Information processing method, information processing device, and program

Abstract

The present disclosure relates to an information processing method, an information processing device, and a program which are capable of digitizing a risk in a vehicle. Provided is an information processing method that includes a computer acquiring an image captured by a camera attached to a vehicle, analyzing at least one of a road and traffic included in the acquired image to evaluate the risk of the vehicle colliding with another moving object, and setting an insurance premium for the vehicle on the basis of the evaluated risk. The present disclosure can be applied to, for example, a system for setting an insurance premium for automobile insurance.

Description

Information processing method, information processing device, and program

[0001] The present disclosure relates to an information processing method, an information processing device, and a program, and more particularly to an information processing method, an information processing device, and a program that are capable of quantifying risks in a vehicle.

[0002] Technologies for setting automobile insurance premiums have been proposed. For example, Patent Literature 1 discloses a technology for analyzing the driving behavior of a driver of a first vehicle, which is an insured vehicle, based on detection data from a rear distance sensor attached to a second vehicle, and setting the insurance premium for the automobile insurance contracted by the driver based on the driving behavior.

[0003] Japanese Patent Publication No. 2022-20918

[0004] In automobile insurance, premiums are generally set based on factors such as the color of a driver's license, accident history, and age, and are not based on an evaluation of the driver's driving behavior relative to road conditions. The technology disclosed in Patent Document 1 estimates insurance premiums by detecting the distance between a first vehicle and a second vehicle to evaluate whether the driver is driving the vehicle safely. However, this requires two vehicles, and the scope of evaluation is limited. Therefore, there has been a demand for a technology that quantifies vehicle risks so that insurance premiums can be estimated by evaluating a wider range of evaluation targets.

[0005] The present disclosure has been made in light of these circumstances, and makes it possible to quantify risks in vehicles.

[0006] An information processing method according to one aspect of the present disclosure is an information processing method including: a computer acquiring an image captured by a camera attached to a vehicle; analyzing at least one of roads and traffic contained in the acquired image to assess the risk of the vehicle colliding with another moving object; and setting an insurance premium for the vehicle based on the assessed risk.

[0007] An information processing device according to one aspect of the present disclosure is an information processing device that includes an acquisition unit that acquires images captured by a camera attached to a vehicle, an evaluation unit that analyzes at least one of roads and traffic contained in the acquired images to evaluate the risk of the vehicle colliding with other moving objects, and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.

[0008] A program according to one aspect of the present disclosure is a program that causes a computer to function as an acquisition unit that acquires images captured by a camera attached to a vehicle, an evaluation unit that analyzes at least one of roads and traffic contained in the acquired images to evaluate the risk of the vehicle colliding with other moving objects, and a setting unit that sets insurance premiums for the vehicle based on the evaluated risk.

[0009] In one aspect of the information processing method, information processing device, and program of the present disclosure, an image captured by a camera attached to a vehicle is acquired, and at least one of the road and traffic contained in the acquired image is analyzed to evaluate the risk of the vehicle colliding with another moving object, and an insurance premium for the vehicle is set based on the evaluated risk.

[0010] An information processing method according to one aspect of the present disclosure is an information processing method including, when a user sets a destination indicating where a vehicle is to be moved, a computer acquiring route information regarding a travel route that is a route traveled by the vehicle from its current location to the destination, analyzing the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and assessing the risk of the vehicle colliding with another moving object on the travel route, and setting an insurance premium for the vehicle based on the assessed risk.

[0011] An information processing device according to one aspect of the present disclosure is an information processing device that includes: an acquisition unit that, when a destination indicating the destination of a vehicle is set by a user, acquires route information regarding a travel route that is a route traveled by the vehicle from its current location to the destination; an evaluation unit that analyzes the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and evaluates the risk of the vehicle colliding with other moving objects on the travel route; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.

[0012] A program according to one aspect of the present disclosure is a program that causes a computer to function as an acquisition unit that, when a destination indicating the destination of a vehicle is set by a user, acquires route information regarding a travel route, which is the route traveled by the vehicle from its current location to the destination; an evaluation unit that analyzes the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and evaluates the risk of the vehicle colliding with other moving objects on the travel route; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.

[0013] In an information processing method, an information processing device, and a program according to one aspect of the present disclosure, when a user sets a destination indicating where a vehicle is to be moved, route information is obtained regarding a travel route, which is the route traveled by the vehicle from its current location to the destination, and the obtained route information is analyzed based on map information including at least one of road information regarding roads and statistical information regarding traffic to evaluate the risk of the vehicle colliding with other moving objects on the travel route, and an insurance premium for the vehicle is set based on the evaluated risk.

[0014] Note that the information processing device according to one aspect of the present disclosure may be an independent device or an internal block constituting a single device.

[0015] FIG. 10 is a diagram illustrating the quantification of Exposure when other traffic emerges from an unobservable area. FIG. 11 is a diagram illustrating the quantification of Exposure when a route overlaps with other traffic in an observable area. FIG. 12 is a diagram illustrating an example configuration of an embodiment of a system to which the present disclosure is applied. FIG. 13 is a block diagram illustrating an example detailed configuration of an edge device and a cloud server. FIG. 14 is a flowchart illustrating the flow of processing executed by an edge device. FIG. 15 is a flowchart illustrating the flow of processing executed by a risk assessment unit of the cloud server. FIG. 16 is a diagram illustrating a schematic representation of an example process of risk assessment. FIG. 17 is a flowchart illustrating the flow of processing executed by an insurance premium setting unit of the cloud server. FIG. 18 is a block diagram illustrating another example configuration of a system to which the present disclosure is applied. FIG. 19 is a flowchart illustrating the flow of processing executed by a route selection unit of the cloud server. FIG. 19 is a block diagram illustrating another example configuration of a system to which the present disclosure is applied. FIG. 19 is a flowchart illustrating the flow of processing executed by a risk assessment unit of the cloud server.

[0016] <Risk Indicators> The international basic safety standard IEC 61508 requires risk to be evaluated using three indices: the magnitude of damage (Severity), the probability of exposure (Exposure), and the possibility of avoidance (Controllability). Of these, Severity depends on the kinetic energy of the vehicle, while Controllability depends on the vehicle's speed. Therefore, statistical data from existing vehicles driven and operated by humans can be used for autonomous mobility and automated driving systems. On the other hand, there is no statistical data for exposure from existing vehicles driven and operated by humans. Therefore, quantifying exposure is important when estimating insurance plans.

[0017] Exposure does not represent the probability of a hazard occurring, but the probability that the prerequisites for a hazard will be met. Therefore, it is necessary to calculate the probability that there is a person on the route, not the probability of hitting a person on the route. There are two main scenarios where other traffic (cars, motorcycles, people, etc.) is on the route: when other traffic emerges from an unobservable area, and when the route overlaps with other traffic in an observable area. If the target vehicle (your vehicle) is considered a moving object, then other traffic can also be considered other moving objects.

[0018] Fig. 1 is a diagram for explaining the quantification of Exposure when other traffic emerges from an unobservable area. Fig. 1 shows a case where an area cannot be observed from a traveling vehicle 1 due to an obstacle 2 present around the vehicle 1. Fig. 1 schematically shows a situation when the vehicle 1 is moving in the direction indicated by arrow D11 in the figure, and Fig. 1A shows a situation earlier in time than Fig. 1B.

[0019] In FIG. 1A, as indicated by the dashed line L11 in the figure, an invisible area A11 exists, which is an area that cannot be observed from the vehicle 1 because a portion of the view is blocked by an obstacle 2. The probability that other traffic exists in the invisible area A11 can be calculated using statistical data acquired in advance, such as the traffic volume per unit time on the route (road). For example, if the route is a public road, public statistical data used in urban planning can be used. Furthermore, if the route is on private property or indoors, observed values ​​over a certain period of time or estimated values ​​based on architectural design can be used. Note that while FIG. 1A illustrates an example in which the view is blocked by a fixed obstacle 2, the same applies when the view is blocked by other traffic and an invisible area A11 exists.

