Driver assistance systems

Abstract

To provide a driving assistance system, a driving assistance method and a program, which prevent an accident from occurring by notifying a driver of a sign of the occurrence of a specific event before the event occurs.SOLUTION: A driving assistance system 10 comprises: a DMS 110 that includes an event detection unit 11 for detecting a specific event; a TCU 120 that includes a data acquisition unit 12 for acquiring factor data; and a server 200 that includes a prediction value calculation unit 13 which calculates a prediction value for determining whether or not there is a sign of the occurrence of the event based on a history in which the factor data received from the TCU 120 are accumulated, wherein the TCU 120 includes a sign determination unit 14 that compares current factor data with the prediction value so as to determine whether or not there is the sign of the occurrence of the event, and the DMS 110 includes a notification unit 15 that notifies the driver of a determination result.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present disclosure relates to a driving assistance system, a driving assistance method, and a program. [Background technology]

[0002] BACKGROUND ART There is known a technology for detecting an event (state) occurring in a driver driving a vehicle such as an automobile, and providing assistance to the driver based on the detection result.

[0003] For example, Patent Document 1 discloses a vehicle control circuit that includes an input unit that inputs detection results from at least one sensor mounted on the vehicle that detects the driver's driving condition; a recommendation processing unit that outputs recommended data regarding the use of products or services to improve the driver's driving condition indicated by the detection results to an alarm device mounted on the vehicle; a payment processing unit that executes payment processing for the use of the corresponding product or service in accordance with the driver's input regarding the recommended data; and a guidance processing unit that provides route guidance to the destination of the product or service via the alarm device based on completion of the payment processing. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-155036 Summary of the Invention [Problem to be solved by the invention]

[0005] In such a technology, if a function for notifying (warning) the driver that a specific event, such as drowsiness, has been detected is installed, the driver is notified after the specific event has occurred. However, if the driver is not in a situation where an accident can be prevented at the time the notification is given, an accident may occur. Furthermore, such a notification may hinder the driver's safe driving, which may result in an accident. Therefore, there is a problem in that it is difficult to determine whether there are signs of a specific event occurring and notify the driver before the event occurs.

[0006] In view of the above-mentioned problems, an object of the present disclosure is to provide a driving assistance system, a driving assistance method, and a program that prevent accidents from occurring by notifying the driver of signs that a specific event is about to occur before the event occurs. [Means for solving the problem]

[0007] A driving assistance system according to one embodiment includes a first vehicle-mounted device having an event detection unit that detects a specific event that occurs to a driver while driving a vehicle; a second vehicle-mounted device having a data collection unit that acquires, as factor data, at least one of physical condition data indicating the driver's physical condition that is a factor causing the event and driving situation data indicating the driver's driving situation that is a factor causing the event; and a server connected to the second vehicle-mounted device via a network. The second vehicle-mounted device transmits the factor data acquired when an event is detected to the server via the network. The server has a predicted value calculation unit that calculates a predicted value for determining whether there are signs of an event occurring based on a history of accumulated factor data received from the second vehicle-mounted device. The second vehicle-mounted device has a sign determination unit that compares the current factor data with the predicted value received from the server via the network to determine whether there are signs of an event occurring. The first vehicle-mounted device has a notification unit that notifies the driver of the determination result when it is determined that there are signs of an event occurring.

[0008] In addition, a driving assistance method according to one embodiment includes the steps of: detecting a specific event that occurs to a driver while driving a vehicle; acquiring, as factor data, at least one of physical condition data indicating the physical condition of the driver that is a factor causing the event and driving situation data indicating the driving situation of the driver that is a factor causing the event; calculating a predicted value for determining whether or not there are signs that the event will occur based on a history of accumulated factor data acquired when the event is detected; comparing the current factor data with the predicted value to determine whether or not there are signs that the event will occur; and, if it is determined that there are signs that the event will occur, notifying the driver of the determination result.

[0009] In addition, a program according to one embodiment causes a computer to perform the following processes: detecting a specific event that occurs to a driver while driving a vehicle; acquiring as factor data at least one of physical condition data indicating the driver's physical condition that is a factor in causing the event and driving situation data indicating the driver's driving situation that is a factor in causing the event; calculating a predicted value to determine whether there are signs that the event will occur based on a history of accumulated factor data acquired when an event is detected; comparing the current factor data with the predicted value to determine whether there are signs that the event will occur; and notifying the driver of the determination result if it is determined that there are signs that the event will occur. [Effects of the Invention]

[0010] The present disclosure makes it possible to provide a driving assistance system, a driving assistance method, and a program that prevent accidents from occurring by notifying the driver of signs that a specific event is about to occur before the event occurs. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of a driving assistance device according to a first embodiment. [Figure 2]3 is a flowchart illustrating a driving assistance method according to the first embodiment. [Figure 3] FIG. 10 is a block diagram showing the overall configuration of a driving assistance system according to a second embodiment. [Figure 4] 4 is a block diagram showing the configurations of a DMS, a TCU, a mobile terminal, and a server of the driving assistance system shown in FIG. 3. [Figure 5] 10 is a flowchart illustrating a process of a driving assistance system according to a second embodiment. [Figure 6] 10 is a flowchart illustrating a symptom determination process. [Figure 7] 10 is a flowchart illustrating an event detection process. [Figure 8] 10 is a flowchart illustrating a predicted value calculation process. [Figure 9] FIG. 10 is a diagram illustrating an example of a method for calculating a predicted value of a driving time. [Figure 10] FIG. 10 is a diagram illustrating an example of a method for calculating a predicted value for a time period. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, the present embodiment will be described with reference to the drawings. However, the present disclosure is not limited to the following embodiment. In addition, for clarity of explanation, the following description and drawings have been simplified as appropriate. Furthermore, in the following description, the same or equivalent elements are given the same reference numerals, and duplicate explanations will be omitted.

