Characteristic change detection system, characteristic change detection device, characteristic change detection method, and computer program

Abstract

This characteristic change detection system includes a characteristic change detection device that detects, using vehicle information related to travel of a vehicle, a change in characteristics of a driver who drives the vehicle. The characteristic change detection device includes: a driving characteristic value calculation unit that calculates a driving characteristic value representing the characteristics of the driver from a normal distribution indicating normal characteristics prepared in advance and an evaluation distribution generated from the vehicle information; and a detection unit that detects the presence or absence of a change in characteristics by using the driving characteristic value and a predetermined reference value. The reference value is calculated on the basis of the normal distribution and a danger distribution indicating dangerous characteristics.

Description

Characteristic change detection system, characteristic change detection device, characteristic change detection method, and computer program

[0001] The present disclosure relates to a characteristic change detection system, a characteristic change detection device, a characteristic change detection method, and a computer program. This application claims priority to Japanese Application No. 2024-123016, filed July 30, 2024, the entire contents of which are incorporated herein by reference.

[0002] As society ages, the proportion of elderly drivers is increasing, and so is the proportion of traffic accidents caused by elderly drivers. Elderly people experience changes in their physical functions as they age, such as a decline in dynamic visual acuity, difficulty processing multiple pieces of information simultaneously, and a decline in their ability to make instant decisions. Due to these changes in physical functions, elderly people may experience delays in steering or braking when driving.

[0003] However, because human functions tend to gradually decline, it is difficult for drivers to notice when they are making slow decisions when steering or braking during their daily driving.

[0004] As a technology related to driver driving, there is conventionally known a technology that warns a driver when the driver's condition is estimated to be abnormal. Such a technology is disclosed in Patent Document 1, which will be described later. The technology disclosed in Patent Document 1 defines a probability distribution of normal driving conditions and abnormal driving conditions, and estimates the driver's driving fitness based on this probability distribution. Patent Document 1 further displays a warning message on a display device according to the estimated driver's driving fitness.

[0005] JP 2009-145951 A

[0006] A characteristic change detection system according to one aspect of the present disclosure includes a characteristic change detection device that detects a change in the characteristics of a driver driving a vehicle using vehicle information related to the vehicle's travel, and the characteristic change detection device includes a driving characteristic value calculation unit that calculates a driving characteristic value representing the driver's characteristics from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information, and a detection unit that detects the presence or absence of a characteristic change using the driving characteristic value and a predetermined reference value, and the reference value is calculated based on a risk distribution indicating risky characteristics and the normal distribution.

[0007] The present disclosure can be realized not only as a characteristic change detection system, a characteristic change detection device, a characteristic change detection method, and a computer program that include such characteristic configurations, but also as other systems, devices, methods, or computer programs that include a characteristic change detection system, a characteristic change detection device, a characteristic change detection method, or a computer program.

[0008] FIG. 1 is a diagram illustrating an example of the configuration of a characteristic change detection system according to a first embodiment. FIG. 2 is a block diagram illustrating an example of the hardware configuration of an in-vehicle device illustrated in FIG. 1. FIG. 3 is a block diagram illustrating an example of the hardware configuration of a server illustrated in FIG. 1. FIG. 4 is a block diagram illustrating an example of the functional configuration of an in-vehicle system including the in-vehicle device illustrated in FIG. 1. FIG. 5 is a block diagram illustrating an example of the functional configuration of the server illustrated in FIG. 1. FIG. 6 is a diagram illustrating a characteristic change detection method. FIG. 7 is a graph illustrating an example of driving data of a driver with normal characteristics. FIG. 8 is a graph illustrating an example of driving data of a driver who may have risky characteristics. FIG. 9 is a graph illustrating the relative frequency distribution of accelerator opening difference per unit time (normal distribution and risky driving characteristic distribution). FIG. 10 is a graph illustrating the relative frequency distribution of accelerator opening difference per unit time (normal distribution and distribution of an evaluation subject). FIG. 11 is a flowchart illustrating an example of the control structure of a program executed in the server according to the first embodiment. FIG. 12 is a flowchart illustrating an example of the control structure of a program executed in the server according to the first embodiment. FIG. 13 is a diagram illustrating, in a table format, an example of each group when driver characteristics are grouped. Fig. 14 is a flowchart showing an example of a control structure of a program executed in a server according to a first modified example. Fig. 15 is a diagram showing an example of the configuration of a characteristic change detection system according to a second embodiment. Fig. 16 is a diagram showing an example of the configuration of a system according to a third embodiment. Fig. 17 is a block diagram showing an example of the functional configuration of the in-vehicle device shown in Fig. 16.

[0009] The technology disclosed in Patent Document 1 estimates a driver's state that is different from normal, such as a drowsy state, a drunk state, or a distracted state, as a driver's driving suitability. That is, the technology disclosed in Patent Document 1 estimates the driver's state while driving. Therefore, even if the technology disclosed in Patent Document 1 is used, it is difficult to detect changes in the driver's driving characteristics due to aging.

[0010] The present disclosure has been made to solve such problems, and one object of the present disclosure is to provide a characteristic change detection system, a characteristic change detection device, a characteristic change detection method, and a computer program that can detect changes in a driver's characteristics.

[0011] According to the present disclosure, it is possible to provide a characteristic change detection system, a characteristic change detection device, a characteristic change detection method, and a computer program that can detect a change in a driver's characteristics.

[0012] Preferred embodiments of the present disclosure will be described below. At least some of the embodiments described below may be combined in any combination.

[0013] (1) A characteristic change detection system according to a first aspect of the present disclosure includes a characteristic change detection device that detects a change in a characteristic of a driver who drives a vehicle using vehicle information related to the vehicle's travel, and the characteristic change detection device includes a driving characteristic value calculation unit that calculates a driving characteristic value representing the driver's characteristics from a normal distribution that indicates normal characteristics and an evaluation distribution generated from the vehicle information, and a detection unit that detects the presence or absence of a characteristic change using the driving characteristic value and a predetermined reference value, where the reference value is calculated based on a risk distribution that indicates risk characteristics and the normal distribution. This makes it possible to detect a change in the driver's characteristics even if the driver's characteristics change over time.

[0014] (2) In the above (1), the detection unit may include a dangerous driving characteristic value calculation unit that calculates a dangerous driving characteristic value by dividing the driving characteristic value by a reference value, and the detection unit may detect whether or not there is a change in the characteristics by comparing the dangerous driving characteristic value with a predetermined threshold value. This makes it possible to easily detect changes in the driver's characteristics.

[0015] (3) In the above (1) or (2), the vehicle information may include time-series information of at least one of an accelerator opening, a brake pressure, and a steering angle, thereby enabling changes in driver characteristics to be detected with high accuracy.

[0016] (4) In the above (3), the normal distribution may be a first frequency distribution generated by classifying, into a plurality of ranges, the difference values ​​between data per unit time in the time-series information indicating normal characteristics and the data per unit time immediately preceding the data, and the driving characteristic value calculation unit may calculate the difference values ​​between data per unit time in the time-series information regarding the driver and the data per unit time immediately preceding the data, and classify the plurality of difference values ​​into a plurality of ranges to create a second frequency distribution as the evaluation distribution, or may calculate, as the driving characteristic value, the sum of the differences in frequency between the second frequency distribution and the first frequency distribution in a range in which the second frequency distribution is larger than the first frequency distribution. This allows for accurate detection of changes in the driver's characteristics.

[0017] (5) In any one of (1) to (4) above, the characteristic change may include a change in the driver's characteristic from a normal characteristic to a dangerous characteristic, thereby making it possible to notify the driver that the driver's characteristic has changed to a dangerous characteristic.

[0018] (6) In the above (5), the characteristic change detection device may further include a notification unit that notifies at least one of the driver, the driver's relatives, and the manager who manages the driver that the driver's characteristic has changed to a dangerous characteristic. This makes it possible to make at least one of the driver, the driver's relatives, and the manager aware of the change in the driver's characteristic.

[0019] (7) In any one of (1) to (6) above, the characteristic change detection device may further include an information acquisition unit that acquires at least one of biometric information of the driver, vehicle driving information, and environmental information outside the vehicle while the vehicle is driving, and a characteristic estimation unit that estimates the driver's characteristics based on the information acquired by the information acquisition unit. This makes it possible to identify what characteristics have changed when the driver's characteristics have changed.

[0020] (8) In the above (7), the characteristic change detection system may further include an on-board device that is mounted on the vehicle and transmits vehicle information to the characteristic change detection device, and a seating sensor provided in the seat where the driver sits, and the on-board device may further transmit sensor data from the seating sensor as biological information to the characteristic change detection device. This makes it possible to infer, for example, a change in characteristics related to the driver's driving posture.

