DETECTION DEVICE, DETECTION METHOD, AND PROGRAM

Abstract

Provides technology that enables detection of changes in a driver's driving habits To do. An acquisition unit acquires driving data of a vehicle, and a driving control unit calculates the driving status of a driver based on the driving data. A driving time model for driving time and a driving range model for the range in which the driver drives the vehicle. a decision unit for determining a driving style model relating to how the driver drives the vehicle; and a driving time At least one of the following: model change, range model change, and driving style model change and a detection unit that detects a change in the driving tendency of the driver based on the above. Provide.

Description

[Technical field]

[0001] The present invention relates to a detection device, a detection method, and a program. [Background technology]

[0002] Currently, in order to ensure the safety of drivers who drive vehicles, ADAS (Advanced Driver Assistance System) is being developed. ADAS is currently being developed with a function called Automotive Assistance Systems. It includes functions such as warning you when there is an obstacle ahead and when a collision is deemed unavoidable. These include automatic braking when the vehicle is in a critical position and warnings when the vehicle deviates from its lane. Literature 1 describes a system that predicts the driver's drowsiness level based on information obtained from the vehicle and the driver. An apparatus for doing so is disclosed. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-68935 A Summary of the Invention [Problem to be solved by the invention]

[0004] There are many reasons why accidents occur, but they are not necessarily limited to those that occur while driving. The bar is filled with fatigue, mental stress, and other reasons for aggressive driving. One of the causes of accidents is a state that differs from normal. Therefore, understanding the driving tendencies of drivers can help prevent accidents. This is considered necessary to prevent

[0005] ADAS can detect a moment of inattention when the driver is driving a car and automatically stop the vehicle. It is possible to take some action, such as raking or issuing a warning. However, ADAS is being used to prevent the driver's mental health from gradually deteriorating, resulting in poor driving. It is not possible to detect changes in the driver's driving habits, such as the driver's gradually becoming more violent. I can't come.

[0006] Therefore, the present invention is capable of detecting a change in a driver's driving tendency. The purpose of this study is to provide technology that [Means for solving the problem]

[0007] The detection device according to one aspect of the present invention includes an acquisition unit that acquires driving data of a vehicle, and Based on this, a driving time model for the driver's driving time and the range in which the driver drives the vehicle are and a driving style model that describes how the driver drives the vehicle. A decision unit for determining whether or not a driving time model change, a driving range model change, and a driving method model change A detection unit that detects a change in the driver's driving tendency based on at least one of the following: A detection device having a Effect of the Invention

[0008] According to the present invention, it is possible to detect a change in a driver's driving tendency. We can provide technology. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 illustrates an example of a detection system according to an embodiment of the present invention. [Diagram 2] FIG. 2 illustrates an example of a hardware configuration of a detection device. [Diagram 3] FIG. 2 is a diagram illustrating an example of a functional block configuration of a detection device. [Figure 4] 11 is a flowchart illustrating an example of a processing procedure for determining and updating a driving model of a driver. [Diagram 5] FIG. 2 illustrates an example of a driving time cluster. [Figure 6] FIG. 11 is a diagram illustrating an example of the frequency of driving time. [Figure 7] FIG. 2 illustrates an example of a driving range cluster. [Figure 8] FIG. 11 is a diagram showing an example of the frequency of travel locations. [Figure 9] FIG. 2 illustrates an example of a driving method cluster. [Figure 10] 10 is a flowchart showing an example of a processing procedure for detecting a change in a driving model of a driver. [Figure 11] FIG. 11 is a diagram showing an example of a predetermined control according to the content of a change in a driver's driving tendency. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] An embodiment of the present invention will be described with reference to the accompanying drawings. Those marked with the same symbol have the same or similar configuration.

[0011] <System configuration> FIG. 1 is a diagram illustrating an example of a detection system 1 according to the present embodiment. Detects changes in the driver's driving habits and takes appropriate action according to the changes The system is designed to take action according to the changes, such as driving a car to prevent accidents. This also includes alerting the driver (i.e., detecting signs of accidents). The system includes a detection device 10 and one or more vehicles 20. The detection device 10 and the vehicles 20 are wirelessly They are connected via a communication network and can communicate with each other.