[0013] Embodiment 1 The configuration of a driving assistance device 1 according to the first embodiment will be described with reference to FIG. 1. FIG. 1 is a block diagram showing the configuration of the driving assistance device according to the first embodiment. The driving assistance device 1 shown in FIG. 1 is, for example, an information processing device such as an on-board device mounted on a vehicle such as an automobile. As shown in FIG. 1, the driving assistance device 1 includes an event detection unit 11, a data collection unit 12, a predicted value calculation unit 13, a symptom determination unit 14, and a notification unit 15. In the following description, a "specific event" that occurs to the driver may be simply referred to as an "event."

[0014] The event detection unit 11 detects a specific event that occurs to the driver while driving the vehicle. The data collection unit 12 acquires, as factor data, at least one of physical condition data indicating the driver's physical condition that is a factor causing the event and driving situation data indicating the driver's driving situation that is a factor causing the event. The predicted value calculation unit 13 calculates a predicted value for determining whether there is a sign that the event will occur based on a history of accumulated factor data acquired when an event is detected. The sign determination unit 14 compares the current factor data with the predicted value to determine whether there is a sign that the event will occur. If it is determined that there is a sign that an event will occur, the notification unit 15 notifies the driver of the determination result.

[0015] The event detection unit 11 detects events based on the output of a monitoring sensor that monitors events that occur to the driver. Specific events are events that can cause accidents, such as drowsiness, inattention, and poor posture. Examples of monitoring sensors that can be used include a camera that captures an image of the driver and a motion sensor that measures the driver's movements. The monitoring sensor may be connected to the driving assistance device 1 or may be built into the driving assistance device 1.

[0016] The physical condition data is data indicating the physical condition of the driver, such as heart rate, amount of exercise, body temperature, blood pressure, and electromyography, measured by a biosensor. The physical condition data is preferably heart rate data indicating the driver's heart rate measured by a biosensor capable of measuring heart rate. The lower the heart rate, the more drowsy the driver tends to feel, and therefore the more likely he or she is to fall asleep at the wheel. The biosensor may be connected to the driving assistance device 1 or may be built into the driving assistance device 1.

[0017] The driving situation data is data that indicates the driving situation of the driver, such as driving time, driving date and time, and vehicle location. Driving time is the time elapsed since driving began, and can be identified using, for example, a timer that measures the time since a driving power source such as an ignition power source was turned on, or the date and time when the driving power source was turned on. Driving date and time can be identified using, for example, a timer that measures the current date and time while driving. Vehicle location can be identified using, for example, a global navigation satellite system (GNSS) such as GPS (Global Positioning System), QZSS (Quasi-Zenith Satellite System), GLONASS (Global Navigation Satellite System), and Galileo.

[0018] For example, the longer a driver is driving, the less concentration they have, which makes them more likely to look away or lose their posture. Furthermore, during times when concentration is likely to decline, drivers are more likely to look away or lose their posture. Furthermore, if there are conspicuous buildings or scenery around the road on which the vehicle is traveling, the driver's gaze will be drawn to these, making them more likely to look away or lose their posture.

[0019] The data collection unit 12 may acquire, for example, one type of data from among the various types of physical condition data and the various types of driving situation data as factor data, or may acquire two or more types of data each as factor data.

[0020] The predicted value calculation unit 13 calculates the predicted value using a calculation method according to the type of factor data. The predicted value calculation unit 13 may calculate the predicted value based on, for example, the arithmetic mean of the accumulated factor data. The predicted value calculation unit 13 may calculate the predicted value based on, for example, the result of weighting the accumulated factor data. The predicted value calculation unit 13 may calculate the predicted value based on, for example, feature points extracted from the accumulated factor data.

[0021] The sign determination unit 14 determines whether there is a sign of an event occurring using a determination method according to the type of factor data. When comparing the current factor data with the predicted value, the sign determination unit 14 may determine that there is a sign of an event occurring, for example, when the value of the occurrence factor indicated by the current factor data is higher or lower than the predicted value of the occurrence factor. When comparing the current factor data with the predicted value, the sign determination unit 14 may determine that there is a sign of an event occurring, for example, when the value of the occurrence factor indicated by the current factor data is close to the predicted value of the occurrence factor.

[0022] When the result of the determination by the symptom determination unit 14 indicates that there is a symptom of an event occurring, the notification unit 15 notifies the driver of the determination result by voice, vibration, light, display, etc. The notification unit 15 can use a voice output device such as a speaker, a vibration generator, a light emitting device equipped with a light source such as an LED (light emitting diode), a display device such as a liquid crystal display, etc., either alone or in combination.

[0023] FIG. 2 is a flowchart illustrating a driving assistance method according to the first embodiment. As shown in FIG. 2, first, the event detection unit 11 detects a specific event that occurs to the driver while driving the vehicle (step S1). Next, the data collection unit 12 acquires, as factor data, at least one of physical condition data indicating the physical condition of the driver that is a factor causing the event and driving situation data indicating the driving situation of the driver that is a factor causing the event (step S2). Next, the predicted value calculation unit 13 calculates a predicted value for determining whether or not there is a sign of the event occurring based on a history of accumulated factor data acquired when the event is detected (step S3). Next, the sign determination unit 14 compares the current factor data with the predicted value to determine whether or not there is a sign of the event occurring (step S4). Thereafter, if it is determined that there is a sign of the event occurring, the notification unit 15 notifies the driver of the determination result (step S5).

