation AI model 100 Information processing system 101 Insurance premium review device 102 In-vehicle device 103 Terminal device N Communication network

Claims (20)

Obtain near miss information about the near miss that occurred,
outputting accident type information for the near miss that has occurred by referring to an accident type database that associates the near miss information with an accident type relating to an accident that is feared to occur from the near miss;
An accident estimation device including a controller. The controller
By referring to an accident database in which the output accident type information is associated with virtual accident information indicating an accident event that may occur due to the accident, the virtual accident information for the near miss that has occurred is output.
The accident estimation device according to claim 1 . The controller
Instead of the accident type database, the near-miss information is used as input information, and an accident type model trained with learning data in which the accident type information is correct answer data is used to output the accident type information for the near-miss that has occurred.
The accident estimation device according to claim 1 . The controller
Instead of the accident type database, the near-miss information is used as input information, and an accident type model trained with learning data in which the accident type information is correct answer data is used to output the accident type information for the near-miss that has occurred.
The accident estimation device according to claim 2 . The virtual accident information
Includes simulated images of virtual accidents,
The accident estimation device according to claim 2 or 4. The virtual accident information
Contains hypothetical damage information regarding damages resulting from hypothetical accidents;
The accident estimation device according to claim 2 or 4. The virtual damage information
Including information showing the percentage of fault between us and the other party in the accident,
The accident estimation device according to claim 6. The virtual damage information
information indicating at least one of a personal injury situation, an image of the personal injury situation, a property damage situation, and an image of the property damage situation in the virtual accident;
The accident estimation device according to claim 6. The virtual damage information
Contains information showing the change in premiums resulting from the occurrence of hypothetical accidents;
The accident estimation device according to claim 6. Obtain near miss information about the near miss that occurred,
The acquired near-miss information is input into a virtual accident model trained with learning data in which near-miss information on the occurrence status of the near-miss is used as input information and virtual accident information on accidents that are feared to occur as a result of the near-miss is used as correct answer information, thereby outputting the virtual accident information.
Accident estimation device. Obtain near miss information about the near miss that occurred,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
Calculating insurance premium calculation information related to the calculation of insurance premiums based on the hypothetical accident information;
outputting the calculated insurance premium calculation information;
Insurance premium review device. Obtain near miss information about the near miss that occurred,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
notifying a vehicle occupant of the virtual accident information;
Onboard machine. Obtain near miss information about the near miss that occurred,
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
Outputting information on teaching materials for traffic safety education including virtual accident information.
Terminal device. The virtual accident information
This is a simulation video showing the situation of an accident that may occur due to a near miss.
The terminal device according to claim 13. An information processing system having an accident estimation device and an on-board device mounted on a vehicle,
The accident estimation device
Acquiring near-miss information relating to the near-miss that has occurred from the in-vehicle device;
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
transmitting the calculated hypothetical accident information to the in-vehicle device;
The in-vehicle device
When a near miss occurs, near miss information is collected from the vehicle's sensors,
The collected near-miss information is transmitted to the accident estimation device,
receiving the hypothetical accident information transmitted from the accident estimation device;
notifying a vehicle occupant of the received virtual accident information;
Information processing system. An information processing system having an accident estimation device, an in-vehicle device mounted on a vehicle, and an insurance premium review device,
The in-vehicle device
When a near miss occurs, near miss information is collected from the vehicle's sensors,
The collected near-miss information is transmitted to the accident estimation device,
The accident estimation device
Acquiring near-miss information relating to the near-miss that has occurred from the in-vehicle device;
Based on the near miss information, hypothetical accident information regarding an accident that is feared to occur from the near miss is calculated;
The calculated hypothetical accident information is transmitted to the insurance premium examination device,
The insurance premium review device
receiving the hypothetical accident information transmitted from the accident estimation device;
Calculating insurance premium calculation information related to the calculation of insurance premiums based on the received hypothetical accident information;
outputting the calculated insurance premium calculation information;
Information processing system. The computer
Obtain near miss information about the near miss that occurred,
outputting accident type information for the near miss that has occurred by referring to an accident type database that associates the near miss information with an accident type relating to an accident that is feared to occur from the near miss;
Accident estimation method. Obtain near miss information about the near miss that occurred,
a computer is caused to execute a procedure of outputting accident type information for the near miss that has occurred by referring to an accident type database that associates the near miss information with an accident type relating to an accident that is feared to occur from the near miss, using the acquired near miss information;
Accident estimation program. The computer
Near miss information on the occurrence of near misses is used as input information, and learning information is used to train an AI (Artificial Intelligence) model, with the correct answer being the accident type of the accident that is feared to occur from the near miss.
AI model generation method. A computer is caused to execute a procedure for making an AI (Artificial Intelligence) model learn using learning information in which near-miss information relating to the occurrence status of near-misses is used as input information and the accident types of accidents that are feared to occur as a result of the near-misses are used as correct answer information.
AI generation program.