[0020] Next, consider the rush-out area A12, which is an area where other traffic emerges from the invisible area A11. The size of the rush-out area A12 can be calculated from the speed and braking distance of the other traffic if the other traffic is a vehicle. Therefore, both the speed and braking distance can be estimated using statistical data obtained through observation, or by using the legal speed limit for the speed and braking distance for the braking performance. Furthermore, the probability that other traffic actually exists in the rush-out area A12 is an extension of the probability that other traffic exists in the invisible area A11. The overlapping area between the rush-out area A12 and the braking area A13, which is an area based on the braking distance of the vehicle 1, is the risk area A14, which is an area where there is a risk of collision.

[0021] In A of Figure 1, the rush-out area A12 and the braking area A13 do not overlap, but as the vehicle 1 moves in the direction indicated by arrow D11, the rush-out area A12 and the braking area A13 overlap in B of Figure 1, and this overlapping area becomes the risk area A14. By integrating the area of ​​the risk area A14 and the probability that other traffic is present over time while the vehicle 1 is moving, the risk borne by the vehicle 1 can be quantified. Note that the invisible area A11 can also be said to be an unobservable area, the rush-out area A12 an area where other moving objects may appear, and the braking area A13 a braking distance area.

[0022] Fig. 2 is a diagram for explaining the quantification of Exposure when the route of a vehicle 1 overlaps with the route of another vehicle 3 or 4 that is traveling in the area observable from vehicle 1. Fig. 2A and Fig. 2B respectively show cases where the route of a vehicle 1 overlaps with the route of another vehicle 3 or 4 that is traveling in the area observable from vehicle 1.

[0023] In FIG. 2A, another known vehicle 3 is present in an area observable from a traveling vehicle 1. The other vehicle 3 is traveling at a constant speed and direction of travel in the direction indicated by arrow D22 in the figure. In this situation, when vehicle 1 needs to cut in, overtake, pass, or cross other vehicles, and travels in the direction indicated by arrow D21 in the figure, it is assumed that the routes of vehicle 1 and the other vehicle 3 may overlap. In this case, the area where route area A21, which is an area according to the route of vehicle 1, overlaps with route area A22, which is an area according to the route of the other vehicle 3, can be defined as risk area A23, which is an area where there is a risk of collision. The area of ​​risk area A23 and the probability of the presence of other vehicles can be integrated over time while vehicle 1 is moving, thereby quantifying the risk posed to vehicle 1.

[0024] In B of Fig. 2, another vehicle 4 is present in an area observable from a traveling vehicle 1, but vehicle 1 is traveling in the direction indicated by arrow D31 in the figure, and the other vehicle 4 is traveling in the direction indicated by arrow D32 in the figure. In this case, in B of Fig. 2, as in A of Fig. 2, the area where route area A31, which is an area according to the route of vehicle 1, and route area A32, which is an area according to the route of the other vehicle 4, overlap is defined as risk area A33, and it is possible to estimate that there is a risk of collision. It should be noted that route areas A21 and A31 can also be said to be the route area of ​​the host vehicle, and route areas A22 and A32 can be said to be the route areas of the other moving object.

[0025] In this way, by determining risk areas when other traffic emerges from an unobservable area and when a route overlaps with other traffic in an observable area, the risk in the risk area can be quantified and Exposure can be determined. By quantifying Exposure, if the Exposure value is the same, the magnitude of risk is the same whether the vehicle 1 is driven or operated by a person or a system, and the same insurance plan (premium) can be estimated. This disclosure proposes a system that can quantify such risks and estimate insurance plans (premiums).

[0026] <System Configuration> Figure 3 is a diagram showing an example configuration of an embodiment of a system to which the present disclosure is applied. In Figure 3, a system 10 to which the present disclosure is applied is composed of edge devices 11-1 to 11-N (N: an integer greater than or equal to 1) and a cloud server 12. The edge devices 11-1 to 11-N are each mounted on a different vehicle 1. The edge devices 11-1 to 11-N can be connected to the cloud server 12 via a communication network 21.

[0027] The edge device 11-1 is a device equipped with an omnidirectional camera, an AI accelerator, and the like. The edge device 11-1 is mounted on the vehicle 1-1. The edge device 11-1 processes images of the periphery of the vehicle 1-1 and transmits the resulting transmission data to the cloud server 12 via the communication network 21. The edge device 11-2 is configured similarly to the edge device 11-1 and processes images of the periphery of the vehicle 1-2 in which the edge device 11-2 is mounted, and transmits the resulting transmission data to the cloud server 12 via the communication network 21. Similarly, the edge devices 11-3 to 11-N are mounted on different vehicles 1 and transmit the transmission data obtained by processing the images they have captured to the cloud server 12.

[0028] The cloud server 12 is a server built in a cloud environment by a cloud service provider. The cloud server 12 receives transmission data transmitted from each of the edge devices 11-1 to 11-N via the communication network 21. The cloud server 12 processes the transmission data received from the edge device 11-1 to evaluate the risk of the vehicle 1-1 colliding with other traffic, and sets an insurance premium for the vehicle 1-1 based on the evaluation results. The other traffic includes automobiles, motorcycles, pedestrians, etc. The cloud server 12 also evaluates the risk of the vehicle 1-2 colliding with other traffic based on the transmission data received from the edge device 11-2, and sets an insurance premium for the vehicle 1-2 based on the evaluation results. The cloud server 12 performs similar processing on the transmission data received from each of the edge devices 11-3 to 11-N, and sets insurance premiums for the vehicles 1-3 to 1-N.

[0029] In the following description, when there is no need to particularly distinguish between edge devices 11-1 to 11-N, they will be referred to as edge device 11. Furthermore, when there is no need to particularly distinguish between vehicles 1-1 to 1-N, they will be referred to as vehicle 1. Vehicle 1 may be a vehicle driven and operated by a person, or an autonomous vehicle that travels automatically without being driven and operated by a person. The level of automation of the autonomous driving may be any level. Alternatively, vehicle 1 may be an autonomous driving robot such as an automatic delivery robot or a delivery robot. For example, an autonomous driving robot includes an autonomous driving robot that can travel on roads such as public roads.

[0030] Fig. 4 is a block diagram showing an example of a detailed configuration of the edge device 11 and the cloud server 12 in Fig. 3. In Fig. 4, the edge device 11 includes an imaging unit 111, an AI processing unit 112, a position estimation unit 113, a position information addition unit 114, and a communication unit 115.

[0031] The imaging unit 111 is configured with an omnidirectional camera or the like and can capture images in a 360-degree direction based on the location where the imaging unit 111 is installed. The imaging unit 111 captures images of the surroundings of the vehicle 1, generates image data, and supplies the image data to the AI ​​processing unit 112. The imaging unit 111 may not only capture images in all directions with a single camera and generate image data, but may also generate image data in a 360-degree direction by synthesizing images captured by multiple cameras. Alternatively, the imaging unit 111 is not limited to an omnidirectional camera and may be a camera capable of capturing images in a specific direction, such as the front, rear, or side of the vehicle 1. Images captured by a camera of another device, such as a camera mounted on an in-vehicle video recording device (drive recorder) installed in the vehicle 1, may also be used.