[0024] As described above, in this embodiment, whether or not there is a sign of an event occurring is determined using a predicted value of the occurrence factor obtained by learning the tendency of the occurrence factor when an event occurs from a history that accumulates factor data acquired when the occurrence of an event is detected. This makes it possible to determine with high accuracy whether there is a sign of an event occurring. Then, in this embodiment, it is possible to notify the driver that there is a sign of an event occurring. According to this embodiment, by notifying the driver that there is a sign of a specific event occurring before the event occurs, it is possible to prevent accidents from occurring.

[0025] The driving assistance device 1 includes a processor, a memory, and a storage device (not shown). The storage device stores a computer program that implements the processing of the driving assistance method according to this embodiment. The processor then loads the computer program from the storage device into the memory and executes the computer program. As a result, the processor implements the functions of an event detection unit 11, a data collection unit 12, a predicted value calculation unit 13, a symptom determination unit 14, and a notification unit 15.

[0026] Alternatively, the event detection unit 11, the data collection unit 12, the predicted value calculation unit 13, the symptom determination unit 14, and the notification unit 15 may each be realized by dedicated hardware. Furthermore, some or all of the components of the driving assistance device 1 may be realized by a general-purpose or dedicated circuit, a processor, or a combination thereof. These may be configured by a single chip, or may be configured by multiple chips connected via a bus. Some or all of the components of the driving assistance device 1 may be realized by a combination of the above-mentioned circuits, etc., and a program. Furthermore, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field-Programmable Gate Array), etc. may be used as the processor.

[0027] Furthermore, when some or all of the components of the driving assistance device 1 are realized by multiple information processing devices, circuits, etc., the multiple information processing devices, circuits, etc. may be centrally or decentralizedly arranged within the vehicle.

[0028] Embodiment 2 The configuration of a driving assistance system 10 according to a second embodiment will be described with reference to Fig. 3. Fig. 3 is a block diagram showing the overall configuration of the driving assistance system according to the second embodiment. As shown in Fig. 3, the driving assistance system 10 according to this embodiment includes a vehicle 100 equipped with a driver monitoring system (DMS) 110, a telematics control unit (TCU) 120, and a mobile terminal 130 carried by the driver, and a server 200 connected to the TCU 120 via an external network NW. The TCU 120 wirelessly communicates with the server 200, which is connected to the network NW configured by a communication network such as the Internet, a public network, a dedicated line, or a mobile communication network, via a wireless base station BS.

[0029] In the driving assistance system 10, the DMS 110 detects a specific event that occurs to the driver while driving the vehicle 100. The TCU 120 acquires, as factor data, at least one of physical condition data indicating the physical condition of the driver that is a factor causing the event and driving situation data indicating the driving situation of the driver that is a factor causing the event. The TCU 120 compiles various data including at least one type of factor data collected from the DMS 110, the mobile terminal 130, and the inside of the TCU 120 and transmits the collected data to the server 200. The server 200, having received the various data, calculates a predicted value for determining whether or not there is a sign of an event occurring based on a history of accumulated factor data. The server 200 calculates various predicted values ​​for each type of factor data and transmits the calculated values ​​to the TCU 120. The TCU 120 compares the current factor data with the various predicted values ​​received from the server 200 to determine whether or not there is a sign of an event occurring and transmits the determination result to the DMS 110. When the TCU 120 determines that there is a sign of an event occurring, the DMS 110 notifies the driver of the determination result. The DMS 110 notifies the driver of the determination result by voice or the like.

[0030] The specific event is preferably at least one of dozing at the wheel, looking away from the driver, and poor posture. This can prevent accidents from occurring. In this embodiment, the present invention will be described in detail as a driving assistance system 10 that determines whether there are signs of the occurrence of three types of events: dozing at the wheel, looking away from the driver, and poor posture. In the following description, the "specific event" may be simply referred to as an "event."

[0031] Fig. 4 is a block diagram showing the configurations of the DMS, TCU, mobile terminal, and server of the driving assistance system shown in Fig. 3. In the example shown in Fig. 4, the DMS 110 includes an event detection unit 11 and a notification unit 15, the TCU 120 includes a data collection unit 12 and a symptom determination unit 14, and the server 200 includes a predicted value calculation unit 13. However, the arrangement of these functional units is not limited to the example shown in Fig. 4.

[0032] The DMS 110 is a first in-vehicle device configured by a computer or the like mounted on the vehicle 100. The DMS 110 monitors events that occur to the driver while driving the vehicle 100, and transmits event detection data of detected events to the TCU 120. As described above, the DMS 110 also receives a determination result of whether or not there is a sign of an event occurring from the TCU 120, and notifies the driver of the determination result. The DMS 110 includes a monitoring sensor 111, a notification unit 15, an in-vehicle communication unit 113, a control unit 114, and a storage unit 115. The monitoring sensor 111, the notification unit 15, the in-vehicle communication unit 113, and the storage unit 115 are connected to the control unit 114 by signal lines.

[0033] The monitoring sensor 111 that monitors events occurring to the driver is configured by a camera equipped with an image sensor such as a CCD (Charged Coupled Device) or a CMOS (Complementary Metal-Oxide-Semiconductor).