[0021] (9) In the above (8), the in-vehicle device may further transmit vehicle position information while the vehicle is traveling as driving information to the characteristic change detection device, thereby making it possible to estimate changes in characteristics related to, for example, dynamic visual acuity or information processing ability.

[0022] (10) In the above (7), the characteristic change detection device may communicate with an external server, and the information acquisition unit may acquire, as environmental information, weather information of the driving area when the vehicle is driving from the external server. This allows for estimation of changes in characteristics related to, for example, eyesight or information processing ability.

[0023] (11) A characteristic change detection device according to a second aspect of the present disclosure includes an acquisition unit that acquires vehicle information related to vehicle travel, and a processing unit that executes processing to detect a change in the characteristics of a driver who drives the vehicle using the vehicle information acquired by the acquisition unit, the processing unit including a driving characteristic value calculation unit that calculates a driving characteristic value representing the driver's characteristics from a normal distribution that indicates normal characteristics prepared in advance and an evaluation distribution generated from the vehicle information, and a detection unit that detects the presence or absence of a characteristic change using the driving characteristic value and a predetermined reference value, the reference value being calculated based on a risk distribution that indicates risk characteristics and the normal distribution. This makes it possible to detect a change in the driver's characteristics even if the driver's characteristics change over time.

[0024] (12) In the above (11), the characteristic change detection device may be a server device that communicates with the vehicle, thereby making it possible to easily detect changes in the driver's characteristics.

[0025] (13) In the above (11), the characteristic change detection device may be an on-board device mounted in a vehicle, thereby making it possible to easily detect changes in the driver's characteristics.

[0026] (14) A characteristic change detection method according to a third aspect of the present disclosure is a method for detecting a characteristic change of a driver who drives a vehicle, the method including the steps of: acquiring vehicle information related to the traveling of the vehicle by a computer; calculating a driving characteristic value representing the driver's characteristics from a normal distribution indicating normal characteristics that has been prepared in advance by the computer and an evaluation distribution generated from the vehicle information; and detecting whether or not the characteristic change has occurred by the computer using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risk characteristics and the normal distribution. This makes it possible to detect a change in the driver's characteristics even if the driver's characteristics change over time.

[0027] (15) A fourth aspect of the present disclosure provides a computer program for causing a computer to execute a detection step of detecting a change in a characteristic of a driver who drives a vehicle, the detection step including the steps of: acquiring vehicle information related to the traveling of the vehicle; calculating a driving characteristic value representing the driver's characteristic from a prepared normal distribution indicating normal characteristics and an evaluation distribution generated from the vehicle information; and detecting whether or not the characteristic has changed using the driving characteristic value and a predetermined reference value, the reference value being calculated based on a risk distribution indicating risk characteristics and the normal distribution. This makes it possible to detect a change in the driver's characteristic even if the driver's characteristic changes over time.

[0028] [Details of the embodiments of the present disclosure] Specific examples of a characteristic change detection system, a characteristic change detection device, a characteristic change detection method, and a computer program according to embodiments of the present disclosure will be described below with reference to the drawings. Note that in the following embodiments, identical components are assigned the same reference numerals. Their names and functions are also identical. Therefore, detailed descriptions thereof will not be repeated.

[0029] First Embodiment [Overall Configuration] Referring to FIG. 1 , a characteristic change detection system 50 according to this embodiment includes an in-vehicle device 100 mounted on a vehicle 60 and a server computer (hereinafter referred to as a server) 200 that communicates with the in-vehicle device 100. The characteristic change detection system 50 detects changes in the driving characteristics (hereinafter simply referred to as characteristics) of the driver of the vehicle 60 over time and notifies the driver or the like. By notifying the driver or the like of a change in the driver's characteristics, the characteristic change detection system 50 makes the driver aware that, for example, a change in physical function has occurred. If an administrator 70 managing the driver is present, the notification to the driver may be made via the administrator 70. Furthermore, a configuration may be adopted in which the change in the driver's characteristics is notified to the driver's relatives 80 or the like, and the relatives 80 or the like notify the driver.

[0030] In this embodiment, the server 200 notifies the driver, the driver's manager 70, the driver's relatives 80, or the like of a change in the driver's characteristics. To this end, the server 200 communicates with the in-vehicle device 100, as well as with a terminal device 72 used by the manager 70 and a mobile terminal 82 carried by the driver's relatives 80, or the like. However, the present disclosure is not limited to this configuration. For example, the server 200 may be configured to notify only the manager 70 of a change in the driver's characteristics. In this case, the manager 70 may be configured to notify the driver or the driver's relatives 80, or the like, of the change in the driver's characteristics. The server 200 may further be configured to notify only the driver's relatives 80, or the like, of a change in the driver's characteristics. In this case, the driver's relatives 80, or the like, may be configured to notify the driver of the change in the driver's characteristics.

[0031] The vehicle 60 may be, for example, a delivery vehicle such as a truck, and the driver of the vehicle 60 may be, for example, an employee of a delivery company. In this case, the administrator 70, for example, manages the vehicle 60, which is a vehicle to be monitored, and the driver of the vehicle 60. The administrator 70 may have a role of registering the vehicle and driver to be monitored in the system. In this embodiment, such a configuration will be described as an example.

[0032] The server 200 uses the data transmitted from the in-vehicle device 100 to detect changes in the characteristics of the driver of the vehicle 60 over time. Therefore, the server 200 can also be called a characteristic change detection device. The server 200 is, for example, a cloud server. However, the server 200 is not limited to this and may be another server device such as an on-premise server or an edge server.

[0033] The in-vehicle device 100 transmits to the outside vehicle data and other auxiliary data related to the traveling of the vehicle 60 on which the in-vehicle device 100 is installed. Specifically, the in-vehicle device 100 communicates with the server 200 and transmits the vehicle data and auxiliary data to the server 200.

[0034] The vehicle data includes position data indicating the traveling position of the vehicle 60 and traveling data while the vehicle 60 is traveling. The position data is, for example, position coordinate data such as GPS (Global Positioning System) data, and includes time information while traveling. The position data can be, for example, data acquired by a car navigation device (hereinafter also referred to as "on-board navigation") installed in the vehicle 60. The traveling data includes, for example, time-series information of the accelerator opening, brake pressure, and steering angle. The traveling data may be time-series information of at least one of these. The traveling data can be, for example, CAN (Controller Area Network) information.

[0035] The incidental data includes biometric data of the driver. The biometric data is sensor data detected by a biometric sensor. The biometric sensor includes a seating sensor, which is a body pressure (pressure) sensor installed in the driver's seat inside the vehicle 60. The biometric sensor may include a wearable device worn by the driver or a biometric data measuring device such as a fitness tracker, instead of or in addition to the seating sensor. In this case, the measurement data used as biometric data may include the driver's heart rate data, the driver's stress data, or the driver's physical condition data. Furthermore, if the vehicle 60 is equipped with an on-board camera that captures images of the exterior of the vehicle 60, the image data (image data) captured by the on-board camera may be used as incidental data. The image data captured by the on-board camera can provide environmental information about the vehicle 60 while it is traveling. For example, the image data can provide information such as whether the vehicle is traveling at night, whether there is traffic congestion, or whether visibility ahead of the vehicle is poor due to rain or other factors.

[0036] The in-vehicle device 100 receives vehicle data and auxiliary data (including the driver's biological data) of the vehicle 60 from the vehicle 60 and the biological sensor, respectively (analog signals are sampled at a predetermined sampling period and converted into digital data), and transmits the received data to the server 200.

[0037] The server 200 receives the vehicle data and the additional data transmitted from the in-vehicle device 100. The server 200 further communicates with an external server 90, which is an external data source. The external server 90 may be a plurality of servers or a single server. The external server 90 provides environmental information external to the vehicle 60 while the vehicle 60 is traveling. In this case, the environmental information includes, for example, weather data and time data. Of the received data, the server 200 mainly uses the vehicle data to detect changes in the characteristics of the driver of the vehicle 60.

[0038] Changes in driver characteristics include changes from normal driving characteristics (hereinafter simply referred to as normal characteristics) to dangerous driving characteristics (hereinafter simply referred to as dangerous characteristics). Normal driving characteristics, for example, refer to characteristics in a state in which a driver is able to drive normally. Dangerous driving characteristics, for example, refer to characteristics that are more likely to lead to dangerous driving situations than normal driving characteristics due to a decline in the driver's physical functions. Such changes in characteristics often occur over time. Therefore, it is difficult for drivers to notice that their own characteristics have changed into dangerous characteristics during daily driving.