[0032] The AI ​​processing unit 112 is composed of an AI accelerator and the like. The AI ​​processing unit 112 applies semantic segmentation to the image data supplied from the imaging unit 111 and extracts road pixels from the image. Semantic segmentation is a deep learning algorithm that associates a label or category with each pixel in an image, making it possible to recognize groups of pixels that form distinctive categories. Here, the image may be converted into a vector image to reduce its size.

[0033] The AI ​​processing unit 112 also applies semantic segmentation to the image data to extract pixels of other traffic (moving objects) from the image. The AI ​​processing unit 112 calculates (or estimates) information about the extracted other traffic. The information about the other traffic includes the speed and direction of the other traffic, and the distance from the vehicle (vehicle 1). For example, the AI ​​processing unit 112 can calculate information such as the speed and direction of the other traffic by acquiring speed information (information about the speed of vehicle 1) output from another device mounted on vehicle 1 and detecting how the other traffic extracted from the images input in a time series changes relative to vehicle 1 moving at the speed indicated by the acquired speed information.

[0034] Note that the speed information is an example of output information output from another device, and other information may be used as long as it can be used to calculate the speed and direction of other traffic, the distance from the vehicle 1, etc. For example, it may be detection information detected by various sensors mounted on the vehicle 1. The AI ​​processing unit 112 may recognize roads and other traffic from the image using other recognition processing, not limited to semantic segmentation. The AI ​​processing unit 112 supplies the processed data to the location information addition unit 114.

[0035] The position estimation unit 113 is configured with a device or the like having a function of estimating odometry information of the vehicle 1. The odometry information is the result of estimating the position and attitude of the vehicle 1 based on output information from equipment (e.g., sensors such as a global positioning system (GPS) and an inertial measurement unit (IMU)) attached to measure the behavior of the vehicle 1. Note that GPS is an example of a positioning system, and other positioning systems may also be used. The position estimation unit 113 estimates the odometry information of the vehicle 1 to obtain position information indicating the position of the vehicle 1, and supplies the position information to the position information addition unit 114.

[0036] The location information adding unit 114 adds the location information supplied from the location estimation unit 113 to the semantic data included in the data supplied from the AI ​​processing unit 112. For example, the location information is a geotag, which is metadata indicating latitude and longitude, and the geotag is embedded in the semantic data. In addition to the semantic data obtained by applying semantic segmentation, the data from the AI ​​processing unit 112 also includes other information such as the speed and direction of traffic. Data including this information and the semantic data with the location information added is supplied from the location information adding unit 114 to the communication unit 115.

[0037] The communication unit 115 is configured with a communication device or the like having a communication function using a mobile communication network such as 5G (5th Generation: fifth generation mobile communication system). The communication unit 115 is connected to a communication network 21 external to the vehicle 1 and communicates with external devices including the cloud server 12 via the communication network 21. The communication unit 115 transmits the data supplied from the location information addition unit 114 as transmission data to the cloud server 12 via the communication network 21.

[0038] 4, the cloud server 12 includes a map information storage unit 121, a communication unit 122, a processing unit 123, and an evaluation value storage unit 124. Map information is prepared in advance for operation of the system 10. The map information includes road information, including the structure of roads in a specific area, and statistical data (statistical information) related to traffic. The road information includes information on the width, shape, and direction of traffic of roads in the specific area. The statistical data includes information on the traffic volume per unit time and vehicle speed (vehicle speed) in the specific area. The specific area includes at least the area in which the vehicle 1 travels. For example, when providing insurance services within Japan, map information covering the entire country of Japan can be prepared.

[0039] The map information storage unit 121 is configured with a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The map information storage unit 121 stores map information in advance and supplies it to the processing unit 123 as needed. While FIG. 4 shows a configuration in which the map information storage unit 121 on the cloud server 12 side stores map information, the map information may also be stored on the edge device 11 side. When a storage unit (not shown) of the edge device 11 stores map information, the map information corresponding to the area in which the vehicle 1 travels may be stored and transmitted to the cloud server 12 as needed.

[0040] The communication unit 122 is configured with a communication device such as a network card. The communication unit 122 is connected to the communication network 21 and communicates with external devices including the edge device 11 via the communication network 21. The communication unit 122 receives transmission data sent from the edge device 11 and supplies the data to the processing unit 123.

[0041] The processing unit 123 is configured with a processor such as a CPU (Central Processing Unit). The processing unit 123 includes a risk assessment unit 131 and an insurance premium setting unit 132. For example, the risk assessment unit 131 and the insurance premium setting unit 132 are realized by the CPU executing a program. The risk assessment unit 131 processes transmission data supplied from the communication unit 122 based on map information acquired from the map information storage unit 121, thereby assessing the risk of collision between the vehicle 1 and other traffic. For example, as described with reference to FIG. 1 , the risk assessment unit 131 assesses the risk of collision when other traffic emerges from an area that cannot be observed from the vehicle 1 (invisible area). Furthermore, as described with reference to FIG. 2 , the risk assessment unit 131 assesses the risk of collision when the route of the vehicle 1 overlaps with that of other traffic in an area that can be observed from the vehicle 1.

[0042] The risk assessment unit 131 records the assessment value obtained by assessing the risk in the assessment value storage unit 124. The insurance premium setting unit 132 reads the assessment value of the target vehicle 1 from the assessment values ​​recorded in the assessment value storage unit 124, and sets the insurance premium for the target vehicle 1 based on the read assessment value. The assessment value storage unit 124 is configured with a storage device such as an HDD or SSD. The assessment value storage unit 124 may be configured with the same storage device as the map information storage unit 121.

[0043] 3 and 4, the cloud server 12 has multiple functions, but a different server may be provided for each function of the cloud server 12. For example, in FIG. 4, the processing unit 123 of the cloud server 12 has a risk assessment unit 131 and an insurance premium setting unit 132, but the function of the insurance premium setting unit 132 may be provided in another server, so that the server that assesses the risk and the server that sets the insurance premium are different servers. Also, in FIG. 4, the cloud server 12 has a map information storage unit 121 and an evaluation value storage unit 124, but another server having a storage unit that records map information and evaluation values ​​may be provided.

[0044] <Processing in Edge Device> The flow of processing executed by the edge device 11 in FIG. 4 will be described with reference to the flowchart in FIG.

[0045] In step S11, the imaging unit 111 captures an image of the area around the vehicle 1. In step S12, the AI ​​processing unit 112 applies semantic segmentation to extract road pixels from the image captured by the imaging unit 111. In step S13, the AI ​​processing unit 112 applies semantic segmentation to extract pixels of other traffic from the image captured by the imaging unit 111. In step S14, the AI ​​processing unit 112 uses information such as the speed of vehicle 1 to calculate the speed and direction of the other traffic, and the distance between the other traffic and the vehicle (vehicle 1).

[0046] In step S15, the position estimation unit 113 estimates position information indicating the position of the vehicle 1. In step S16, the position information addition unit 114 adds position information such as a geotag to the semantic data. In step S17, the communication unit 115 transmits the semantic data with the added position information and transmission data including information such as the speed and direction of other traffic to the cloud server 12 via the communication network 21. When the processing of step S17 ends, the series of processes ends.

[0047] In this way, the edge device 11 processes images captured by a camera attached to the vehicle 1 moving from one point to another, and transmits the resulting transmission data to the cloud server 12 via the communication network 21. The timing of transmitting the transmission data can be set appropriately depending on the operating environment, such as when processing performed in predetermined units such as image frame units is completed.

[0048] <Processing in Cloud Server> The flow of processing executed by the risk assessment unit 131 of the cloud server 12 in FIG. 4 will be described with reference to the flowchart in FIG.