[0034] The monitoring sensor 111 mounted on the vehicle 100 is positioned so as to be able to capture an image of the driver's face. The monitoring sensor 111 is attached, for example, near the driver's seat of the vehicle 100. The monitoring sensor 111 generates facial image data of a captured image (moving image or still image) capturing the face of the driver who drives the vehicle 100, and transmits the facial image data to the control unit 114. The facial image data can be used for personal authentication to identify the driver, detection of the occurrence of an event, detection of the type of event, and detection of the date and time when the event occurred.

[0035] The monitoring sensor 111 may have a function to detect a specific event. In this case, the monitoring sensor 111 performs a predetermined image analysis on the captured image of the driver's face, detects an event based on the driver's face image data, and transmits the event detection data together with the face image data to the control unit 114 as necessary.

[0036] The notification unit 15 is configured with a speaker that outputs various sounds. The notification unit 15 notifies the driver of the determination result by outputting a sound based on the audio data output from the notification control unit 118. The sound output by the notification unit 15 includes a message or sound that notifies that there is a sign that a specific event is about to occur.

[0037] The notification unit 15 may be configured as a vibration generator that notifies the driver by outputting vibrations. Such a vibration generator may be built into the steering wheel, the driver's seat, or the like that is operated by the driver. The notification unit 15 configured as a vibration generator can more reliably notify the driver of the determination result by vibration even in situations where the driver has difficulty hearing the voice due to ambient noise while driving.

[0038] The notification unit 15 is not limited to being mounted on an in-vehicle device such as the DMS 110, but may be realized in the mobile terminal 130 by the mobile terminal 130 executing a specific application, for example.

[0039] The in-vehicle communication unit 113 is configured by a CAN (Controller Area Network) (registered trademark) transceiver that uses a communication protocol of CAN. The in-vehicle communication unit 113 is connected to the in-vehicle communication unit 125 via an in-vehicle network such as CAN. The DMS 110 is connected to the TCU 120 by the in-vehicle communication unit 113 so that they can communicate with each other. The in-vehicle communication unit 113 is not limited to a CAN transceiver, and may be configured by, for example, an Ethernet PHY unit that uses an Ethernet communication protocol. The DMS 110 transmits event detection data to the TUC by the in-vehicle communication unit 113, and receives a signal related to an operation instruction for the DMS 110 from the TCU 120 by the in-vehicle communication unit 113.

[0040] The control unit 114 includes, for example, at least one processor. The storage unit 115 includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The processor reads various programs stored in the ROM into the RAM and executes them, thereby providing overall control over the operation of each functional unit of the DMS 110. The processor may be, for example, a general-purpose circuit such as a CPU.

[0041] The RAM is a work memory used, for example, when the DMS 110 is operating, and temporarily stores data or information used in the operation of the DMS 110 and data or information obtained by the operation of the DMS 110. The ROM stores, in advance, various programs for controlling the DMS 110, for example.

[0042] The following describes the functional units included in the control unit 114. The control unit 114 includes an image data acquisition unit 116, an authentication unit 117, an event detection unit 11, and a notification control unit 118. The image data acquisition unit 116 acquires facial image data from the monitoring sensor 111 and transmits the facial image data to the authentication unit 117 and the event detection unit 11.

[0043] The authentication unit 117 performs personal authentication to identify the driver using facial image data and the like acquired by the image data acquisition unit 116. The authentication unit 117 may perform personal authentication such as facial authentication by, for example, comparing a feature amount extracted from the facial image data with registered data (e.g., registered feature amounts) for performing personal authentication. The registered data for performing personal authentication is associated with the driver's identifier and stored in advance in the storage unit 115. The authentication unit 117 may perform personal authentication other than facial authentication using authentication data other than facial image data as long as it can perform personal authentication to identify the driver.

[0044] The event detection unit 11 can detect an event occurring to the driver by performing a predetermined image analysis on the face image data acquired by the image data acquisition unit 116. The event detection unit 11 can determine an event based on, for example, the face direction, gaze direction, degree of eye opening, head position, eye position, etc. indicated by the face image data.

[0045] When the event detection unit 11 detects an event, it transmits event detection data indicating the detected event to the TCU 120 via the in-vehicle communication unit 113. The event detection data includes data indicating the type of event and data indicating the date and time when the event occurred. When the driver is identified by personal authentication in the authentication unit 117, the event detection unit 11 transmits the event detection data associated with the identifier of the identified driver to the TCU 120. At this time, the event detection unit 11 may transmit face image data together with the event detection data to the TCU 120 via the in-vehicle communication unit 113.

[0046] The notification control unit 118 causes the notification unit 15 to output a voice corresponding to the voice data according to the determination result received from the TCU 120 via the in-vehicle communication unit 113. This enables the notification unit 15 to notify the driver of the determination result.

[0047] The storage unit 115 may include a non-volatile storage device such as a hard disk drive (HDD), a solid state drive (SSD), or a flash memory. The storage unit 115 stores voice data corresponding to various voices output from the notification unit 15. For example, voice data synthesized in advance may be used as this voice data. The storage unit 115 also stores registration data used in the authentication unit 117 for personal authentication.