[0039] For example, a delay in judgment of steering or braking, which falls under the category of risk characteristics, can occur even in normal characteristics. In other words, there is a possibility of fluctuation within the normal characteristics. Therefore, even if a delay in judgment occurs, it is difficult to immediately determine that the driver's characteristics have changed from normal characteristics to risk characteristics. This also applies to the driver himself.

[0040] To address these problems, the server 200 according to this embodiment uses vehicle data transmitted from the in-vehicle device 100 to detect subtle changes in the driver's usual accelerator work, braking, and the like. From these changes, the server 200 detects that the driver's characteristics have changed to dangerous characteristics. When the characteristic change detection system 50 detects that the driver's characteristics have changed to dangerous characteristics, it notifies the driver of this change so that the driver is aware of the change in characteristics.

[0041] Changes in characteristics are often accompanied by a decline in physical functions due to aging, and are therefore more likely to occur in elderly drivers. Therefore, here, dangerous characteristics are treated as equivalent to characteristics seen in elderly drivers. Examples of characteristics of elderly drivers include the following:

[0042] Characteristics of elderly drivers: (1) Accurate driving becomes difficult due to an overall decline in physical strength, and it also becomes difficult to continue driving for long periods of time. (2) Deteriorating eyesight makes it difficult to obtain information about the surrounding situation, making it difficult to make appropriate judgments. (3) Slower reflexes can lead to delayed responses. (4) Driving tends to become self-centered, making it difficult to objectively grasp traffic conditions.

[0043] Generally, the characteristics of elderly drivers are dominated by delayed reactions. Therefore, it is important to detect changes in the characteristics of elderly drivers before the delayed reactions become significant (i.e., before the characteristics of elderly drivers become apparent) and to provide a system that calls for caution or prevents delayed reactions.

[0044] As described above, the characteristics of elderly drivers include multiple types of characteristics. Therefore, if it is possible to understand what type of characteristics have changed, it will be more effective to alert the driver or to guide them to a system that does not cause a delayed reaction.

[0045] Taking these points into consideration, the characteristic change detection system 50 of this embodiment estimates risky characteristics, i.e., the characteristics of elderly drivers, in more detail based on the vehicle position data and accompanying data received from the in-vehicle device 100, as well as environmental information from the external server 90.

[0046] 2 , the in-vehicle device 100 includes a computer 102. The computer 102 includes a control unit 110 that controls the entire in-vehicle device 100, a memory 120 that stores various data, an in-vehicle communication unit 130 that communicates with the in-vehicle network, and a communication unit 140 that communicates with the external wireless device 62. The control unit 110, the memory 120, the in-vehicle communication unit 130, and the communication unit 140 are all connected to a communication bus 150, and data exchange between them is performed via the communication bus 150.

[0047] The control unit 110 includes a calculation unit 112, a read-only memory (ROM) 114 that stores a boot-up program for the computer 102, and a random-access memory (RAM) 116 that can be written to and read from as needed. The calculation unit 112 includes, as a calculation element (processor), a central processing unit (CPU) or a micro processing unit (MPU). The memory 120 includes, for example, a non-volatile memory such as a flash memory. The ROM 114 or the memory 120 stores computer programs (hereinafter simply referred to as programs) executed by the calculation unit 112 and various information (data).

[0048] The in-vehicle communication unit 130 provides an interface (IF) for communicating with an in-vehicle network. The in-vehicle communication unit 130 communicates with the in-vehicle network in accordance with a predetermined communication protocol. The in-vehicle network may be any of CAN, LIN (Local Interconnect Network), MOST (Media Oriented Systems Transport), FlexRay, 10BASE-T1S, 100BASE-T1, 1000BASE-T1, CXPI (Clock Extension Peripheral Interface), ASRB (Automotive Safety Restraints Bus), and CAN XL. In this embodiment, although not limited to this, a CAN is used as the in-vehicle network.

[0049] Under the control of the control unit 110, the in-vehicle device 100 (computer 102) receives data (vehicle data and additional data) for detecting changes in driver characteristics via an in-vehicle network or the like, and transmits the received data to an external server 200 (see FIG. 1 ). The communication unit 140 provides an IF for communication with an exterior wireless device 62. The exterior wireless device 62 is a wireless device for communicating with devices external to the vehicle 60. The communication unit 140 communicates with the server 200 via the exterior wireless device 62.

[0050] 3 , the server 200 includes a computer 202. The computer 202 includes a control unit 210, a storage device 220, and a communication unit 230. The control unit 210 includes a CPU 212, a GPU (Graphics Processing Unit) 214, a ROM 216, and a RAM 218. The control unit 210, the storage device 220, and the communication unit 230 are all connected to a communication bus 240, and data exchange between them is performed via the communication bus 240.

[0051] The storage device 220 includes a non-volatile storage device such as a flash memory or a hard disk drive. The storage device 220 stores various information and programs to be executed by the CPU 212. The communication unit 230 provides a connection to the network 64 that enables communication with other devices including the in-vehicle device 100.

[0052] The server 200 communicates with the in-vehicle device 100 (see FIG. 1) via the network 64 and the communication unit 230. The server 200 also communicates with external devices other than the in-vehicle device 100 via the network 64 and the communication unit 230.

[0053] The server 200 acquires data for detecting a change in the characteristics of the driver of the vehicle 60 in which the in-vehicle device 100 is installed from the in-vehicle device 100. The server 200 detects a change in the characteristics of the driver of the vehicle 60 based on the acquired data and feeds back the detection result to the driver. In other words, the server 200 notifies the driver that the driver's characteristics have changed to characteristics that may lead to a dangerous driving situation.

[0054] A program for causing server 200 to function as each functional unit of server 200 according to this embodiment is stored and distributed on a predetermined storage medium such as a DVD (Digital Versatile Disc) or a USB (Universal Serial Bus) memory, and is then transferred from this medium to storage device 220. Alternatively, the program may be transmitted from an external device to computer 202 via network 64 and stored in storage device 220.

[0055] [Functional Configuration] (In-Vehicle System 300) Referring to Fig. 4, vehicle 60 includes in-vehicle system 300. In-vehicle system 300 includes in-vehicle camera 302, biometric sensor 310, in-vehicle navigation system 320, and in-vehicle device 100. In-vehicle camera 302, biometric sensor 310, in-vehicle navigation system 320, and in-vehicle device 100 are communicably connected to one another via in-vehicle network 330. In-vehicle device 100 is further communicably connected to various ECUs (Electronic Control Units) via in-vehicle network 330.

[0056] The on-board camera 302 captures images of the external situation of the vehicle 60. The on-board camera 302 may also include a camera that captures images of the internal situation of the vehicle 60. For example, this camera may be used for driver facial recognition. This makes it possible to determine whether the driver driving the vehicle 60 is a driver registered as a detection target, thereby improving the accuracy of detecting changes in the driver's characteristics.

[0057] The biological sensor 310 includes a seating sensor 312 and a biological data measurement device 314. The seating sensor 312 is a body pressure (pressure) sensor provided in the driver's seat 66 in the vehicle. The seating sensor 312 periodically measures the distribution of the body pressure of the driver seated in the seat 66 and outputs the measured value as sensor data. The sensor data output by the seating sensor 312 includes seat pressure data. The biological data measurement device 314 is, for example, a heart rate sensor that measures the driver's heart rate. As described above, the biological data measurement device 314 may be a wearable device worn by the driver, a fitness tracker, or the like. In this case, the biological data from the biological data measurement device 314 may be transmitted to the in-vehicle device 100 without going through the in-vehicle network 330.

[0058] The in-vehicle navigation system 320 includes a location information acquisition unit 322, a route guidance unit 324, a display unit 326, and an audio output unit 328. The location information acquisition unit 322 acquires location data indicating the traveling position of the vehicle 60. The route guidance unit 324 searches for a route to the destination and displays it on the display unit 326, and receives traffic congestion information and displays the status on the display unit 326. The display unit 326 displays various information including map information. The audio output unit 328 notifies the driver of various information such as route guidance and traffic congestion information by audio. When the in-vehicle navigation system 320 receives a notification indicating a change in the driver's characteristics via the in-vehicle device 100, it issues a warning to the driver.

[0059] In addition to the communication unit 140, the in-vehicle device 100 includes, as functional units, a vehicle data receiving unit 160, an incidental data receiving unit 170, and a warning notification unit 180. Each of the vehicle data receiving unit 160, the incidental data receiving unit 170, and the warning notification unit 180 is connected to an in-vehicle network 330. Each of the vehicle data receiving unit 160, the incidental data receiving unit 170, and the warning notification unit 180 is also connected to the communication unit 140, and communicates with the server 200 (see FIG. 1 ) via the communication unit 140.