[0049] In step S21, the risk assessment unit 131 acquires transmission data transmitted from the edge device 11. In step S22, the risk assessment unit 131 acquires map information pre-recorded in the map information storage unit 121. Here, the risk assessment unit 131 can acquire map information of an area identified by the location information by using the location information added to the semantic data included in the transmission data. In step S23, the risk assessment unit 131 processes the transmission data based on the map information to assess the risk of a collision when other traffic emerges from an area that cannot be observed from the vehicle 1 (an invisible area).

[0050] Here, an example of a method for assessing the risk of a collision when other traffic emerges from an invisible area will be described with reference to Fig. 7 . As shown in Fig. 7 , first, a road structure included in a semantic image is subjected to coordinate transformation to correspond to the road structure included in map information (Coordinate transformation to map view). For example, if the road structure in the map information is viewed from above, a road image R11 extracted from image I1 is subjected to coordinate transformation to become a road image viewed from above (Road structure from map view) similar to the road structure in the map information. This results in a road image R12 after coordinate transformation.

[0051] When the edge device 11 is mounted on a vehicle 1, the installation position of the camera of the imaging unit 111 can be estimated from information about the vehicle model of the vehicle 1 and the car structure C11 included in the image I1 of the semantic data, thereby enabling coordinate transformation to be performed according to the installation position of the camera that captured the image. Note that, when the installation position of the camera is determined by design, such as in a robot, the design value can be used.

[0052] Next, by comparing the coordinate-transformed road image R12 with the road structure (Ideal Road Structure in Map Database) R21 of the map information, an invisible area A11, which is an area that cannot be observed from the vehicle 1, is estimated as shown in a risk map (Estimate Risk Map). Furthermore, the probability that other traffic is present in the invisible area A11 is calculated based on statistical data included in the map information, and a rush-out area A12, which is an area where other traffic will emerge from the invisible area A11, is estimated. The area where the rush-out area A12 overlaps with a braking area A13, which is an area corresponding to the braking distance of the vehicle 1, is estimated as a risk area A14, which is an area where there is a risk of the vehicle 1 colliding with other traffic. The risk assessment unit 131 calculates a value indicating the probability that other traffic is present in the risk area A14.

[0053] Here is a more specific example of risk assessment. Suppose the probability that there is a vehicle passing by on a road in a certain unobservable area (invisible area) per unit time is 0.1 vehicles / s (10%). In this case, if the average speed of other vehicles is 50 km / h (13.8 m / s), then it can be said that approximately 68% of vehicles are traveling at speeds within 50 km / h ±1 σ (1 standard deviation). Therefore, the probability that a vehicle traveling at the average speed is in the invisible area per unit time can be expressed as the product of these (approximately 6.8%). Probability can also be calculated in a similar manner for other speed ranges.

[0054] Furthermore, if we assume that this vehicle brakes in the normal deceleration range (0.2G ± 1σ), it can stop in approximately 10.9 m, which corresponds to the area where this braking distance occurs (the rush-out area). The collision speed at each point within the rush-out area can be mapped from speed and deceleration. Therefore, when a vehicle traveling at an average speed brakes in the average deceleration range, the collision probability per unit time is 4.6% (6.8% x 68%), and the collision speed at that time is as described above. It is also possible for vehicles to decelerate more rapidly (0.2G + 1σ or more), in which case the rush-out area is narrower and the collision probability is lower. On the other hand, if vehicles decelerate more slowly (0.2G + 1σ or less), the rush-out area is wider and the collision probability is higher. By performing this calculation for each speed range and deceleration range, we can map the relationship between the collision probability per unit time and collision speed at each point on the map.

[0055] Returning to the explanation of FIG. 6 , in step S24, the risk assessment unit 131 processes the transmitted data to assess the risk of collision when the route overlaps with that of other traffic in an area observable from vehicle 1. Here, for example, in the situation shown in A of FIG. 2 , based on information about the other traffic included in the transmitted data (such as the speed and direction of the other traffic, and the distance from the vehicle (vehicle 1)), the area where route area A21 corresponding to the route of vehicle 1 overlaps with route area A22 corresponding to the route of the other traffic is estimated as risk area A23, which is an area where there is a risk of vehicle 1 colliding with the other traffic. The risk assessment unit 131 calculates a value indicating the probability that other traffic exists in risk area A23.

[0056] In step S25, the risk assessment unit 131 records in the assessment value storage unit 124 an assessment value indicating the risk of collision when other traffic emerges from an area that cannot be observed from the vehicle 1, and an assessment value indicating the risk of collision when the route overlaps with that of other traffic in an area that can be observed from the vehicle 1. For example, the former assessment value may be a value indicating the probability that other traffic is present in the risk area A14, and the latter assessment value may be a value indicating the probability that other traffic is present in the risk area A23. These assessment values ​​are accumulated over time while the vehicle 1 is moving and recorded in the assessment value storage unit 124. Note that the assessment value may include the area of ​​the risk area A14 and the area of ​​the risk area A23. When the processing of step S25 ends, the series of processes ends.

[0057] Next, the flow of processing executed by the insurance premium setting unit 132 of the cloud server 12 in FIG. 4 will be described with reference to the flowchart in FIG.

[0058] In step S31, it is determined whether to estimate an insurance premium for the target vehicle 1, and if it is determined that an insurance premium should be estimated, the process proceeds to step S32. In step S32, the insurance premium setting unit 132 reads out the evaluation value for the target vehicle 1 stored in the evaluation value storage unit 124. In step S33, the insurance premium setting unit 132 sets the insurance premium (automobile insurance premium) for the target vehicle 1 based on the read evaluation value. When the process of step S33 ends, the series of processes ends.

[0059] In this way, the cloud server 12 uses the transmission data transmitted from the edge device 11 to evaluate the risk of the vehicle 1 colliding with other vehicles and stores the evaluation value, thereby quantifying the risk that the target vehicle 1 is taking while moving and determining Exposure. When estimating the insurance premium for the target vehicle 1, if the Exposure value is the same, the magnitude of the risk is the same regardless of whether the target vehicle 1 is driven or operated by a person or a system, and the same insurance plan (premium) can be estimated.

[0060] For example, an insurance company can use a cloud service (or a cloud server 12 provided by the cloud service) to build a system 10 to which the present disclosure is applied, thereby quantifying the risk associated with the vehicle and estimating an insurance plan, regardless of whether the vehicle 1 is being driven or operated by a person or a system. In other words, the system 10 to which the present disclosure is applied does not set insurance premiums based on the color of the driver's license, accident history, age, or the like, but instead evaluates the driving or operation itself in relation to road conditions, regardless of whether the driver or operator is a person or a system, and can set higher insurance premiums if the risk is high, or lower insurance premiums if the risk is low. Furthermore, the system 10 to which the present disclosure is applied can estimate insurance premiums for a single vehicle by evaluating a wide range of evaluation targets, not limited to inter-vehicle distance, thereby enabling more appropriate insurance premium estimates.

[0061] As described above, the system 10 to which the present disclosure is applied acquires an image captured by the imaging unit 111 attached to the target vehicle 1, analyzes at least one of the roads and traffic included in the acquired image, evaluates the risk of the target vehicle 1 colliding with other traffic (such as automobiles, motorcycles, or pedestrians), and sets an insurance premium for the target vehicle 1 based on the evaluated risk. More specifically, the system 10 to which the present disclosure is applied analyzes the roads and traffic included in the acquired image based on map information including road information and statistical information, calculates an evaluation value indicating the risk, and sets an insurance premium for the target vehicle 1 based on the calculated evaluation value. For example, the system 10 to which the present disclosure is applied can analyze the roads included in the acquired image based on the map information to calculate an evaluation value when other traffic emerges from an area not observable from the target vehicle 1, or analyze the traffic included in the acquired image to calculate an evaluation value when the route overlaps with other traffic in an area observable from the target vehicle 1. This allows the risk of the target vehicle 1 to be quantified, thereby more appropriately estimating insurance premiums.