[0048] The mobile terminal 130 is, for example, a multi-function mobile phone such as a smartphone, or a wearable terminal such as a smart watch. When worn by the driver, the mobile terminal 130 can measure and record the driver's physical condition and activity state. The mobile terminal 130 acquires physical condition data indicating the driver's physical condition that will cause an event to occur, and transmits the physical condition data to the TCU 120. The mobile terminal 130 also acquires activity state data indicating the driver's activity state that will cause an event to occur, and transmits the activity state data to the TCU 120. The mobile terminal 130 includes a biosensor 131, an activity management unit 132, a short-range communication unit 133, a control unit 134, and a memory unit 135. The biosensor 131, the activity management unit 132, the short-range communication unit 133, and the memory unit 135 are connected to the control unit 134 by signal lines.

[0049] The biosensor 131 generates the measurement results of the driver's physical condition as physical condition data and transmits the physical condition data to the control unit 134. In this embodiment, the physical condition data is heart rate data indicating the driver's heart rate. In this case, a heart rate sensor that measures the driver's heart rate can be used as the biosensor 131. The heart rate data can be used to determine, for example, whether or not there are signs of drowsiness. In addition to the heart rate sensor, the mobile terminal 130 may also include other biosensors such as an acceleration sensor that measures the driver's exercise amount, a temperature sensor that measures the driver's body temperature, a blood pressure sensor that measures the driver's blood pressure, and an electromyography sensor that measures the driver's electromyography.

[0050] The activity management unit 132 generates the calculated result of the driver's activity state as activity state data and transmits the activity state data to the control unit 134. In this embodiment, the activity state data is sleep state data that indicates the sleep state of the driver based on the sleep time and sleep quality of the driver. The activity state may be calculated using, for example, the measurement results from the biosensor 131. The sleep state data can be used to determine whether there are signs of at least one of inattentiveness and poor posture, for example.

[0051] For example, the shorter the sleep time on the previous day and the worse the quality of sleep on the previous day, the more likely the driver is to experience an event while driving. Therefore, in this embodiment, the mobile device 130 acquires sleep state data indicating the driver's sleep state on the previous day (before driving) using the activity management unit 132.

[0052] The sleep state data acquired by the mobile device 130 is not limited to sleep state data indicating the sleep state of the driver on the previous day, but may be sleep state data indicating the sleep state of the driver for a predetermined period before driving. The activity state data may be data indicating the activity state of the driver that is the cause of an event. In this embodiment, sleep state data is used as an example of the activity state data, but the activity state data is not limited to this and may be, for example, exercise state data indicating an exercise state based on the number of steps and walking speed.

[0053] The short-range communication unit 133 is configured by Bluetooth (registered trademark) which performs wireless communication using a short-range wireless communication protocol. When a driver carrying the mobile terminal 130 is in the vehicle 100, the mobile terminal 130 is connected to the TCU 120 by the short-range communication unit 133 so that they can communicate with each other. The mobile terminal 130 transmits sleep state data and heart rate data to the TCU by the short-range communication unit 133, and receives signals related to operation instructions for the mobile terminal 130 from the TCU 120 by the short-range communication unit 133.

[0054] The control unit 134 includes, for example, at least one processor. The storage unit 135 includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The processor reads various programs stored in the ROM into the RAM and executes them, thereby comprehensively controlling the operation of each functional unit of the mobile terminal 130. The processor may be, for example, a general-purpose circuit such as a CPU.

[0055] The RAM is, for example, a work memory used during operation of the mobile terminal 130, and temporarily stores data or information used in the operation of the mobile terminal 130 and data or information obtained by the operation of the mobile terminal 130. The ROM stores, in advance, various programs for controlling the mobile terminal 130, for example.

[0056] The functional units included in the control unit 134 will be described. The control unit 134 includes a physical condition data acquisition unit 136 and an activity state data acquisition unit 137. The physical condition data acquisition unit 136 acquires heart rate data (physical condition data) from the biosensor 131 and stores the data in the memory unit 135. The physical condition data acquisition unit 136 transmits the heart rate data to the TCU 120 via the short-range communication unit 133 in response to a signal requesting the heart rate data received from the TCU 120. The activity state data acquisition unit 137 acquires sleep state data (activity state data) from the activity management unit 132 and stores the data in the memory unit 135. The activity state data acquisition unit 137 transmits the sleep state data to the TCU 120 via the short-range communication unit 133 in response to a signal requesting the sleep state data received from the TCU 120.

[0057] The storage unit 135 may include a non-volatile storage device such as a flash memory, etc. The storage unit 135 stores the heart rate data obtained by the biosensor 131 and the sleep state data obtained by the activity management unit 132.

[0058] The TCU 120 is a second in-vehicle device configured by a computer or the like mounted on the vehicle 100. The TCU 120 collects various data including sleep state data acquired before driving and various cause data acquired when an event is detected while driving, and transmits the collected data to the server 200.

[0059] The TCU 120 receives from the server 200 various predicted values ​​calculated for each type of factor data based on the history of the various types of accumulated factor data, and determines whether there is a sign of an event occurring using the received various predicted values. The various predicted values ​​are calculated by a predicted value calculation unit 13 of the server 200, which will be described later.

[0060] The TCU 120 includes an input / output unit 121, a timing unit 122, a positioning unit 123, an exterior communication unit 124, an interior communication unit 125, a short-range communication unit 126, a control unit 127, and a storage unit 128. The input / output unit 121, the timing unit 122, the positioning unit 123, the exterior communication unit 124, the interior communication unit 125, the short-range communication unit 126, and the storage unit 128 are connected to the control unit 127 by signal lines.

[0061] The input / output unit 121 is configured by, for example, a communication interface for serial communication. An ignition (IG) switch or the like that starts or stops the vehicle 100 is communicably connected to the input / output unit 121 via a wire harness such as a serial cable. When the IG switch is turned on or off, the input / output unit 121 receives a signal transmitted from the IG switch and transmits the signal to the control unit 127.