[0012] The detection device 10 acquires driving data from the vehicle 20 and detects the driving status based on the acquired driving data. The driving model of the driver is determined. The driving model in this embodiment is an index or pattern that represents the tendency when driving a vehicle. At least three models are used: driving time model, driving range model, and driving method model. include.

[0013] The "driving time model" is a model that considers the time of day when a driver drives a car (e.g., driving in the morning). (e.g., if you drive a lot, or if you drive a lot in the morning and at night, etc.) is the tendency of drivers to drive their cars over a wide area (e.g., driving short distances) The "driving style model" represents the driving style of The driving habits of drivers (average speed, accelerator operation, steering, etc.) (e.g., whether the operation is careful or rough).

[0014] For example, if you often drive in the morning to go shopping in the neighborhood, and you drive carefully, For the driver, the driving time model is "driving in the morning" and the driving range model is "driving around the house". The driving method model is "driving within a 10km radius with the driver's attention focused on the accelerator and steering wheel operation" It is modeled as "carefully made."

[0015] The vehicle 20 is equipped with an on-board device that transmits and receives data to and from the detection device 10. The device detects various data acquired by various sensors provided in the vehicle 20 or the on-board device. In response to an instruction from the detection device 10, the in-vehicle device transmits the AD of the vehicle 20. A function for changing the settings of the AS function and displaying messages received from the detection device 10. It has Noh.

[0016] When the detection device 10 detects that the driving model of the driver has changed, the detection device For example, the vehicle 20 that the driver drives or the vehicle that the driver owns may be changed based on the content of the change. The device performs a predetermined notification to the user, or changes the settings of the ADAS function installed in the vehicle 20. It performs controls such as:

[0017] <Hardware configuration> FIG. 2 is a diagram illustrating an example of a hardware configuration of the detection device 10. The detection device 10 includes a CPU (Central Processing Unit), GPU (Graphical processing unit) and other processors1 1. Memory, HDD (Hard Disk Drive) and / or SSD (Solid State Drive) A storage device 12, a communication IF (Interface) 13 for performing wired or wireless communication, and a The input device 14 inputs information and the output device 15 outputs information. 4 is, for example, a keyboard, a touch panel, a mouse and / or a microphone. The chair 15 is, for example, a display, a touch panel, and / or a speaker.

[0018] The detection device 10 may be configured as one or more physical servers, or may be configured as a hypervisor. It may be configured using a virtual server that runs on a hypervisor. , may be configured using a cloud server.

[0019] <Function block configuration> FIG. 3 is a diagram illustrating an example of a functional block configuration of the detection device 10. The detection device 10 includes an acquisition unit The image forming apparatus includes a processing unit 101, a determining unit 102, a detecting unit 103, a control unit 104, and a memory unit 105. The storage unit 105 can be realized by using the storage device 12 included in the detection device 10. The acquisition unit 101, the determination unit 102, the detection unit 103, and the control unit 104 constitute a detection device 1. 0 by the processor 11 executing a program stored in the storage device 12. The program can be stored in a storage medium. The storage medium storing the program is a computer-readable non-transitory storage medium (Non- A non-transitory storage medium may be a non-transitory computer readable medium. The storage medium is not limited to this, but may be, for example, a USB memory or a CD-ROM.

[0020] The acquisition unit 101 acquires driving data of the vehicles 20 from a plurality of vehicles 20, and stores the data in the storage unit 105. The data is stored in the operation data DB (DataBase).

[0021] The determination unit 102 determines a driving model based on the driving data of the driver to be detected. The driving model of the driver to be detected is stored in the driving model history DB. In detail, the determination unit 102 determines whether the driver is a target driver based on the driving data of the target driver. A driving time model for the driving time of the vehicle 20 and a driving distance model for the range in which the driver drives the vehicle 20. Determine a range model and a driving style model for how the driver drives the vehicle 20. The determined driving time model, driving range model, and driving method model are combined into a driving model history D Store in B.