[0060] The vehicle data receiving unit 160 includes a traveling data receiving unit 162 and a position data receiving unit 164. The traveling data receiving unit 162 receives traveling data of the vehicle 60 while the vehicle is traveling from a vehicle control unit (not shown) such as an ECU. The traveling data receiving unit 162 receives, for example, CAN information as traveling data of the vehicle 60. The received CAN information includes at least one of information on the accelerator opening, brake pressure, and steering angle. The traveling data may also include speed data or acceleration data while the vehicle 60 is traveling. The position data receiving unit 164 receives, from the in-vehicle navigation system 320, position data acquired by the in-vehicle navigation system 320. The traveling data receiving unit 162 receives the traveling data of the vehicle 60 via the in-vehicle network 330 and transmits the received traveling data together with accompanying data to the server 200 via the communication unit 140.

[0061] The incidental data receiving unit 170 includes a biometric data receiving unit 172 and an image data receiving unit 174. The biometric data receiving unit 172 receives sensor data from the seating sensor 312 via the in-vehicle network 330 and transmits the received sensor data as incidental data to the server 200 via the communication unit 140. More specifically, the biometric data receiving unit 172 receives seat pressure data of the driver from the seating sensor 312 and calculates the driver's center of gravity position based on the received seat pressure data. That is, the biometric data receiving unit 172 extracts data on the driver's center of gravity position from the received seat pressure data. The biometric data receiving unit 172 transmits the seat pressure data or the extracted center of gravity position data to the server 200 as sensor data. The biometric data receiving unit 172 may be configured to receive the driver's biometric data measured by the biometric data measuring device 314 and transmit the data to the server 200 via the communication unit 140. The image data receiving unit 174 receives images captured by the vehicle-mounted camera 302 as image data, and transmits the received sensor data as additional data to the server 200 via the communication unit 140 .

[0062] The warning notification unit 180 receives a notification of a change in the driver's characteristics from the server 200 (see FIG. 1) via the communication unit 140, and transmits the received notification to, for example, the in-vehicle navigation system 320 via the in-vehicle network 330. Upon receiving the notification from the warning notification unit 180, the in-vehicle navigation system 320 issues a warning to the driver (for example, that the characteristics have changed and that this may pose a risk).

[0063] 5, control unit 210 of server 200 includes, as functional units, an information acquisition unit 250, a detection processing unit 280, and a notification unit 256. Information acquisition unit 250 controls communication unit 230 to acquire various data such as vehicle data and auxiliary data, and stores the acquired data in storage device 220. Detection processing unit 280 uses the data stored in storage device 220 to execute processing for detecting changes in the characteristics of the driver who drives the vehicle over time.

[0064] The detection processing unit 280 includes a driving characteristic value calculation unit 282, a dangerous driving characteristic value calculation unit 284, a characteristic change detection unit 286, and an estimation unit 288. The driving characteristic value calculation unit 282 creates an evaluation distribution from vehicle data (driving data) related to the driver to be evaluated, and calculates the driver's driving characteristic value using a distribution indicating normal driving characteristics (hereinafter referred to as a normal driving characteristic distribution or simply a normal distribution) and the evaluation distribution. The normal driving characteristic distribution is pre-stored in, for example, the storage device 220. The dangerous driving characteristic value calculation unit 284 calculates the dangerous driving characteristic value using the driving characteristic value calculated by the driving characteristic value calculation unit 282 and a reference value (a dangerous driving characteristic reference value described below). The reference value is calculated using a distribution indicating the normal driving characteristic distribution and a dangerous driving characteristic (e.g., a characteristic that causes a delayed reaction) (hereinafter referred to as a dangerous driving characteristic distribution or simply a dangerous distribution). The reference value is pre-stored in, for example, the storage device 220. The characteristic change detection unit 286 determines whether the characteristic of the driver to be evaluated has changed to a risky characteristic, based on the risky driving characteristic value calculated by the risky driving characteristic value calculation unit 284. The estimation unit 288 estimates more detailed characteristics of the driver based on data such as the driver's biometric data and environmental data (e.g., weather data) outside the vehicle while the vehicle 60 is traveling.

[0065] When the characteristic change detection unit 286 detects that the driver's characteristic has changed to a dangerous characteristic, the detection processing unit 280 notifies the notification unit 256 of this. The notification unit 256 receives the notification from the detection processing unit 280 and executes processing to notify the driver that the characteristic has changed via the communication unit 230. The notification unit 256 transmits a notification indicating that the driver's characteristic has changed to a dangerous characteristic to the in-vehicle device 100, for example, via the communication unit 230. The in-vehicle device 100 that has received the notification issues a warning to the driver.

[0066] The characteristic change detection unit 286 may detect that the driver's characteristic has returned from a dangerous characteristic to a normal characteristic. When the characteristic change detection unit 286 detects that the driver's characteristic has returned to the normal characteristic, the detection processing unit 280 notifies the notification unit 256 of this. The notification unit 256 receives the notification from the detection processing unit 280 and executes processing to notify the in-vehicle device 100 via the communication unit 230 that the driver's characteristic has changed. The notification unit 256 transmits a notification indicating that the driver's characteristic has changed from a dangerous characteristic to a normal characteristic to the in-vehicle device 100 via the communication unit 230, for example. Upon receiving the notification, the in-vehicle device 100 notifies the driver that the warning has been canceled.

[0067] <<Characteristic Change Detection Method>> A method for detecting a change in a driver's characteristics will be described with reference to Fig. 6. In summary, as described above with respect to the functions of the server 200, a driving characteristic value is calculated from vehicle data related to the driver to be evaluated, and based on the dangerous driving characteristic value calculated using the calculated driving characteristic value, it is determined whether the driver's characteristics have changed to a state (risky characteristic) that may pose a risk to driving.

[0068] -Definition of Distribution- An example of a distribution is shown in the top row of Figure 6. The horizontal axis x of the graph in Figure 6 represents the analysis target value, and the vertical axis represents frequency (number). For the analysis target value x, the normal distribution indicating normal driving characteristics is defined as pn(x) (see solid line). The risk distribution indicating dangerous driving characteristics (e.g., characteristics that cause delayed reaction) is defined as pi(x) (see dashed line). pn(x) and pi(x) represent the frequency when the analysis target value is x. Driving data (e.g., accelerator opening from CAN information) can be used as the analysis target value x. Note that to compare two distributions, relative frequencies (the ratio of each frequency to the total number) are used. Relative frequencies correspond to probability, and the sum of relative frequencies is 1. For two distributions, if the total numbers are the same, the frequency (number) itself may be used.

[0069] - Setting the dangerous driving characteristic reference value - As shown in the middle of Figure 6, the normal distribution pn(x) has values ​​greater than 0 in the range from x1 to x4 (pn(x1) = pn(x4) = 0). The risky distribution pi(x) is wider than the normal distribution pn(x) and has values ​​greater than 0 in the range from x0 (x0 < x1) to x5 (x5 > x4) (pi(x0) = pi(x5) = 0). The normal distribution pn(x) and the risky distribution pi(x) intersect at x2 and x3. The dangerous driving characteristic reference value (hereinafter simply referred to as the reference value) is defined as the area difference (see the shaded area) in the range where the value of the risky distribution pi(x) is greater than the value of the normal distribution pn(x). Specifically, assuming that the normal distribution pn(x) and the risky distribution pi(x) are continuous, the dangerous driving characteristic reference value Ib is calculated using the following equation 1. In reality the distribution is discrete and summations are performed instead of integrations.

[0070]

[0071] - Calculation of driving characteristic value of driver to be evaluated - From the vehicle data of the driver to be evaluated, an evaluation distribution pu(x) (see dashed line) as shown in the lower part of Figure 6 is obtained. The degree of spread of the evaluation distribution pu(x) shown in the lower part of Figure 6 is between the normal distribution pn(x) and the risk distribution pi(x). As described above, the normal distribution pn(x) has values ​​greater than 0 in the range from x1 to x4. The evaluation distribution pu(x) has values ​​greater than 0 in the range from x6 to x9 (pu(x6) = pu(x9) = 0). The normal distribution pn(x) and the evaluation distribution pu(x) intersect at x7 and x8. The driving characteristic value of the driver to be evaluated is defined as the area difference (see shaded area) in the range where the value of the evaluation distribution pu(x) is greater than the value of the normal distribution pn(x). Specifically, assuming that the normal distribution pn(x) and the evaluation distribution pu(x) are continuous, the driving characteristic value Iu is calculated by the following equation 2. In reality, the distributions are discrete, and this is performed instead of integration. This corresponds to the function of the driving characteristic value calculation unit 282 (see FIG. 5).