[0062] The timing unit 122 is composed of a timer that performs various types of timing. The timing unit 122 starts and stops timing in response to a signal from the control unit 127. The timing unit 122 generates timing data and transmits the timing data to the control unit 127. The timing data includes driving time data that indicates a driving time equivalent to the time that has elapsed since the IG switch-on signal was acquired from the control unit 127 until the IG switch-off signal was acquired. The timing data also includes date and time data of the current date and time (driving date and time).

[0063] The positioning unit 123 measures the current position of the vehicle 100. The positioning unit 123 includes, for example, a GPS (Global Positioning System) receiver, and generates vehicle position data indicating the current position (vehicle position) of the vehicle 100 based on GPS signals received from GPS satellites. The positioning unit 123 sequentially acquires the vehicle position from the time it receives an IG switch-on signal from the control unit 127 until it receives an IG switch-off signal, and transmits the generated vehicle position data to the control unit 127. The vehicle position data is, for example, two-dimensional coordinates or three-dimensional coordinates of the vehicle 100 at the current date and time.

[0064] The positioning unit 123 is not limited to including a GPS receiver, and may generate vehicle position data by locating the current position of the vehicle 100 based on the distance from the radio base station BS, or may acquire position data from a mobile terminal 130 equipped with a positioning function.

[0065] The exterior communication unit 124 is configured by a communication interface that performs wireless communication using a mobile communication protocol such as LTE (Long Term Evolution), 4G (4th Generation), 5G (5th Generation), Wifi (registered trademark), etc. The exterior communication unit 124 allows the TCU 120 to transmit and receive various data or information to and from the server 200 via the network NW and the wireless base station BS.

[0066] The in-vehicle communication unit 125 is configured by a CAN transceiver that uses the CAN communication protocol. The TCU 120 is connected to the DMS 110 by the in-vehicle communication unit 125 so that they can communicate with each other.

[0067] The short-distance communication unit 126 is configured by Bluetooth, which performs wireless communication using a short-distance wireless communication protocol. The TCU 120 is connected to the mobile terminal 130 by the short-distance communication unit 126 so that they can communicate with each other.

[0068] The control unit 127 includes, for example, at least one processor. The storage unit 128 includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The processor reads various programs stored in the ROM into the RAM and executes them, thereby comprehensively controlling the operation of each functional unit of the TCU 120. The processor may be, for example, a general-purpose circuit such as a CPU.

[0069] The RAM is a work memory used, for example, when the TCU 120 is operating, and temporarily stores data or information used in the operation of the TCU 120 and data or information obtained by the operation of the TCU 120. The ROM stores, in advance, various programs for controlling the TCU 120, for example.

[0070] The control unit 127 includes a data collection unit 12 and a symptom determination unit 14. The data collection unit 12 acquires sleep state data transmitted from the mobile terminal 130 and received by the short-range communication unit 126. For example, when the TCU 120 starts communication with the mobile terminal 130, the data collection unit 12 acquires the sleep state data from the mobile terminal 130 via the short-range communication unit 126. In response to a signal requesting the sleep data received from the server 200, the data collection unit 12 transmits the sleep state data to the server 200 via the exterior communication unit 124.

[0071] Furthermore, the data collection unit 12 acquires event detection data transmitted from the DMS 110 during driving and received by the in-vehicle communication unit 125. The data collection unit 12 also acquires heart rate data transmitted from the mobile terminal 130 and received by the short-range communication unit 126 as factor data. The data collection unit 12 also acquires vehicle position data from the positioning unit 123 as factor data. The data collection unit 12 also acquires timing data including driving time data and date and time data from the timing unit 122 as factor data.

[0072] In response to a signal requesting various types of factor data received from the server 200, the data collection unit 12 transmits various types of factor data acquired when an event is detected to the server 200 via the exterior communication unit 124. The data collection unit 12 can identify the various types of factor data acquired when an event is detected based on the event detection data. The data collection unit 12 also transmits the various types of factor data acquired sequentially to the symptom determination unit 14.

[0073] The symptom determination unit 14 accesses the server 200 via the exterior communication unit 124 and acquires various predicted values ​​accumulated in the predicted value storage unit 212. The symptom determination unit 14 compares the current factor data acquired by the data collection unit 12 with the predicted values ​​stored in the predicted value storage unit 212 to determine whether there is a sign of an event occurring.

[0074] When the factor data is heart rate data, the sign determination unit 14 can determine whether or not there is a sign of dozing off by comparing the current heart rate data acquired from the mobile terminal 130 with the predicted heart rate value acquired from the server 200. The sign determination unit 14 determines that there is a sign of dozing off when the heart rate indicated by the current heart rate data is lower than the predicted heart rate value.

[0075] When the factor data is driving time data, the symptom determination unit 14 can determine whether there is a symptom of at least one of inattentive driving and poor posture occurring by comparing the current driving time acquired from the timing unit 122 with the predicted value of driving time acquired from the server 200. When the driving time indicated by the current driving time data is close to the predicted value of driving time, the symptom determination unit 14 determines that there is a symptom of at least one of inattentive driving and poor posture occurring.

[0076] When the factor data is date and time data, the symptom determination unit 14 can determine whether there is a symptom of at least one of inattentive driving and poor posture occurring by comparing the current date and time data acquired from the clock unit 122 with the predicted value of the time period acquired from the server 200. When the driving date and time indicated by the current date and time data is close to the predicted value of the time period, the symptom determination unit 14 determines that there is a symptom of at least one of inattentive driving and poor posture occurring.