[0062] <Configuration for Presenting a Recommended Route> When the destination to which the vehicle 1 is to travel is decided, the user sets the destination using a device such as a car navigation system or a smartphone, and route guidance from the current location to the destination is provided. Here, areas on the route to the destination that cannot be observed (invisible areas) caused by fixed features can be estimated to some extent from the route and direction of travel. Therefore, when there are multiple routes to the destination, the risk of each route can be evaluated, and the route with the lowest risk can be presented as the route recommended by the system (hereinafter also referred to as the "recommended route").

[0063] 9 is a block diagram showing an example of a configuration of the system 10 to which the present disclosure is applied, in which a recommended route is presented. In FIG. 9, parts corresponding to those in FIG. 4 are denoted by the same reference numerals, and their description will be omitted as appropriate. In FIG. 9, in addition to the edge device 11 and the cloud server 12, a UI device 13 is provided.

[0064] The UI device 13 is composed of a car navigation system installed in the vehicle 1, a smartphone or tablet terminal carried by a user in the vehicle 1, etc. When using a smartphone or the like, an application that provides car navigation functions (car navigation app) needs to be installed. The UI device 13 has a processing unit 141, an input unit 142, an output unit 143, and a communication unit 144.

[0065] The processing unit 141 is composed of a processor such as a CPU. The processing unit 141 performs various calculations and controls the operation of each unit. The input unit 142 is composed of physical buttons, etc. The input unit 142 supplies operation signals corresponding to user operations to the processing unit 141. The output unit 143 has a display unit composed of a panel such as a liquid crystal panel or an OLED (Organic Light Emitting Diode) panel, a signal processing circuit, etc. The output unit 143 displays text, images, etc. under control of the processing unit 141. The display unit may be composed of a touch panel in which a touch sensor and the display unit are integrated. The output unit 143 may also have a speaker, etc. The communication unit 144 is composed of a communication device, etc., having a communication function using a mobile communication network such as 5G. Under control of the processing unit 141, the communication unit 144 communicates with external devices, including the cloud server 12, via the communication network 21.

[0066] The communication unit 144 transmits the data supplied from the processing unit 141 as transmission data to the cloud server 12 via the communication network 21. The transmission data includes at least information regarding a destination (Goal) set by the user and a current location (Current Location). For example, the current location may be obtained using a GPS or the like, or may be obtained from another device. The communication unit 144 receives data transmitted from the cloud server 12 via the communication network 21 and supplies the data to the processing unit 141 as received data. The received data includes information regarding a recommended route. The processing unit 141 displays a recommended route on the output unit 143 based on the received data supplied from the communication unit 144.

[0067] In the cloud server 12, the processing unit 123 includes a route selection unit 133 in addition to the risk assessment unit 131 and the insurance premium setting unit 132. Transmission data transmitted from the UI device 13 is received by the communication unit 122 via the communication network 21 and supplied to the route selection unit 133. The route selection unit 133 references map information acquired from the map information storage unit 121, selects candidate routes (hereinafter also referred to as candidate routes) from the destination to the location of the destination based on information about the destination and current location included in the transmission data, and supplies information about the candidate routes to the risk assessment unit 131. At least two or more candidate routes can be selected. In FIG. 9 , the map information recorded in the map information storage unit 121 includes information for selecting candidate routes. Note that the information for selecting candidate routes is not limited to the map information recorded in the map information storage unit 121, and may be acquired from other map information.

[0068] The risk assessment unit 131 assesses the risk of collision between the vehicle 1 and other traffic for each candidate route based on the information about the candidate routes supplied from the route selection unit 133. For example, the risk assessment unit 131 assesses the risk of collision when other traffic emerges from an unobservable area (invisible area) on the candidate route. Since the unobservable area (invisible area) caused by fixed features can be estimated to some extent based on the route and traveling direction, the risk can be assessed using the calculation method shown in FIG. 7 described above. The risk assessment unit 131 supplies the assessment value obtained by assessing the risk for each candidate route to the route selection unit 133.

[0069] The route selection unit 133 selects a recommended route from among the candidate routes based on the evaluation value for each candidate route supplied from the risk evaluation unit 131. Here, of the evaluation values ​​obtained by evaluating the risk for each candidate route, the candidate route with the evaluation value evaluated as having the lowest risk can be selected as the recommended route. The route selection unit 133 supplies information about the selected recommended route (recommended route information) to the communication unit 122, and the data is transmitted to the UI device 13 via the communication network 21.

[0070] The flow of processing executed by the route selection unit 133 of the cloud server 12 in FIG. 9 will be described with reference to the flowchart in FIG.

[0071] In step S41, the route selection unit 133 acquires transmission data transmitted from the UI device 13. The transmission data includes information regarding the destination and current location set by the user using the UI device 13. In step S42, the route selection unit 133 selects a candidate route from the current location identified from the map information to the destination, based on the information regarding the destination and current location included in the transmission data. In step S43, the route selection unit 133 acquires an evaluation value for each selected candidate route from the risk assessment unit 131.

[0072] In step S44, the route selection unit 133 selects, as a recommended route, the candidate route that is evaluated as having the lowest risk from among the candidate routes based on the evaluation value for each candidate route. In step S45, the route selection unit 133 supplies the recommended route information to the communication unit 122, thereby transmitting the information to the UI device 13 via the communication network 21. When the processing of step S45 ends, the series of processes ends.

[0073] As described above, when the destination of the vehicle 1 is determined, a recommended route evaluated as having low risk is presented to the UI device 13, allowing the vehicle 1 to travel to the destination according to the presented recommended route. For example, when the vehicle 1 is driven by a human, the human drives the vehicle 1 according to a map or voice guidance showing the recommended route. Note that, when the vehicle 1 is an autonomous vehicle, the vehicle 1 may be autonomously driven based on the recommended route. When the vehicle 1 travels, the edge device 11 and the cloud server 12 can evaluate the risk of the vehicle 1 colliding with other vehicles by performing the processes shown in the flowcharts of FIGS. 5 and 6 described above, and record the evaluation value obtained by the risk evaluation. Then, the cloud server 12 can set the insurance premium for the vehicle 1 when traveling along the recommended route (a route evaluated as having low risk) based on the recorded evaluation value by performing the process shown in the flowchart of FIG. 8 described above.

[0074] <Configuration for evaluation using travel route> When a destination indicating the destination of the vehicle 1 is set using the UI device 13, information on the route traveled by the vehicle 1 from the current location to the destination (hereinafter also referred to as the travel route) can be obtained. Using this information on the travel route (hereinafter also referred to as route information), it becomes possible to estimate a certain degree of risk based on traffic volume, blind spots, etc. on the travel route based on advance statistical data. When using route information, data from the edge device 11 is not required, so a configuration without the edge device 11 can be used.

[0075] 11 is a block diagram showing an example of a configuration of a system 10 to which the present disclosure is applied, in which evaluation is performed using travel routes. In Fig. 11, parts corresponding to those in Fig. 4 and Fig. 9 are assigned the same reference numerals, and descriptions thereof will be omitted as appropriate. In Fig. 11, the edge device 11 is not provided, and the system 10 to which the present disclosure is applied is composed of a cloud server 12 and a UI device 13.

[0076] The UI device 13 is composed of a car navigation system installed in the vehicle 1, a smartphone carried by a user in the vehicle 1, etc. The UI device 13 transmits transmission data including at least route information to the cloud server 12 via the communication network 21. For example, the route information includes a destination set by the user operating the UI device 13, information about the current location acquired using GPS or the like, etc.