[0072]

[0073] - Calculation of dangerous driving characteristic value - Using the dangerous driving characteristic reference value Ib and driving characteristic value Iu calculated as described above, the dangerous driving characteristic value Iv is calculated by Iv = Iu / Ib. This corresponds to the function of the dangerous driving characteristic value calculation unit 284 (see FIG. 5 ). For example, if the dangerous driving characteristic reference value Ib is 0.25 and the driving characteristic value Iu of the driver to be evaluated is 0.12, the dangerous driving characteristic value Iv of the driver to be evaluated is Iv = 0.12 / 0.25 = 0.48.

[0074] Since the dangerous driving characteristic reference value Ib is a constant calculated by identifying the normal distribution pn(x) and the risky distribution pi(x) in advance, if the dangerous driving characteristic value Iv is 1 or greater, the evaluation distribution pu(x) is considered to have a wider spread than the risky distribution pi(x), i.e., a larger variation, and it can be determined that the risky driving characteristic is highly likely. Furthermore, if the dangerous driving characteristic value Iv is less than 1 but equal to or greater than a predetermined value close to 1 (e.g., 0.8), it can be determined that the risky driving characteristic is highly likely. If the dangerous driving characteristic value Iv is less than the predetermined value, it can be determined that the risky driving characteristic is low. That is, if Iv≧1, it is determined that the risky driving characteristic is highly likely; if 0.8≦Iv<1, it is determined that the risky driving characteristic is highly likely; and if Iv<0.8, it is determined that the risky driving characteristic is low. This corresponds to the function of the characteristic change detection unit 286 (see FIG. 5 ).

[0075] It is considered that the dangerous driving characteristic value Iv reflects the individuality of the driver. Therefore, when the dangerous driving characteristic value Iv is less than 1, instead of determining whether or not the dangerous driving characteristic is present from a single calculated value, the dangerous driving characteristic value Iv may be calculated multiple times and the dangerous driving characteristic may be evaluated in conjunction with the change trend of these values. For example, after a dangerous driving characteristic value Iv satisfying 0.8≦Iv<1 is calculated, if the dangerous driving characteristic values ​​Iv calculated multiple times tend to increase over time, it may be determined that the driver is progressing toward a dangerous driving characteristic and that there is a high possibility that the driver will develop a dangerous driving characteristic. On the other hand, if no increasing trend is observed in the dangerous driving characteristic value Iv calculated multiple times (it is almost constant or tends to decrease), it may be determined that there is a low possibility that the driver will develop a dangerous driving characteristic.

[0076] <<Example of Data to be Analyzed>> Figure 7 shows an example of driving data of a driver with normal characteristics. Figure 8 shows an example of driving data of a driver who may have dangerous characteristics. The horizontal axis of Figures 7 and 8 represents time t, and the vertical axis represents speed v(t) and accelerator opening ac(t). Drivers with normal characteristics may be, for example, driving instructors or experienced drivers in their 30s. Drivers who may have dangerous characteristics may be, for example, elderly people over a certain age (e.g., 65, 70, or 75 years old).

[0077] As shown in Figure 7, the accelerator opening of a driver with normal characteristics changes relatively smoothly. As shown in Figure 8, the accelerator opening of a driver who may have dangerous characteristics changes significantly and is not as smooth as that of a driver with normal characteristics. This is thought to be due to the fact that as people get older, their ankle joints become stiffer and they are unable to fully grasp the situation, which causes them to make decisions such as slowing down at the last moment, resulting in inconsistent accelerator work.

[0078] Thus, since the accelerator pedal depression while driving is likely to reflect the characteristics of elderly drivers, the time-series data of the accelerator pedal depression during acceleration and cruising (the difference in the accelerator pedal depression per unit time) can be used as the data to be analyzed (the above-mentioned analysis value x).

[0079] <<Examples of Distributions Indicating Normal Characteristics and Dangerous Characteristics>> Figure 9 shows the frequency distribution (specifically, relative frequency distribution) of the accelerator opening difference per unit time for driving instructors and elderly people. The "accelerator opening difference per unit time" is the time series data x(t ｎ ) as the difference value between adjacent data (i.e., x(t ｎ ) - x(t ｎ－１ )), where n is a natural number and t ｎ = t ｎ－１+Δt. The unit time may be a natural number multiple of the sampling time Δt. That is, a frequency distribution of accelerator depression differences may be created using time series data created by extracting (i.e., thinning) data at a predetermined time interval longer than the sampling time Δt from the time series data of accelerator depression. In FIG. 9 , the hatched bars show the frequency distribution of accelerator depression differences per unit time for driving instructors. This corresponds to a normal distribution. The relative frequency is calculated by classifying the data (accelerator depression differences) into multiple ranges of five intervals, and the number of data belonging to each range is the ratio to the total. The open bars show the frequency distribution of accelerator depression differences per unit time for elderly people. This corresponds to the distribution of dangerous driving characteristics. In FIG. 9 , two bar graphs representing the frequency of the same classified accelerator depression differences are displayed side by side to facilitate comparison between the two distributions.

[0080] Similarly, Figure 10 shows the frequency distribution of the accelerator opening difference per unit time for driving instructors and evaluation subjects. In Figure 10, the hatched bars show the frequency distribution of the accelerator opening difference per unit time for driving instructors (normal distribution). The open bars show the frequency distribution of the accelerator opening difference per unit time for evaluation subjects (evaluation subject distribution).

[0081] Referring to Figure 9, it can be seen that for driving instructors, the accelerator opening differences are concentrated at 0 and 5, indicating uniform accelerator work. On the other hand, for elderly people, the frequency of accelerator opening differences of 0 and 5 is lower than for driving instructors, and conversely, relatively large accelerator opening differences such as 30 and -30 are also observed. In terms of distribution, the distribution for elderly people tends to be more spread out horizontally than the distribution for driving instructors. In Figure 10, the distribution of the evaluation subjects is wider than the distribution for driving instructors (normal distribution), but not as wide as the distribution for elderly people (risky driving characteristic distribution) shown in Figure 9.

[0082] To calculate the dangerous driving characteristic reference value Ib from the two distributions shown in FIG. 9 , the sum of the difference values ​​(i.e., the frequency of the dangerous driving characteristic distribution minus the frequency of the normal distribution) is calculated where the frequency of the dangerous driving characteristic distribution is greater than that of the normal distribution. Specifically, the difference values ​​are summed for values ​​on the horizontal axis from −40 to −5 and from 10 to 40. To calculate the driving characteristic value Iu from the two distributions shown in FIG. 10 , the sum of the difference values ​​(i.e., the frequency of the distribution of the evaluation subject minus the frequency of the normal distribution) is calculated where the frequency of the distribution of the evaluation subject is greater than that of the normal distribution. Specifically, the difference values ​​are summed for values ​​on the horizontal axis from −40 to −5 and from 10 to 40. Using the dangerous driving characteristic reference value Ib obtained from the data in FIG. 9 and the driving characteristic value Iu obtained from the data in FIG. 10 , the dangerous driving characteristic value Iv can be calculated as Iv = Iu / Ib, as described above. As described above, the risky driving characteristic can be evaluated based on the risky driving characteristic value Iv itself, or the risky driving characteristic value Iv and its tendency of change.

[0083] The driving data used as the analysis target data (the above-mentioned analysis target value x) is not limited to the time series data of the accelerator pedal position. For example, time series data of the brake pressure, the steering angle, etc. may be used as the analysis target data.

[0084] [Software Configuration] The following describes the configuration of software executed by the server 200 to detect changes in driver characteristics. In the following, it is assumed that the storage device 220 of the server 200 stores vehicle data and auxiliary data transmitted from the in-vehicle device 100, as well as environmental data acquired from the external server 90.

[0085] <<Characteristic Change Detection Process>> Referring to Figure 11, the control structure of a program executed by server 200 (see Figure 1) to detect a change in driver characteristics based on the accelerator pedal position will be described. The program shown in the flowchart of Figure 11 is started, for example, according to a preset schedule. For example, at a predetermined time every day, CPU 212 of control unit 210 shown in Figure 3 reads and executes the corresponding program from storage device 220.

[0086] In step 500, the CPU 212 acquires driving data from the storage device 220 (see FIG. 3 ). Thereafter, control proceeds to step 502. As described above, the driving data received by the server 200 from the in-vehicle device 100 (computer 102) is stored in advance in the storage device 220.