[0022] The determination unit 102 may determine the driving model in any way. Analyze driving data (from the past to the present, etc.) and select data that appears more than a specified number of times during the specified period. The driving model may be determined according to data (for example, data that appears frequently).

[0023] The detection unit 103 detects the driving status of the driver based on the presence or absence of a change in the driving model of the driver. More specifically, the detection unit 103 detects a change in the driving time model. Based on at least one of the changes in the driving range model and the changes in the driving style model, Detects changes in the driver's driving habits.

[0024] The control unit 104 detects a change in the driving tendency of the driver by the detection unit 103. When the driver has changed his / her driving habits, the control unit 100 executes a predetermined control according to the change in the driver's driving habits. The driver may be notified that their driving habits are changing. Providing a notification that driving habits are changing can be achieved, for example, by displaying a message on the The notification may be displayed on the screen, or an alarm may be sounded or a vibrator may be activated. The predetermined control may be performed by an ADAS (adaptive driving support system) mounted on the vehicle 20. The setting of the remote control device may be changed.

[0025] The storage unit 105 stores a driving data DB and a driving model history DB.

[0026] <Processing Procedure> (Decide on driving model) FIG. 4 is a flowchart showing an example of a process for determining and updating a driver's driving model. It is.

[0027] In step S10, the acquisition unit 101 acquires driving data from a plurality of vehicles 20. The data is stored in the data database.

[0028] The driving data may be any data, but for example, the position information (G PS output information), speed, acceleration, direction of travel, accelerator opening, braking strength and hand The driving data may include the driver's heart rate, the driver's steering angle, etc. The driving data may include facial images showing facial expressions. It contains time data that indicates when

[0029] In step S11, the determination unit 102 determines, based on the driving data of the driver to be detected, The driving method model, the driving range model, and the driving method model are determined. 2 is a model of the driving style of the driver to be detected, a driving range model, and a driving style model. The legal model and the date information indicating when the operating model is (e.g., January 2020) (e.g., operation model determined using operation data from January 20 to January 26) The data is then recorded in the driving model history DB.

[0030] The driver to be detected may be any selected driver, for example. , a driver selected by a company that provides accident warning services using the detection system 1 The driver may be a person who is not aware of the accident (for example, a driver who subscribes to an accident prediction service).

[0031] The determination unit 102 determines how the driving method model, the driving range model, and the driving method model are determined. However, the operation data for a predetermined period (for example, one month) may be analyzed to determine the predetermined Data that appears a predetermined number of times or more during a period (for example, data that appears frequently) is The driving model is determined by determining which cluster the vehicle belongs to among the clusters selected. You can do so.

[0032] FIG. 5 is a diagram illustrating an example of a driving time cluster. In the example of FIG. 5, Then, "1. Night travel (18:00-3:00)", "2. Afternoon travel (13:00-18:00)", "3. 1. Morning travel (6:00-12:00)", 2. Early morning (3:00-6:00) and evening (15:00-18:00) There are five clusters: "5. Morning (6:00-9:00) and nighttime (18:00-3:00) travel" and "6. Morning (6:00-9:00) and nighttime (18:00-3:00) travel". It is determined.

[0033] FIG. 6 is a diagram showing an example of the frequency of driving time. The graph shows one month's worth of operation data, aggregated on a weekly basis. The bright areas indicate the times when the driver is most likely to drive. For example, Driver A drives Therefore, the determination unit 102 determines the driving status of the driver A as follows: The time model is determined to be "1. Night travel." Similarly, Driver D travels early in the morning and It can be seen that the driver D drives more in the evening. Therefore, the determination unit 102 determines the driving mode of the driver D as follows: The model is determined to be "4. Early morning and evening travel."