[0077] In the cloud server 12, the processing unit 123 includes a risk assessment unit 131 and an insurance premium setting unit 132. The risk assessment unit 131 processes route information included in the transmission data based on map information acquired from the map information storage unit 121 to assess the risk of the vehicle 1 colliding with other traffic. Here, the risk assessment unit 131 can estimate the traffic volume and blind spots on the travel route from statistical data included in the map information. Risk assessment can also be performed by estimating the width and shape of roads on the travel route from road information included in the map information. In FIG. 11 , the map information recorded in the map information storage unit 121 includes at least one of road information and traffic statistical data (statistical information) related to roads in a specific area. The specific area includes at least the travel route of the vehicle 1. In FIG. 11 , the map information may be the same as that in FIG. 4 , or other map information may be prepared.

[0078] The risk assessment unit 131 records the assessment value obtained by assessing the risk in the assessment value storage unit 124. The insurance premium setting unit 132 reads the assessment value of the target vehicle 1 from the assessment value storage unit 124, and sets the insurance premium for the target vehicle 1 based on the read assessment value. Note that in Fig. 11, similar to Fig. 9, the processing unit 123 of the cloud server 12 may be configured to include a route selection unit 133 and present a recommended route to the UI device 13.

[0079] The flow of processing executed by the risk assessment unit 131 of the cloud server 12 in FIG. 11 will be described with reference to the flowchart in FIG.

[0080] In step S51, the risk assessment unit 131 acquires transmission data transmitted from the UI device 13. The transmission data includes route information. In step S52, the risk assessment unit 131 acquires map information pre-recorded in the map information storage unit 121. In step S53, the risk assessment unit 131 processes the route information included in the transmission data based on the map information, thereby assessing the risk of the vehicle 1 colliding with other traffic on the travel route.

[0081] For example, the risk of a collision can be evaluated by estimating traffic volume and blind spots on a travel route using statistical data included in map information. Specifically, the risk can be evaluated as higher the traffic volume per unit time on the travel route. Furthermore, the risk of a collision can be evaluated by estimating the width and shape of roads on the travel route using road information included in map information. Specifically, the risk can be evaluated as lower the wider the roads on the travel route. Alternatively, speed information regarding the speed of the target vehicle 1 can be acquired, and the risk can be evaluated based on the speed at which the vehicle traveled on the travel route. Specifically, the risk can be evaluated as higher the faster the speed of the vehicle 1 traveling on the travel route. For example, the speed information can be acquired from a device installed in the target vehicle 1 and transmitted as part of the transmission data.

[0082] In this way, by acquiring route information, it is possible to more easily evaluate risk using statistical data and road information contained in pre-prepared map information, based only on the conditions of the route traveled by vehicle 1 (road conditions). However, the configuration shown in Fig. 11 does not include edge device 11, and therefore it is not possible to acquire images of the area around vehicle 1, making it impossible to recognize the traffic conditions on the spot, and therefore it is not possible to determine the risk between vehicle 1 and other vehicles in the observation area.

[0083] In step S54, the risk assessment unit 131 records the assessment value obtained by assessing the risk of collision in the assessment value storage unit 124. When the processing of step S54 ends, the series of processes ends. Then, the insurance premium setting unit 132 performs the processing shown in the flowchart of FIG. 8 described above, thereby setting the insurance premium for the target vehicle 1 based on the assessment value recorded in the assessment value storage unit 124. For example, the insurance premium can be changed depending on the speed at which the target vehicle 1 travels on the roads on the travel route indicated by the route information.

[0084] As described above, in the system 10 to which the present disclosure is applied, when a user sets a destination indicating where the target vehicle 1 is to move, route information regarding the route traveled by the target vehicle 1 from its current location to the destination is obtained, and the obtained route information is analyzed based on map information including at least one of road information regarding roads and statistical information regarding traffic, to evaluate the risk of the target vehicle 1 colliding with other traffic (cars, motorcycles, people, etc.) on the travel route, and an insurance premium for the vehicle can be set based on the evaluated risk.

[0085] By acquiring information (route information) regarding the route traveled by the target vehicle 1, it is possible to perform a sufficiently accurate risk assessment with a simplified configuration and processing, although the accuracy of the assessment is lower than in a configuration that includes the above-described edge device 11. In other words, it is possible to perform risk assessment using a UI device 13, such as a car navigation system installed in the target vehicle 1 or a smartphone carried by the user, without providing the edge device 11, thereby simplifying the device configuration on the vehicle 1 side. Furthermore, on the cloud server 12 side, the processing performed by the processing unit 123 does not require processing such as coordinate conversion, thereby simplifying the processing.

[0086] Incidentally, as autonomous mobility and autonomous driving become more widespread in the future, it is expected that there will be demand for insurance related to autonomous mobility and autonomous driving. Currently, there is insurance for short-term events for autonomous mobility and autonomous driving, but there is no insurance that can be contracted for long periods, like automobile insurance or life insurance. Insurance for short-term events is based on the premise of short-term contracts, and is therefore realized by offering expensive insurance plans that ignore economic efficiency.

[0087] Autonomous mobility and automated driving have a short history and are not yet widespread in society. Therefore, unlike human-driven vehicles, there is no long-term statistical data available. Therefore, it is difficult to realize a reasonable insurance plan that maintains economic efficiency. Conventional automobile insurance sets premiums based on factors such as the color of the driver's license, accident history, and age, and does not evaluate the driver's driving performance relative to road conditions. Therefore, the same criteria cannot be used for autonomous mobility and automated driving. In light of this current situation, this disclosure proposes a method for quantifying the aforementioned risks to realize a reasonable insurance plan that maintains economic efficiency.

[0088] One possible approach to insurance for autonomous mobility and automated driving is as follows: First, quantify the risk for vehicles such as cars, motorcycles, and forklifts driven or operated by humans, and accumulate long-term statistical data. Strict quantification is not necessary; the goal is to evaluate the driver's driving and operation in relation to road conditions, so that insurance companies can use it to set premiums, replacing the color of the driver's license, accident history, or age. Next, assuming that human driving and operation has been replaced by autonomous mobility and automated driving, the system's driving and operation in relation to road conditions is evaluated using the same scale as human driving and operation, and the risk is quantified. If the assessed risk value is the same, the same insurance plan can be estimated with the same level of risk, regardless of whether the driver or operator is a human or a system. Even with such an approach, the present disclosure can quantify the risk that the target vehicle is taking while in motion and determine Exposure. Therefore, if the Exposure value is the same, whether the target vehicle is being driven or operated by a person or a system, the magnitude of the risk is the same and the same insurance plan (premium) can be estimated.

[0089] <Modifications> <<Measures for Reducing Risk>> Since exposure can be reduced by reducing the areas that cannot be observed from the vehicle 1 (invisible areas), a configuration may be adopted in which the invisible areas are observed by some means. For example, a convex mirror (road reflector) installed on the road allows the vehicle 1 to check the situation in areas that cannot be observed. Therefore, the invisible areas can be reduced by analyzing information about other traffic reflected in the convex mirror contained in the captured image. Alternatively, the situation in the invisible areas may be recognized by installing a sensor in an invisible area, such as around a corner on the road, receiving data from the sensor, or by sharing data from other traffic via wireless communication. The situation in the invisible areas may also be recognized by using images captured from the sky by a camera mounted on a satellite, for example.