[0087] In step 502, CPU 212 extracts evaluation data to be processed in subsequent steps from the driving data acquired in step 500. Specifically, CPU 212 extracts time-series data of accelerator pedal depression. CPU 212 may use, as evaluation data, time-series data with a changed unit time, created by thinning out the time-series data of accelerator pedal depression read from storage device 220 as described above. Thereafter, control proceeds to step 504.

[0088] In step 504, the CPU 212 calculates the difference value (accelerator opening difference) per unit time from the time-series data extracted in step 502, and creates a frequency distribution of the difference values ​​(hereinafter referred to as the evaluation frequency distribution). The created evaluation frequency distribution corresponds to the above-mentioned evaluation distribution pu(x). Thereafter, control proceeds to step 506.

[0089] In step 506, the CPU 212 uses the evaluation frequency distribution created in step 504 as the evaluation distribution pu(x) and the normal distribution pn(x) to calculate the driving characteristic value Iu according to Equation 2, as described above as a function of the driving characteristic value calculation unit 282. The normal distribution pn(x) is pre-stored in the storage device 220, and the CPU 212 reads and uses it. The distribution is discrete, and in Equation 2, integration is replaced by addition. Thereafter, control proceeds to step 508.

[0090] In step 508, the CPU 212 calculates the dangerous driving characteristic value Iv using the driving characteristic value Iu calculated in step 504 and the dangerous driving characteristic reference value Ib, using the formula Iv = Iu / Ib. This corresponds to the function of the dangerous driving characteristic value calculation unit 284 (see FIG. 5 ). The dangerous driving characteristic reference value Ib is pre-stored in the storage device 220, and the CPU 212 reads and uses it. The CPU 212 stores the calculated dangerous driving characteristic value Iv in the storage device 220 in association with the time (e.g., year, month, and day) when the original data (time-series data of accelerator opening) used in the calculation was generated. This is to enable the previously calculated dangerous driving characteristic value Iv to be used in step 512 (described later) the next time this program is executed. Control then proceeds to step 510.

[0091] In step 510, the CPU 212 determines whether the dangerous driving characteristic value Iv calculated in step 508 indicates a dangerous characteristic. Specifically, the CPU 212 compares the dangerous driving characteristic value Iv with a predetermined first threshold value Th1 (e.g., 1), and if Iv≧Th1, it determines that a dangerous characteristic is indicated, and control proceeds to step 514. If not (Iv<Th1), control proceeds to step 512.

[0092] In step 512, the CPU 212 determines whether the dangerous driving characteristic value Iv calculated in step 508 has transitioned to a dangerous characteristic. Steps 510 and 512 correspond to the function of the characteristic change detection unit 286 (see FIG. 5 ). Specifically, the CPU 212 compares the dangerous driving characteristic value Iv with a predetermined second threshold value Th2 (e.g., 0.8). If Iv<Th2, the CPU 212 determines that there has been no transition to a dangerous characteristic, and the program ends. If Iv≧Th2, the CPU 212 reads from the storage device 220 dangerous driving characteristic values ​​Iv calculated in previous executions of the program, and combines these values ​​with the currently calculated dangerous driving characteristic value Iv to determine whether the multiple dangerous driving characteristic values ​​Iv are on an increasing trend. If there is an increasing trend, the CPU 212 determines that there has been a transition to a dangerous characteristic, and control proceeds to step 514. Otherwise, the program ends. For example, when multiple risky driving characteristic values ​​Iv and corresponding times are plotted two-dimensionally, the CPU 212 calculates a straight line that approximates them, and determines whether the risky driving characteristic value Iv is on an increasing trend based on the slope of the calculated line. For example, if the slope is positive, it can be determined that there is an increasing trend. If the risky driving characteristic value Iv is not on an increasing trend, the CPU 212 determines that there is no transition to a risky characteristic, and the program ends.

[0093] In step 514, the CPU 212 notifies the in-vehicle device 100 that the driver's characteristics have changed. This corresponds to the function of the notification unit 256 (see FIG. 5 ). Specifically, if the determination in step 510 is YES, the CPU 212 transmits a notification to the in-vehicle device 100 via the communication unit 230 indicating that the driver's characteristics have changed to dangerous characteristics. The in-vehicle device 100, having received the notification, issues an appropriate notification to the driver (such as a warning about the dangerous characteristics). If the determination in step 512 is YES, the CPU 212 transmits a notification to the in-vehicle device 100 via the communication unit 230 indicating that the driver's characteristics are transitioning to dangerous characteristics. The in-vehicle device 100, having received the notification, issues an appropriate notification to the driver (such as a warning that the characteristics are transitioning to dangerous characteristics). Thereafter, the program ends.

[0094] 12, a control structure of a program executed by server 200 (see FIG. 1) for classifying and detailing driver characteristics will be described. This program is started, for example, in response to a determination that a dangerous characteristic exists in the characteristic change detection process.

[0095] This program includes step 3000, which acquires vehicle data, incidental data, environmental data, etc. corresponding to data determined to have a risky characteristic from storage device 220 (see FIG. 3), and step 3010, which is executed after step 3000, which classifies the risky characteristic based on the acquired incidental data and environmental data, and terminates this program. The classification result in step 3010 may be notified together with notifying administrator 70, etc., that the driver's characteristic has changed to a risky characteristic. The classification result is stored in the database of storage device 220 in association with the driver identification information corresponding to the processed data. The classification result is, for example, information representing first to fourth groups, which will be described later.

[0096] The program of FIG. 12 may be executed on the corresponding data when the determination result in step 510 or step 512 in the program of FIG. 11 is YES.

[0097] Referring to FIG. 13, in step 3010 of the program of FIG. 12, the risk characteristics are classified into, for example, four characteristics (groups).

[0098] (1) First group: From the time information and weather information, it is possible to determine whether it is night or day, and whether it is a dark environment (i.e., a dark environment due to bad weather, etc.). If a risk level is counted in such a situation (i.e., if there is a risk characteristic or a transition to a risk characteristic (the determination result in step 510 or step 512 is YES)), it is assumed that the person has difficulty seeing objects in the dark due to a decrease in night vision. Therefore, the characteristic of this group can be said to be a decrease in night vision.

[0099] (2) The second group's vehicle position information (position data) indicates whether the vehicle is traveling on a highway. If the risk level is counted while traveling on a highway, it is assumed that the vehicle is unable to match the speed of other vehicles due to a decline in dynamic visual acuity. Therefore, the characteristics of this group can be said to be a decline in dynamic visual acuity or a decline in information processing ability.

[0100] (3) Third group: Continuous driving time (e.g., 30 minutes or more) and driving posture can be determined from time information and biological information (such as seat pressure and center of gravity position). For example, if the continuous driving time is long or the center of gravity position is shifted frequently, it is assumed that fatigue will accumulate early due to an inability to assume a correct driving position due to a decrease in muscle strength or physical strength. Therefore, the characteristics of this group can be said to be a decrease in muscle strength or physical strength.

[0101] (4) Fourth Group: Vehicle position information (position data) and driving history information indicate whether the vehicle was traveling in an unfamiliar location. In unfamiliar locations, the driver must process a large amount of information while driving. If a risk level is detected while traveling in an unfamiliar location, it is assumed that the brain has difficulty recognizing peripheral vision due to a narrowed field of vision or a decline in dynamic visual acuity. Therefore, the characteristics of this group can be said to be a decline in peripheral vision or a decline in information processing ability.

[0102] In this way, by further classifying risk characteristics into smaller categories, it becomes possible to more effectively warn the driver or guide the driver to a system that does not cause a delayed response. As will be described later, when characteristics of the first group are observed, the on-board navigation system is remotely controlled to select a route with as many streetlights as possible when driving at night. When characteristics of the second group are observed, the on-board navigation system is remotely controlled to select a route that mainly uses ordinary roads and, when using expressways, is remotely controlled to select a route that is as short as possible. When characteristics of the third group are observed, the on-board navigation system is remotely controlled to select a route that includes appropriate rest stops. When characteristics of the fourth group are observed, the on-board navigation system is remotely controlled to select a route that includes easy-to-follow roads even if it takes longer.