[0034] FIG. 7 is a diagram showing an example of a travel range cluster. In the example of FIG. 7, The categories are "1. Short distance (less than 4km)", "2. Medium distance (4km to less than 50km)" and Three clusters have been defined: "1. Long distance (4km or more to less than 50km)" and "2. Long distance (4km or more to less than 50km)".

[0035] FIG. 8 is a diagram showing an example of the frequency of travel locations. FIG. 8 shows the frequency of travel locations for each mesh on the map. , which indicates how often the driver traveled through each mesh. The larger the value, the higher the traveling frequency. Based on the size of the range that includes all meshes with a running frequency of 4 or more, For example, the determination unit 102 may determine a travel range model including all meshes. The radius of the smallest circle that can be drawn and that can contain the vehicle is set as the driver's driving range. For example, assume that the radius of the circle is 10 km. In this case, the determination unit 102 The driver's driving range model is determined to be a model for driving "2. Medium distance". You can do so.

[0036] FIG. 9 is a diagram showing an example of a driving method cluster. In the example of FIG. 9, 1. Low speed range (less than 30km / h) 2. Medium speed range (30km / h to 70km / h) / h), "3. High speed range (70km / h or more)", "7. Medium speed range + sudden deceleration and stopping "Stop + sudden start" etc. are stipulated.

[0037] For example, the determination unit 102 may determine the driving data for a predetermined period of time by determining whether the driving speed is constant, during acceleration, or during deceleration. In addition, the determination unit 102 classifies the vehicle into a low-speed range (less than 30 km / h) during constant-speed driving. Driving in the medium speed range (30km / h or more but less than 70km / h) and in the high speed range (70km During acceleration, the determination unit 102 classifies the driving as normal acceleration. During deceleration, the system classifies the deceleration into normal deceleration and sudden deceleration (sudden deceleration). Next, the determination unit 102 classifies the vehicle into two types: Among the classified scenes, which driving method cluster the scene with the highest frequency of occurrence belongs to? By making this judgment, a driving method model of the driver is determined.

[0038] Returning to FIG. 4, the determination unit 102 performs the processes in steps S10 and S11. The procedure is repeated for each predetermined period to determine the driver's driving model for each predetermined period. In addition, the determination unit 102 does not need to detect a change in the driving model. The process is terminated for the driver who has been determined to be in a wrong position (S12-YES). The transfer model is determined in units of a specified period. The specified period can be, for example, one week or one month. In this way, the driving model history DB contains the driver's driving speed and driving speed. The driving models are recorded in sequence for each specified period. For example, if the specified period is one month, If it is on a monthly basis, the operation model for January 2020 (more precisely, from January 1 to January 31) Driving model based on driving data from ), Driving model for February 2020 (more precisely, Operation model based on operation data from February 1 to February 29), Operation model for March 2020 The driving model is recorded, such as the driving model for December 2020. This will be the case.

[0039] The processing procedure shown in Fig. 4 is carried out as a batch process on the accumulated past operation data. It may be executed in real time whenever new operating data is accumulated. It may be arranged so that

[0040] For example, the determination unit 102 may generate driving data obtained by dividing the driving data for the past few months into predetermined periods. A driving model is determined for each of the data, and a given set of driving data from the past few months is used. The operation model for each period may be determined collectively.

[0041] Alternatively, the determination unit 102 determines a driving model every time new driving data is accumulated. For example, the determination unit 102 may determine that a predetermined period of time has elapsed since the driving model was determined. When the specified period of operation data is passed and new operation data for the specified period is accumulated, the newly accumulated specified period of operation data is A driving model may be determined for the driving data for the period.

[0042] (Detection of changes in driving model) FIG. 10 is a flow diagram showing an example of a processing procedure for detecting a change in a driver's driving model. This is a chart.

[0043] In step S20, the detection unit 103 detects the driver of the detection target from the driving model history DB. Obtain data for the driving model.