[0090] <<Other Configuration Examples>> In the above description, the cloud server 12 evaluates the risk of the vehicle 1 colliding with other traffic using the data transmitted from the edge device 11. However, the risk evaluation may be performed by the edge device 11. That is, while FIG. 4 illustrates a configuration in which the processing unit 123 of the cloud server 12 includes the risk evaluation unit 131, the edge device 11 may include the risk evaluation unit 131, and the edge device 11 may evaluate the risk of the vehicle 1 colliding with other traffic and transmit transmission data including the evaluation value to the cloud server 12 via the communication network 21. In the cloud server 12, the evaluation value included in the transmission data from the edge device 11 is recorded in the evaluation value storage unit 124, and the insurance premium setting unit 132 sets the insurance premium based on the evaluation value. When this configuration is adopted, map information needs to be pre-recorded in a storage unit (not shown) of the edge device 11. Note that the risk evaluation unit 131 is not limited to being provided in the edge device 11, and may be provided in another device mounted on the vehicle 1.

[0091] Although FIG. 4 illustrates a configuration in which the edge device 11 includes the AI ​​processing unit 112, the processing unit 123 of the cloud server 12 may also include the AI ​​processing unit 112. For example, the edge device 11 transmits to the cloud server 12 transmission data in which various information such as location information is added to an image captured by the imaging unit 111. In the processing unit 123 of the cloud server 12, the AI ​​processing unit 112 processes the image data included in the transmission data from the edge device 11, thereby supplying data including semantic data and information such as the speed and direction of other traffic to the risk assessment unit 131. The risk assessment unit 131 uses the data supplied from the AI ​​processing unit 112 to assess the risk of the vehicle 1 colliding with other traffic. Note that in FIGS. 4 and 9, the imaging unit 111 of the edge device 11 or the communication unit 122 of the cloud server 12 can also be considered an acquisition unit that acquires the image captured by the imaging unit 111. In FIG. 11, the input unit 142 of the UI device 13 or the communication unit 122 of the cloud server 12 can also be said to be an acquisition unit that acquires route information related to a travel route.

[0092] The evaluation value stored in the evaluation value storage unit 124 is an evaluation value obtained by evaluating Exposure, one of the risk indices consisting of Severity, Exposure, and Controllability, and therefore evaluation values ​​obtained by evaluating Severity and Controllability may be used as necessary when setting insurance premiums by the insurance premium setting unit 132. Since Severity depends on the kinetic energy of the vehicle and Controllability depends on the vehicle speed, the evaluation value can be calculated using statistical data, for example.

[0093] <Computer Configuration> The above-described series of processes (for example, the processes shown in the steps of the flowcharts in FIGS. 5, 6, 8, 10, and 12) can be executed by hardware or software. When the series of processes is executed by software, a program constituting the software is installed in a computer. FIG. 13 is a block diagram showing an example of the hardware configuration of a computer that executes the above-described series of processes by a program.

[0094] In the computer, a CPU (Central Processing Unit) 1001, a ROM (Read Only Memory) 1002, and a RAM (Random Access Memory) 1003 are interconnected by a bus 1004. The CPU 1001 is an example of a processor, and other processors such as a GPU (Graphics Processing Unit) may also be used. The number of processors is not limited to one, and multiple processors may be provided. An input / output interface 1005 is also connected to the bus 1004. An input unit 1006, an output unit 1007, a storage unit 1008, a communication unit 1009, and a drive 1010 are connected to the input / output interface 1005.

[0095] The input unit 1006 includes a keyboard, a mouse, a microphone, etc. The output unit 1007 includes a display, a speaker, etc. The storage unit 1008 includes a hard disk, a non-volatile memory, etc. The communication unit 1009 includes a network interface, etc. The drive 1010 drives a removable recording medium 1011 such as a semiconductor memory, a magnetic disk, an optical disk, or a magneto-optical disk.

[0096] In a computer configured as described above, the CPU 1001 loads a program recorded in the ROM 1002 or memory unit 1008 into the RAM 1003 via the input / output interface 1005 and the bus 1004 and executes it, thereby performing the above-mentioned series of processes.

[0097] The program executed by the computer (CPU 1001) can be provided by being recorded on a removable recording medium 1011 such as a package medium, for example. The program can also be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

[0098] In a computer, the program can be installed in the storage unit 1008 via the input / output interface 1005 by inserting the removable recording medium 1011 into the drive 1010. The program can also be received by the communication unit 1009 via a wired or wireless transmission medium and installed in the storage unit 1008. Alternatively, the program can be installed in advance in the ROM 1002 or the storage unit 1008.

[0099] The processing performed by a computer according to a program includes processing that is executed in parallel or individually (for example, parallel processing or object-based processing). In addition, the program may be processed by a single computer (processor) or may be distributed among multiple computers.

[0100] It should be noted that the embodiments of the present disclosure are not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure. Furthermore, the effects described in this specification are merely examples and are not intended to be limiting, and other effects may also be obtained.

[0101] The present disclosure can also be configured as follows.

[0102] (1) An information processing method including, by a computer, acquiring an image captured by a camera attached to a vehicle, analyzing at least one of roads and traffic included in the acquired image to assess the risk of the vehicle colliding with another moving object, and setting an insurance premium for the vehicle based on the assessed risk. (2) The information processing method described in (1), including: analyzing roads and traffic included in the acquired image based on map information including road information including the structure of roads in a specific area and statistical information related to traffic, to calculate an evaluation value indicating the risk, and setting the insurance premium based on the calculated evaluation value. (3) The information processing method described in (2), including: analyzing roads included in the acquired image based on the map information to calculate the evaluation value in the case where another moving object appears from an area that cannot be observed from the vehicle, and analyzing traffic included in the acquired image to calculate the evaluation value in the case where a route overlaps with that of another moving object in an area that can be observed from the vehicle. (4) The information processing method described in (3) includes: estimating an unobservable area that is an area that cannot be observed from the vehicle by converting the coordinates of the road included in the acquired image to correspond to the road structure included in the road information and comparing it with the road structure; estimating an other moving object jump-out area that is an area where other moving objects jump out from the unobservable area based on the statistical information; estimating an area where the other moving object jump-out area overlaps with a braking distance area that is an area according to the braking distance of the vehicle as a risk area that is an area where there is risk; accumulating a value indicating the probability that other moving objects exist on the path of the vehicle in the risk area over time and recording it in a memory unit as the evaluation value; and setting the insurance premium based on the recorded evaluation value.(5) The information processing method according to any one of (3) to (5), comprising: applying semantic segmentation to the acquired image to extract pixels of the road; estimating location information indicating the location of the vehicle when the image was captured; adding the location information to semantic data obtained by applying the semantic segmentation; and estimating the unobservable area by converting the coordinates of the road included in the semantic data to correspond to a structure of the road in an area specified by the location information and comparing the coordinates with the structure of the road. (6) The information processing method according to any one of (3) to (5), comprising: estimating, based on traffic included in the acquired image and output information from a device mounted on the vehicle, an area where a host vehicle route area, which is an area according to the route of the vehicle, and another moving object route area, which is an area according to the route of the other moving object, overlap as the risk area, which is the area where there is risk; accumulating over time a value indicating the probability that another moving object exists on the route of the vehicle in the risk area, and recording the result as the evaluation value in a storage unit; and setting the insurance premium based on the recorded evaluation value. (7) The information processing method according to any one of (1) to (3), further including, when a destination indicating a destination of the vehicle is set by a user, selecting, as a recommended route that is a route to be recommended, a candidate route that is evaluated to have the lowest risk from among candidate routes that are candidates for a route along which the vehicle will travel from a current location to the destination, and presenting the selected recommended route. (8) The information processing method according to (2) or (3), wherein the road information includes information on the width, shape, and traffic direction of roads in the specific area, and the statistical information includes information on the traffic volume per unit time and the speed of vehicles in the specific area. (9) An information processing device comprising: an acquisition unit that acquires an image captured by a camera attached to a vehicle; an evaluation unit that analyzes at least one of roads and traffic included in the acquired image to evaluate a risk of the vehicle colliding with another moving object; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.(10) A program causing a computer to function as: an acquisition unit that acquires images captured by a camera attached to a vehicle; an evaluation unit that analyzes at least one of roads and traffic included in the acquired images to evaluate the risk of the vehicle colliding with other moving objects; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk. (11) An information processing method including: when a destination indicating the vehicle's destination is set by a user, acquiring route information on a travel route that is a route traveled by the vehicle from its current location to the destination, analyzing the acquired route information based on map information including at least one of road information on roads and statistical information on traffic, and evaluating the risk of the vehicle colliding with other moving objects on the travel route, and setting an insurance premium for the vehicle based on the evaluated risk. (12) The information processing method according to (11), wherein the statistical information includes information on traffic volume per unit time in a specific area, and evaluating the risk from the traffic volume on the travel route based on the statistical information. (13) The information processing method according to (11) or (12), wherein the road information includes information about the width and shape of roads in a specific area, and the information processing method includes evaluating the risk from the width and shape of roads on the travel route based on the road information. (14) The information processing method according to (11), wherein the road information includes information about the width and shape of roads on the travel route based on the road information. (15) An information processing device comprising: an acquisition unit that, when a destination indicating a destination of a vehicle is set by a user, acquires route information on a travel route that is a route traveled by the vehicle from a current location to the destination, an evaluation unit that analyzes the acquired route information based on map information including at least one of road information on roads and statistical information on traffic, and evaluates a risk of the vehicle colliding with another moving object on the travel route, and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.(16) A program that causes a computer to function as: an acquisition unit that acquires, when a user sets a destination indicating where a vehicle is to move, route information regarding a travel route that is a route traveled by the vehicle from its current location to the destination; an evaluation unit that analyzes the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and evaluates the risk of the vehicle colliding with other moving objects on the travel route; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.