[0103] As described above, the characteristic change detection system 50 according to this embodiment includes a server 200 that functions as a characteristic change detection device that detects a change in the characteristics of a driver driving the vehicle 60 using vehicle information related to the vehicle's travel. The server 200 includes a driving characteristic value calculation unit 282 that calculates a driving characteristic value Iu representing the driver's characteristics from a normal distribution pn(x) indicating normal characteristics that has been prepared in advance and an evaluation distribution pu(x) generated from the vehicle information. The server 200 also includes detection units (a dangerous driving characteristic value calculation unit 284 and a characteristic change detection unit 286) that detect the presence or absence of a characteristic change using the driving characteristic value Iu and a predetermined reference value (a dangerous driving characteristic reference value Ib). The reference value (the dangerous driving characteristic reference value Ib) is calculated based on a risk distribution pi(x) indicating risk characteristics and the normal distribution pn(x). This allows a change in the driver's characteristics to be detected even if the driver's characteristics change over time.

[0104] As described above, the detection unit includes the dangerous driving characteristic value calculation unit 284 that calculates the dangerous driving characteristic value by dividing the driving characteristic value Iu by a reference value (the dangerous driving characteristic reference value Ib), and the detection unit (characteristic change detection unit 286) detects whether or not there is a change by comparing the dangerous driving characteristic value Iv with a predetermined threshold value (the first threshold value). This makes it possible to easily detect changes in the driver's characteristics.

[0105] As described above, the vehicle data (driving data) may include time-series information on at least one of the accelerator pedal position, brake pressure, and steering angle, thereby enabling accurate detection of changes in the driver's driving characteristics.

[0106] As described above, the normal distribution pn(x) may be a first frequency distribution generated by classifying, into multiple ranges, the difference values ​​between data per unit time in time-series information indicating normal characteristics (e.g., accelerator opening) and the data per unit time immediately preceding the data. The driving characteristic value calculation unit 282 may calculate the difference values ​​between data per unit time in time-series information related to the driver and the data per unit time immediately preceding the data, and classify the difference values ​​into multiple ranges to create a second frequency distribution as the evaluation distribution pu(x). The driving characteristic value calculation unit 282 may calculate the sum of the differences in frequency between the second frequency distribution and the first frequency distribution in the range in which the second frequency distribution (pu(x)) is larger than the first frequency distribution (pn(x)), and use this as the driving characteristic value Iu. This allows for accurate detection of changes in the driver's driving characteristics.

[0107] As described above, the characteristic change may include a change in the driver's characteristic from a normal characteristic to a dangerous characteristic, thereby making it possible to notify the driver that the driving characteristic of the driver has changed to a dangerous characteristic.

[0108] As described above, the server 200 as a characteristic change detection device further includes a notification unit 256 that notifies at least one of the driver and the manager 70 who manages the driver that the driver's characteristic has changed to a dangerous characteristic. This makes it possible to make the driver aware of the change in the driver's characteristic.

[0109] As described above, the server 200 functioning as the characteristic change detection device may include the communication unit 230 functioning as an information acquisition unit that acquires at least one of biometric information of the driver, driving information of the vehicle 60, and environmental information outside the vehicle while the vehicle 60 is driving. The server 200 may also include a characteristic estimation unit (estimation unit 288) that estimates the characteristics of the driver based on the information acquired by the information acquisition unit. This makes it possible to identify what characteristics have changed when the driver's characteristics have changed.

[0110] As described above, the characteristic change detection system 50 may further include an in-vehicle device 100 that is mounted on the vehicle 60 and transmits vehicle information to the characteristic change detection device, and a seating sensor 312 that is provided in the seat where the driver sits. The in-vehicle device 100 may transmit sensor data from the seating sensor 312 as biological information to the server 200 that functions as the characteristic change detection device. This makes it possible to infer, for example, a change in characteristics related to the driver's driving posture.

[0111] As described above, the in-vehicle device 100 may further transmit vehicle position information while the vehicle 60 is traveling as traveling information to the server 200, which functions as a characteristic change detection device. This makes it possible to estimate, for example, changes in characteristics related to dynamic visual acuity or information processing ability.

[0112] As described above, the server 200 functioning as the characteristic change detection device may communicate with the external server 90, and the communication unit 230 functioning as the information acquisition unit may acquire, as environmental information, weather information of the driving area when the vehicle 60 is driving from the external server 90. This makes it possible to infer, for example, changes in characteristics related to eyesight or information processing ability.

[0113] (First Modification) In the first embodiment described above, an example in which the characteristic classification process is executed when the driver's characteristic is determined to be a risky characteristic has been described. However, the present disclosure is not limited to such an embodiment. For example, the vehicle data (driving data) received by the server may be grouped in advance based on the vehicle data, incidental data, environmental data, etc., and the characteristic classification process may be executed for each group. In the first modification, such a configuration example will be described.

[0114] The server according to the first modification executes the program shown in Fig. 14 instead of the program shown in Fig. 12. This program is started, for example, according to a preset schedule. For example, this program is started at a certain time every day.

[0115] Referring to FIG. 14, this program includes step 3100 of acquiring vehicle data, incidental data, and environmental data from storage device 220 (see FIG. 3), and step 3110, which is executed after step 3100, of dividing the driving data included in the vehicle data into a plurality of groups based on the vehicle data (such as position data), incidental data, and environmental data, storing the data in storage device 220, and terminating the program.

[0116] In the first modified example, in the characteristic change detection process, the program shown in FIG. 11 is executed for the driving data of each group.

[0117] 15 , a characteristic change detection system 50A according to a second embodiment differs from the first embodiment in that it detects changes in the characteristics of a driver of a general vehicle over time and notifies the driver or the like. A vehicle 60a is a general vehicle such as a private car. In this case, the manager is, for example, a relative of the driver.

[0118] The vehicle 60a is equipped with the same in-vehicle device 100 as in the first embodiment. The other configurations are the same as those in the first embodiment.

[0119] 16, a characteristic change detection device according to the third embodiment differs from the first embodiment in that it is an on-board device 400 mounted on a vehicle 60b. That is, in this embodiment, the on-board device 400 has the function of the server 200 (see FIG. 1) in the first embodiment. That is, the on-board device 400 detects changes in the characteristics of the driver of the vehicle 60b over time and notifies the driver or the driver's relatives, etc.

[0120] 17 , the functional configuration of the in-vehicle device 400 further includes a detection processing unit 450 as a functional unit in addition to the configuration of the in-vehicle device 100 shown in FIG. 4 . The vehicle data received by the vehicle data receiving unit 160 and the incidental data received by the incidental data receiving unit 170 are stored in the storage device 402. The in-vehicle device 400 communicates with an external server 90 (see FIG. 16 ), which is an external data source, via the communication unit 140. The in-vehicle device 400 acquires environmental information and the like from the external server 90 and stores the information in the storage device 402.

[0121] The detection processing unit 450 has the same functions as the detection processing unit 280 (see FIG. 5) of the server 200. The detection processing unit 450 executes a process of detecting changes in the characteristics of the driver of the vehicle 60b (see FIG. 16) over time, using data stored in the storage device 402. The detection processing unit 450 includes a driving characteristic value calculation unit 452, a dangerous driving characteristic value calculation unit 454, a characteristic change detection unit 456, and an estimation unit 458. These functional units have the same functions as the respective functional units of the server 200 (see FIG. 5) in the first embodiment.

[0122] When characteristic change detection unit 456 detects that the driver's characteristic has changed to a dangerous characteristic, detection processing unit 450 notifies warning notification unit 180 of this fact. Upon receiving the notification from detection processing unit 450, warning notification unit 180 notifies (warns) the driver of this fact, or executes processing such as notifying a terminal outside vehicle 60b (for example, a mobile terminal carried by a relative of the driver) via communication unit 140 that the driver's characteristic has changed to a dangerous characteristic.

[0123] When characteristic change detection unit 456 detects that the driver's characteristic has returned to the normal characteristic, detection processing unit 450 notifies warning notification unit 180 of this fact. Upon receiving the notification from detection processing unit 450, warning notification unit 180 notifies the driver of this fact (cancels the warning), or executes processing such as notifying a terminal outside vehicle 60b (for example, a mobile terminal carried by a relative of the driver) via communication unit 140 that the driver's characteristic has returned to the normal characteristic.

[0124] With this configuration, even if the driver's characteristics change over time, the change in the driver's characteristics can be detected. The other configurations and effects are the same as those of the first embodiment.

[0125] (Various Modifications) In the above embodiment, four groups are shown for grouping driver characteristics, but the present disclosure is not limited to such an embodiment. The number of groups for grouping may be two, three, or five or more. The four groups (characteristics) may be other than those shown in the above embodiment. Furthermore, other groups (characteristics) may be added in addition to the four groups.

[0126] In the above embodiment, an example has been described in which CAN information is used as vehicle data. However, the present disclosure is not limited to such an embodiment. The vehicle data may be information other than CAN information.