[0044] In step S21, the detection unit 103 determines whether there is a short-term change in the driving method model. Specifically, the detection unit 103 determines a driving time model and a driving range at the first time. a driving method model, a driving time model at a second time, and a driving range model; By comparing the results with the driving style model, we can determine whether there are short-term changes in the driving model. If there is a short-term change in the driving model (S21-YES), the detection unit 103 A change in the driver's short-term driving tendency is detected (S22).

[0045] In step S23, the detection unit 103 determines whether there is a long-term change in the driving method model. Specifically, the detection unit 103 determines a driving time model and a driving range at the first time. A model, a driving method model, and a driving time at a third time that is a time before the second time By comparing the driving model, the driving range model, and the driving style model, we found that the driving model has long-term effects. If there is a long-term change in the driving model (S23-Y ES), the detection unit 103 detects that the driver's long-term driving tendency has changed (S twenty four).

[0046] Here, the driving model history DB records the driving model of the driver to be detected every month. For example, the driving model history DB contains the driving model le, February 2020 operation model, March 2020 operation model, ..., January 2020 Assume that the driving model for February is recorded. A short-term change is, for example, the latest driving model. This may mean comparing the model with a model that was driven a month ago. The operating model during this period is the latest operating model (i.e. the December 2020 operating model). The driving model at the second time is the driving model one month before the latest driving model (20 The long-term change may be, for example, the latest operation model. This may mean comparing a transitional model with a six-month old operational model. The operating model at any given time is, for example, the operating model six months before the latest operating model (2020 In addition, regarding short-term and long-term changes, The timing of the driving models to be compared is not limited to the above example.

[0047] The detection unit 103 detects that the driver's short-term or long-term driving tendency has changed. If so (S22, S24), the control unit 104 determines whether or not the driver's driving tendency has changed. The controller 100 performs predetermined control in response to the detected error.

[0048] FIG. 11 is a diagram showing an example of a predetermined control according to the contents of a change in the driving tendency of a driver. In FIG. 11, "no change" means that there is no change, and "-" means that there is no change. In addition, in Figure 11, "worsening performance" means that the driving method model This means that the driver's driving behavior has changed to become more dangerous, and "improvement" means that the driving behavior model has improved. In the example of Figure 9, the clusters from which the user moved to which other clusters were When a cluster moves from one to another, it is judged that the cluster is deteriorating. The information on whether the condition is judged to be "improved" or "improved" is stored in advance in the storage unit 105 as setting information. The detection unit 103 may refer to the setting information to determine whether the driving method model is "deteriorating performance." It may also be possible to determine whether the result has been improved or not.

[0049] For example, if the driver's driving time model changes from "2. Afternoon travel" to "1. Night travel," We assume that the driving method model remains unchanged. It is considered that the driving model changed due to the driver coming home late due to the accident. 104 is a predetermined control that enhances the ADAS settings (e.g., the sensitivity of the automatic brake The message to drivers about driving at night was sent to 20 vehicles. The information may be displayed on an in-vehicle device or a mobile device owned by the driver.

[0050] In addition, the driver's driving time model and driving range model remain unchanged, and the driving method model Assume that the situation has changed over the long term from "1. Low speed range" to "5. Low speed range + sudden deceleration stop + departure". In this case, the driving model changed due to distraction caused by stress, etc. Therefore, the control unit 104 strengthens the setting of the ADAS as a predetermined control. (For example, increasing the sensitivity of the automatic brakes), comments that encourage stress relief, and The driver's comments are sent to an in-vehicle device of the vehicle 20 driven by the driver or to a mobile phone owned by the driver. It may also be displayed on a mobile device.

[0051] <Additional information> FIG. 5, FIG. 7, and FIG. 9 show the driving method cluster, the driving range cluster, and the driving method cluster. The above is merely an example, and the detection device may be classified into clusters other than those shown in the figure. The device 10 uses a large amount of driving data stored in the driving data DB to generate a machine learning model (e.g., k- By using clustering methods such as the means method, driving method clusters and driving range clusters can be obtained. Clusters and driving style clusters may be defined.