[0103] DESCRIPTION OF SYMBOLS 1, 1-1 to 1-N Vehicle, 10 System, 11, 11-1 to 11-N Edge device, 12 Cloud server, 13 UI device, 21 Communication network, 111 Imaging unit, 112 AI processing unit, 113 Position estimation unit, 114 Position information addition unit, 115 Communication unit, 121 Map information storage unit, 122 Communication unit, 123 Processing unit, 124 Evaluation value storage unit, 131 Risk evaluation unit, 132 Insurance premium setting unit, 133 Route selection unit, 141 Processing unit, 142 Input unit, 143 Output unit, 144 Communication unit, 1001 CPU

Claims

1. An information processing method including: a computer acquiring images captured by a camera attached to a vehicle; analyzing at least one of roads and traffic contained in the acquired images to assess the risk of the vehicle colliding with other moving objects; and setting insurance premiums for the vehicle based on the assessed risk.

2. The information processing method according to claim 1, comprising: analyzing the roads and traffic contained in the acquired image based on map information including road information including the structure of roads in a specific area and statistical information on traffic, and calculating an evaluation value indicating the risk; and setting the insurance premium based on the calculated evaluation value.

3. The information processing method according to claim 2, comprising: analyzing roads included in the acquired image based on the map information, and calculating the evaluation value when another moving object emerges from an area that cannot be observed from the vehicle; and analyzing traffic included in the acquired image, and calculating the evaluation value when the route overlaps with that of another moving object in an area that can be observed from the vehicle.

4. An information processing method as described in claim 3, comprising: estimating an unobservable area which is an area which cannot be observed from the vehicle by converting the coordinates of a road included in the acquired image to correspond to the road structure included in the road information and comparing the road structure; estimating an other moving object jumping out area which is an area where other moving objects jump out from the unobservable area based on the statistical information; estimating an area where the other moving object jumping out area overlaps with a braking distance area which is an area according to the braking distance of the vehicle, as a risk area which is an area where there is risk; accumulating over time a value indicating the probability that other moving objects exist on the path of the vehicle in the risk area, and recording the value in a memory unit as the evaluation value; and setting the insurance premium based on the recorded evaluation value.

5. The information processing method according to claim 4, comprising: applying semantic segmentation to the acquired image to extract pixels of the road; estimating location information indicating the position of the vehicle at the time the image was captured; adding the location information to semantic data obtained by applying the semantic segmentation; and estimating the unobservable area by performing coordinate transformation of the road included in the semantic data so as to correspond to the structure of the road in the area identified by the location information and comparing the same with the structure of the road.

6. An information processing method as described in claim 3, comprising: estimating an area where a vehicle route area, which is an area according to the route of the vehicle, and another moving object route area, which is an area according to the route of the other moving object, overlap as a risk area, which is an area where there is risk, based on traffic included in the acquired image and output information from equipment mounted on the vehicle; accumulating a value indicating the probability that another moving object exists on the route of the vehicle in the risk area over time, and recording the value in a memory unit as the evaluation value; and setting the insurance premium based on the recorded evaluation value.

7. The information processing method of claim 1, further comprising: when a destination indicating where the vehicle is to travel is set by a user, selecting a candidate route that is evaluated to have the lowest risk from among candidate routes that are candidates for the route along which the vehicle will travel from its current location to the destination, as a recommended route that is a recommended route; and presenting the selected recommended route.

8. The information processing method according to claim 2, wherein the road information includes information regarding the width, shape, and traffic direction of roads in the specific area, and the statistical information includes information regarding the traffic volume per unit time and vehicle speed in the specific area.

9. An information processing device comprising: an acquisition unit that acquires images captured by a camera attached to a vehicle; an evaluation unit that analyzes at least one of roads and traffic contained in the acquired images to evaluate the risk of the vehicle colliding with other moving objects; and a setting unit that sets insurance premiums for the vehicle based on the evaluated risk.

10. A program that causes a computer to function as: an acquisition unit that acquires images captured by a camera attached to a vehicle; an evaluation unit that analyzes at least one of roads and traffic contained in the acquired images to evaluate the risk of the vehicle colliding with other moving objects; and a setting unit that sets insurance premiums for the vehicle based on the evaluated risk.

11. An information processing method comprising: a computer, when a user sets a destination indicating where a vehicle is to move, acquiring route information regarding a travel route which is a route traveled by the vehicle from its current location to the destination; analyzing the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and evaluating a risk of the vehicle colliding with other moving objects on the travel route; and setting an insurance premium for the vehicle based on the evaluated risk.

12. The information processing method according to claim 11, wherein the statistical information includes information regarding traffic volume per unit time in a specific area, and the method further comprises evaluating the risk from the traffic volume on the travel route based on the statistical information.

13. The information processing method according to claim 11, wherein the road information includes information regarding the width and shape of roads in a specific area, and the information processing method further includes evaluating the risk from the width and shape of roads on the travel route based on the road information.

14. The information processing method according to claim 11, comprising: acquiring speed information relating to a speed of the vehicle; and evaluating the risk from the speed of the vehicle on the travel route based on the speed information.

15. An information processing device comprising: an acquisition unit that, when a destination indicating the destination of a vehicle is set by a user, acquires route information regarding a travel route that is a route traveled by the vehicle from a current location to the destination; an evaluation unit that analyzes the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and evaluates a risk of the vehicle colliding with other moving objects on the travel route; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.

16. A program that causes a computer to function as: an acquisition unit that acquires, when a destination indicating the destination of a vehicle is set by a user, route information regarding a travel route that is a route traveled by the vehicle from its current location to the destination; an evaluation unit that analyzes the acquired route information based on map information including at least one of road information regarding roads and statistical information regarding traffic, and evaluates the risk of the vehicle colliding with other moving objects on the travel route; and a setting unit that sets an insurance premium for the vehicle based on the evaluated risk.

Images