[0077] When the factor data is vehicle position data, the symptom determination unit 14 can determine whether there is a symptom of at least one of inattentive driving and poor posture occurring by comparing the current vehicle position data acquired from the positioning unit 123 with the predicted value of the vehicle position received from the server 200. When the vehicle position indicated by the current vehicle position data is close to the predicted value of the vehicle position, the symptom determination unit 14 determines that there is a symptom of at least one of inattentive driving and poor posture occurring.

[0078] When the symptom determination unit 14 determines that there is a symptom of an event occurring, it transmits the determination result to the DMS 110 via the in-vehicle communication unit 125. The determination result includes data indicating that there is a symptom of an event occurring. The determination result may also include data indicating the type of event.

[0079] The storage unit 128 may include a non-volatile storage device such as an HDD, an SSD, or a flash memory. The storage unit 128 stores map data including road information of the roads on which the vehicle 100 is traveling, facility information around the roads, and map information. This map data may be map data that is stored in advance in the storage unit 128, or may be the latest map data obtained from the server 200 by the exterior communication unit 124.

[0080] The server 200 is a database server provided outside the vehicle 100. The server 200 receives various data from the TCU 120 and transmits to the TCU 120 various predicted values, which are prediction results obtained by predicting the occurrence of an event for each type of factor data based on a history of accumulated various factor data. The server 200 performs the main operation of the system and is configured by a server, a computer, etc. However, it does not need to be physically a single server and may be configured to perform distributed processing. The server 200 includes a server communication unit 201, a control unit 202, and a memory unit 203. The server communication unit 201 and the memory unit 203 are connected to the control unit 202 by signal lines.

[0081] The server communication unit 201 is configured by a communication interface for connecting the server 200 to the network NW. The server 200 transmits and receives various data or information to and from the TCU 120 via the network NW and the wireless base station BS, using the server communication unit 201.

[0082] The control unit 202 includes, for example, at least one processor. The storage unit 203 includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The processor reads various programs stored in the ROM into the RAM and executes them, thereby comprehensively controlling the operation of each functional unit of the server 200. The processor may be, for example, a general-purpose circuit such as a CPU.

[0083] The RAM is a work memory used, for example, when the server 200 is operating, and temporarily stores data or information used in the operation of the server 200 and data or information obtained by the operation of the server 200. The ROM stores, in advance, various programs for controlling the server 200, for example.

[0084] The control unit 202 includes a predicted value calculation unit 13. The predicted value calculation unit 13 performs predetermined data analysis on each type of factor data stored in the data storage unit 211, and calculates a predicted value for each type of factor data to determine whether or not there is a sign of an event occurring. The calculated predicted values ​​are stored and accumulated in the predicted value storage unit 212.

[0085] The predicted value calculation unit 13 transmits various predicted values ​​to the TCU 120 via the server communication unit 201 in response to a signal requesting various predicted values ​​received from the TCU 120. The server 200 may transmit all of the predicted values ​​stored in the predicted value storage unit 212 simultaneously, or may transmit each type of predicted value at a different timing.

[0086] When the factor data is heart rate data, the predicted value calculation unit 13 calculates a predicted heart rate value for determining whether or not there is a sign of dozing off, based on the history of heart rate data accumulated in the data storage unit 211. The predicted value calculation unit 13 calculates, as the predicted heart rate value, a value that is larger by a predetermined value than the average heart rate indicated by the accumulated heart rate data (the average heart rate when dozing off occurs), for example.

[0087] When the factor data is driving time data, the predicted value calculation unit 13 calculates a predicted value of the length of driving time for determining whether there is a sign of at least one of inattentive driving and poor posture, based on the history of driving time data accumulated in the data storage unit 211. The predicted value calculation unit 13 calculates the predicted value of driving time based on, for example, a weight assigned to the driving time data and the number of pieces of driving time data corresponding to the number of occurrences of an event. The weighting may be set so that the shorter the driver's driving time, the larger the weight (score).

[0088] Furthermore, the predicted value calculation unit 13 may classify the driving time data based on the sleep state data before calculating the driving predicted value. This allows for more accurate determination of signs of an event occurring. Here, the classification based on the sleep state data may be any classification related to the driver's sleep time and sleep quality, and may be predefined or may be defined for each driver. Note that the classification based on the sleep state data may also be applied to physical condition data such as heart rate data as needed.

[0089] When the factor data is date and time data, the predicted value calculation unit 13 calculates a predicted value of a time period for determining whether there is a sign of at least one of inattentive driving and poor posture occurring, based on the history of date and time data accumulated in the data storage unit 211. The predicted value calculation unit 13 calculates a predicted value of a time period based on, for example, a weight assigned to the date and time data and the number of date and time data pieces corresponding to the number of times an event has occurred. The weighting may be set so that the weight (score) increases as the time period approaches when the driver's concentration is likely to decrease.

[0090] Furthermore, similarly to the case where the predicted value of the driving time is calculated, the predicted value calculation unit 13 may classify the date and time data based on the sleep state data before calculating the predicted value of the time period. This makes it possible to determine with higher accuracy whether there is a sign of an event occurring. The classification based on the sleep state data may be the same as when the predicted value of the driving time is calculated.