[0052] The determination unit 102 may also determine the driving time model taking into account the day of the week. For example, "1. Night travel (weekdays) + no travel (holidays)", "2. Night travel (weekdays) + "Night travel (holidays)", "3. Night travel (weekdays) + afternoon travel (holidays)", "4. Night travel ( "5. Night travel (weekdays) + early morning and evening travel (holidays)", "6. Night travel (weekdays) + morning and night travel (holidays)", "7. Afternoon travel (weekdays) + travel etc." "XX. Morning and night travel (weekdays) + morning and night travel (holidays)", "XX. Morning and night travel (weekdays) + morning and night travel (holidays)", etc. The classification may be more detailed, for example.

[0053] The determination unit 102 may also determine the travel range model from the viewpoint of the starting point. For example, "1. Close distance from address XX", "2. Medium distance from address XX", 3. To determine the driving range model, such as "Long distance centered on address XX" This allows short-term or long-term changes in driving models to be avoided, such as by moving. This makes it possible to detect changes such as

[0054] In addition to the change in the driving model, the control unit 104 also changes the driving speed based on the driver's age. For example, the content of the control may be determined based on the driving style model in the long term. When the change corresponds to "worsening" and the driver is elderly, the control unit 1 04 is a system that alerts drivers to devices owned by third parties other than the driver (such as relatives). A notification message may be sent.

[0055] In addition, the driving method model is subdivided by road type, area and / or time period. For example, when the driving method model is subdivided by road type, the determining unit 102 may There are models for driving on expressways and models for driving on ordinary roads. In this way, the driving method model of the driver may be determined for each road type. In the case where the driving method model is divided into areas, the decision unit 102 selects the area where the driver drives daily. A driving style model for driving in the rear (e.g., the area determined by the driving range model); Areas outside of the area you normally drive in (e.g., areas outside of the area determined by the driving range model) A model of how a driver drives, such as a model of how the driver drives in a public area (outside the airport) The driving method model may be determined for each predetermined area. In the case of subdivision, the determination unit 102 may determine the time period such as early morning, morning, mid-morning, afternoon, evening, night, and late night. In this way, the driving model of the driver may be determined for each predetermined time period. The time period may be divided in any way. Two time periods, etc.), or three or more time periods (first time period, second time period, third time period, etc.) The driving method model may be based on the road type, area and / or time. When dividing into belts, the detection unit 103 detects roads of the same road type, the same area, and / or the same By comparing driving style models for a single time period, changes in the driving style model can be detected. Some drivers have different driving styles on regular roads and expressways. There are drivers who drive differently when going to work and when coming home. It is considered that the driving method model can be subdivided by road type, area and / or time period. This will enable more detailed insight into changes in the driver's driving style.

[0056] In the above description, the determination unit 102 determines the driving data for a predetermined period of time based on the driving data during constant speed driving, acceleration, The driving style that occurs most frequently among the classified scenes was By determining whether the driver belongs to a cluster, a driving method model of the driver is determined. However, the present invention is not limited to this. For example, the determination unit 102 may determine the occurrence frequency of the classified scenes. determining a driving method model of the driver based on the scene where the degree is equal to or greater than a predetermined threshold; In this case, there are a plurality of scenes in which the occurrence frequency is equal to or greater than a predetermined threshold. In this case, the decision unit 102 determines whether each of the scenes belongs to a driving method cluster based on the driving method cluster to which the scenes belong. For example, the driving method model may be determined based on the frequency of occurrence of the driving data. The driving method clusters corresponding to the scenes exceeding the predetermined threshold are "1. Low speed range" and "11. High speed range". If the driving style model of the driver belongs to the "region + rapid deceleration + acceleration" region, the determination unit 102 determines the driving style model of the driver as " It is possible to determine that both "1. Low speed range" and "11. High speed range + sudden deceleration + acceleration" are true. The detection unit 103 detects the driving style model represented by a plurality of driving style clusters. In this case, if any one of the driving style clusters changes, the driving style model is considered to have changed. It is possible to grasp the change in the driver's driving method in more detail. It becomes Noh.