[0127] In the above embodiment, time-series data of at least one of the accelerator opening, brake pressure, and steering angle is used as the vehicle data (driving data), but the present disclosure is not limited to such an embodiment. The vehicle data (driving data) may be data other than these as long as it is data that can detect changes in the driver's characteristics over time.

[0128] Each process (each function) in the above-described embodiments may be realized by a processing circuit (circuitry) including one or more processors. The processing circuit may be configured with an integrated circuit that combines one or more memories, various analog circuits, and various digital circuits in addition to the one or more processors. The one or more memories store programs (instructions) that cause the one or more processors to execute each process. The one or more processors may execute each process according to the program read from the one or more memories, or may execute each process according to a logic circuit designed in advance to execute each process. The processor may be a CPU, GPU, DSP (Digital Signal Processor), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit), or any other processor suitable for computer control. The plurality of physically separated processors may cooperate with each other to execute the respective processes. For example, the processors mounted on the respective physically separated computers may cooperate with each other via a network such as a LAN (Local Area Network), a WAN (Wide Area Network), or the Internet to execute the respective processes.

[0129] Furthermore, a recording medium can be provided that stores a program that causes a computer to execute each of the processes of the above-described embodiments. The recording medium is, for example, an optical disk (such as a DVD) or a removable semiconductor memory (such as a USB memory). Although the computer program can be transmitted via a communication line, the recording medium is a non-transitory recording medium. By loading the program stored in the recording medium into a computer, the computer can detect changes in the characteristics of a driver operating a vehicle over time, as described above.

[0130] (Supplementary Note 1) That is, a computer-readable non-transitory recording medium records a computer program that causes a computer to execute a detection step of detecting a change in the characteristics of a driver who drives a vehicle, the detection step including: a step of acquiring vehicle information related to the vehicle's traveling; a step of calculating a driving characteristic value representing the characteristics of the driver from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information; and a step of detecting the presence or absence of the change in the characteristics using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risk characteristics and the normal distribution.

[0131] (Supplementary Note 2) The computer program product is also a computer program that causes a computer to execute a detection step of detecting a change in the characteristics of a driver who drives a vehicle, the detection step including: a step of acquiring vehicle information related to the traveling of the vehicle; a step of calculating a driving characteristic value representing the characteristics of the driver from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information; and a step of detecting the presence or absence of the change in the characteristics using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risk characteristics and the normal distribution.

[0132] Embodiments obtained by appropriately combining the techniques disclosed above are also included within the technical scope of the present disclosure.

[0133] Although the present disclosure has been described above by explaining the embodiments, the above-described embodiments are merely examples, and the present disclosure is not limited to only the above-described embodiments. The scope of the present disclosure is defined by the claims in the scope of the claims, taking into consideration the description of the detailed description of the invention, and includes all modifications within the meaning and scope equivalent to the wordings described therein.

[0134] 50, 50A Characteristic change detection system 60, 60a, 60b Vehicle 62 Exterior wireless device 64 Network 66 Seat 70 Manager 72 Terminal device 80 Relative 82 Portable terminal 90 External server 100, 400 In-vehicle device 102, 202 Computer 110, 210 Control unit 112 Calculation unit 114, 216 ROM 116, 218 RAM 120 Memory 130 In-vehicle communication unit 140, 230 Communication unit 150, 240 Communication bus 160 Vehicle data receiving unit 162 Traveling data receiving unit 164 Position data receiving unit 170 Accessory data receiving unit 172 Biometric data receiving unit 174 Image data receiving unit 180 Warning notification unit 200 Server 212 CPU 214 GPU 220, 402 Storage device 250 Information acquisition unit 256 Notification unit 280, 450 Detection processing unit 282, 452 Driving characteristic value calculation unit 284, 454 Dangerous driving characteristic value calculation unit 286, 456 Characteristic change detection unit 288, 458 Estimation unit 300 In-vehicle system 302 In-vehicle camera 310 Biometric sensor 312 Seat occupancy sensor 314 Biometric data measurement device 320 In-vehicle navigation 322 Position information acquisition unit 324 Route guidance unit 326 Display unit 328 Audio output unit 330 In-vehicle network

Claims

1. A characteristic change detection system comprising: a characteristic change detection device that detects a change in the characteristics of a driver driving a vehicle using vehicle information related to the vehicle's travel, wherein the characteristic change detection device comprises: a driving characteristic value calculation unit that calculates a driving characteristic value representing the driver's characteristics from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information; and a detection unit that detects the presence or absence of a change in the characteristic using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risky characteristics and the normal distribution.

2. The characteristic change detection system of claim 1, wherein the detection unit includes a dangerous driving characteristic value calculation unit that calculates a dangerous driving characteristic value by dividing the driving characteristic value by the reference value, and the detection unit detects the presence or absence of a change in the characteristic by comparing the dangerous driving characteristic value with a predetermined threshold value.

3. A characteristic change detection system according to claim 1 or 2, wherein the vehicle information includes time-series information on at least one of accelerator opening, brake pressure, and steering angle.

4. The characteristic change detection system of claim 3, wherein the normal distribution is a first frequency distribution generated by classifying into a plurality of ranges of magnitude difference values ​​between data per unit time in the time series information indicating the normal characteristic and the data per unit time immediately preceding the data, and the driving characteristic value calculation unit calculates the difference value between the data per unit time in the time series information regarding the driver and the data per unit time immediately preceding the data, classifies the plurality of difference values ​​into the plurality of ranges to create a second frequency distribution as the evaluation distribution, and calculates the sum of the difference in frequency between the second frequency distribution and the first frequency distribution in a range in which the second frequency distribution is larger than the first frequency distribution, and sets this as the driving characteristic value.

5. A characteristic change detection system according to any one of claims 1 to 4, wherein the characteristic change includes a change in the driver's characteristic from a normal characteristic to a dangerous characteristic.

6. A characteristic change detection system according to claim 5, further comprising a notification unit that notifies at least one of the driver, a relative of the driver, and an administrator who manages the driver that the driver's characteristic has changed to the dangerous characteristic.

7. A characteristic change detection system as described in any one of claims 1 to 6, wherein the characteristic change detection device further includes: an information acquisition unit that acquires at least one of biometric information of the driver, driving information of the vehicle, and environmental information outside the vehicle while the vehicle is driving; and a characteristic estimation unit that estimates the characteristics of the driver based on the information acquired by the information acquisition unit.

8. A characteristic change detection system as described in claim 7, further comprising: an on-board device mounted in a vehicle and transmitting the vehicle information to the characteristic change detection device; and an occupancy sensor provided in a seat where the driver is seated, wherein the on-board device further transmits sensor data from the occupancy sensor as the biological information to the characteristic change detection device.

9. The characteristic change detection system according to claim 8, wherein the in-vehicle device further transmits vehicle position information while the vehicle is traveling to the characteristic change detection device as the traveling information.

10. A characteristic change detection system as described in claim 7, wherein the characteristic change detection device communicates with an external server, and the information acquisition unit acquires weather information of the driving area when the vehicle is driving from the external server as the environmental information.

11. A characteristic change detection device comprising: an acquisition unit that acquires vehicle information related to the driving of a vehicle; and a processing unit that executes processing to detect a change in characteristics of a driver driving the vehicle using the vehicle information acquired by the acquisition unit, wherein the processing unit comprises: a driving characteristic value calculation unit that calculates a driving characteristic value representing the characteristics of the driver from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information; and a detection unit that detects the presence or absence of a change in the characteristic using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risky characteristics and the normal distribution.

12. The characteristic change detection device according to claim 11, wherein the characteristic change detection device is a server device that communicates with the vehicle.

13. The characteristic change detection device according to claim 11, wherein the characteristic change detection device is an on-board device mounted on the vehicle.

14. A method for detecting a change in a characteristic of a driver driving a vehicle, comprising: a step in which a computer acquires vehicle information related to the driving of the vehicle; a step in which the computer calculates a driving characteristic value representing the characteristics of the driver from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information; and a step in which the computer detects the presence or absence of a change in the characteristic using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risky characteristics and the normal distribution.

15. A computer program that causes a computer to execute a detection step of detecting a change in the characteristics of a driver driving a vehicle, the detection step including: a step of acquiring vehicle information related to the vehicle's traveling; a step of calculating a driving characteristic value representing the driver's characteristics from a normal distribution indicating normal characteristics that has been prepared in advance and an evaluation distribution generated from the vehicle information; and a step of detecting the presence or absence of a change in the characteristics using the driving characteristic value and a predetermined reference value, wherein the reference value is calculated based on a risk distribution indicating risk characteristics and the normal distribution.

Images