[0057] <Summary> According to the embodiment described above, a driving model is determined based on the driving data of a driver. This allows the system to detect changes in the driving model. It will be possible to detect short-term changes in the driving model. This allows the system to detect changes in the driver's driving model as well as long-term changes. This makes it possible to detect changes in detail and take more appropriate action in response to them. It becomes Noh.

[0058] The above-described embodiment is intended to facilitate understanding of the present invention and does not limit the present invention. The flowcharts, sequences, and implementations described in the embodiments are not intended to be interpreted as being related to the present invention. The elements of the embodiment, as well as their arrangement, materials, conditions, shapes, sizes, etc. are merely examples. The present invention is not limited to the above and may be modified as appropriate. The components may be partially substituted or combined.

[0059] For example, all or part of the functions of the detection device 10 may be installed in an in-vehicle device. As an example, the function of the detection device 10 may be provided in an in-vehicle device such as a drive recorder. The vehicle may be equipped with a function to detect and notify changes in the driver's driving habits. This makes it possible to provide a drive recorder and the like. [Explanation of symbols]

[0060] 1...detection system, 10...detection device, 11...processor, 12...storage device, 13...communication IF, 14...input device, 15...output device, 20...vehicle, 101...acquisition unit, 102 ...determination unit, 103...detection unit, 104...control unit, 105...storage unit

Claims

1. An acquisition unit that acquires driving data of a vehicle; A determination unit that determines, based on the driving data, a driving time model related to a driving time of a driver, a driving range model indicating a range in which the driver drives the vehicle, and a driving method model related to a method in which the driver drives the vehicle; A detection unit that detects a change in a driving tendency of the driver based on at least one of a change in the driving time model, a change in the driving range model, and a change in the driving method model; having The determination unit determines the driving method model by dividing the driving method model into a driving method model corresponding to a range defined by the driving range model and a driving method model corresponding to a range outside the range defined by the driving range model. Detection device.

2. The determination unit records the driving time model, the driving range model, and the driving method model in association with time, The detection unit is Detecting a change in the driver's short-term driving tendency by comparing the driving time model, the driving range model, and the driving method model at a first time with the driving time model, the driving range model, and the driving method model at a second time; By comparing the driving time model, the driving range model, and the driving method model at a first time with the driving time model, the driving range model, and the driving method model at a third time that is a time in the past than the second time, a change in the long-term driving tendency of the driver is detected. The detection device of claim 1 .

3. A control unit that performs a predetermined control according to the change in the driving tendency of the driver when a change in the driving tendency of the driver is detected.

3. The detection device according to claim 1 or 2.

4. The control unit performs a predetermined control by notifying the driver of a predetermined event. The detection device of claim 3.

5. The control unit changes a setting of a driving support device mounted on the vehicle as the predetermined control.

5. A detection device according to claim 3 or 4.

6. A detection method performed by a detection device, comprising: acquiring driving data of a vehicle; determining, based on the driving data, a driving time model relating to a driver's driving time, a driving range model indicating a range in which the driver drives the vehicle, and a driving style model relating to a manner in which the driver drives the vehicle; Detecting a change in the driver's driving tendency based on at least one of a change in the driving time model, a change in the driving range model, and a change in the driving style model; Including, The determining step determines the driving method model by dividing the driving method model into a driving method model corresponding to a range defined by the driving range model and a driving method model corresponding to a range outside the range defined by the driving range model. Detection method.

7. acquiring driving data of a vehicle; determining, based on the driving data, a driving time model relating to a driver's driving time, a driving range model indicating a range in which the driver drives the vehicle, and a driving style model relating to a manner in which the driver drives the vehicle; Detecting a change in the driver's driving tendency based on at least one of a change in the driving time model, a change in the driving range model, and a change in the driving style model; Run the following on your computer: The determining step determines the driving method model by dividing the driving method model into a driving method model corresponding to a range defined by the driving range model and a driving method model corresponding to a range outside the range defined by the driving range model. program.

Images