[0091] When the factor data is vehicle position data, the predicted value calculation unit 13 calculates a predicted value of the vehicle position for determining whether or not there is a sign of an event occurring, based on the history of vehicle position data accumulated in the data storage unit 211. The predicted value calculation unit 13 may, for example, compare the vehicle position data with map data, extract points such as conspicuous buildings or scenery where the driver is likely to be distracted or lose his / her posture, and calculate a value (coordinate) indicating the position of the extracted point as the predicted value of the vehicle position.

[0092] Furthermore, similarly to the case where the predicted value of the driving time is calculated, the predicted value calculation unit 13 may classify the vehicle position data based on the sleep state data before calculating the predicted value of the vehicle position. This makes it possible to determine with higher accuracy whether there is a sign of an event occurring. The classification based on the sleep state data may be the same as when the predicted value of the driving time is calculated.

[0093] It should be noted that, since the age, height, physique, etc. differ from driver to driver, it is considered that the tendency for an event to occur also differs depending on the age, height, physique, etc. When calculating various predicted values ​​in the predicted value calculation unit 13, it is advisable to take into account driver data indicating the driver's age, height, physique, etc., and obtain predicted values ​​that take into account changes in the tendency for an event to occur due to changes in age, height, physique, etc. This makes it possible to determine with even higher accuracy whether there are signs of an event occurring.

[0094] Furthermore, for example, when targeting a new driver with no accumulated data or a driver with a small amount of accumulated data, it is advisable to use driver data obtained from other drivers with similar age, height, physique, etc. In this case, it is advisable to confirm the correlation between the driver data obtained from the other drivers and the driver data obtained from the target driver before using the data. This makes it possible to obtain appropriate predicted values ​​for each driver even when targeting a new driver or a driver with a small amount of data, and as a result, it is possible to accurately determine whether there are signs of an event occurring for each driver.

[0095] The storage unit 203 may include a non-volatile storage device such as an HDD, an SSD, or a flash memory. The storage unit 203 includes a data storage unit 211 and a predicted value storage unit 212. The data storage unit 211 accumulates the sleep state data and various types of factor data (heart rate data, vehicle position data, driving time data, and date and time data) received from the TCU 120. The data storage unit 211 may classify each type of factor data according to the type of event, for example, and store each type of factor data.

[0096] The predicted value storage unit 212 accumulates various predicted values ​​calculated by the predicted value calculation unit 13. The predicted value storage unit 212 may classify each predicted value according to the type of event, for example, and store each predicted value for each type of predicted value.

[0097] Some or all of the components of each device may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits, etc., and programs. In addition to a CPU, a GPU, FPGA, etc. may be used as the processor.

[0098] Next, the processing (operation) of the driving assistance system 10 according to this embodiment will be described with reference to Fig. 5 to Fig. 8. Fig. 5 is a flowchart illustrating the processing of the driving assistance system according to the second embodiment. When the driver turns on the IG switch after getting into the vehicle 100, the driving assistance system 10 starts the processing shown in Fig. 5.

[0099] As shown in Fig. 5, when the IG switch is turned on, the TCU 120 receives an IG switch-on signal (step S11). In response to the IG switch-on signal, the TCU 120 performs processing to establish a communication connection, such as pairing processing using the short-range communication units 126 and 133, with the mobile terminal 130 worn by the driver, and starts communication with the mobile terminal 130 (step S12). Also, in response to the IG switch-on signal, the TCU 120 activates the DMS 110 (step S13). When the DMS 110 is activated, the authentication unit 117 performs personal authentication using face image data acquired from the monitoring sensor 111. Next, the TCU 120 acquires sleep state data indicating the sleep state of the previous day from the mobile terminal 130 using the data collection unit 12 (step S14).

[0100] The TCU 120 receives various predicted values ​​from the server 200 via the symptom determination unit 14 (step S15). If the driver has been identified by personal authentication in the authentication unit 117, in step S15 the TCU 120 acquires a predicted value for each driver according to the driver's identifier from the server 200. In addition, if classification based on sleep state data has been performed, in step S15 the TCU 120 acquires a predicted value according to the sleep state data acquired in step S14 from the server 200.

[0101] Thereafter, while the IG switch is on, the driving assistance system 10 repeatedly executes a symptom determination process (step S16) for determining whether there are symptoms of a specific event occurring as shown in FIG. 6, and an event detection process (step S17) for detecting the occurrence of a specific event as shown in FIG. 7.

[0102] First, the TCU 120 executes the symptom determination process shown in Fig. 6. Fig. 6 is a flowchart illustrating the symptom determination process. When the symptom determination process starts, first, in steps S21 to S23, the TCU 120 acquires various types of factor data from the mobile terminal 130, the positioning unit 123, and the clocking unit 122. The various types of factor data acquired by the TCU 120 here are heart rate data, vehicle position data, driving time data, and date and time data. Specifically, the TCU 120 acquires current heart rate data from the mobile terminal 130 using the data collecting unit 12 (step S21). Furthermore, the TCU 120 acquires current vehicle position data from the positioning unit 123 using the data collecting unit 12 (step S22). Furthermore, the TCU 120 acquires current driving time data and current date and time data from the clocking unit 122 using the data collecting unit 12 (step S23).

[0103] Next, the TCU 120 determines whether the heart rate indicated by the current heart rate data is lower than the predicted heart rate value (step S24) using the symptom determination unit 14. If the symptom determination unit 14 determines that the heart rate indicated by the heart rate data is lower than the predicted heart rate value (step S24: YES), the process proceeds to step S25.

[0104] Following step S24, the TCU 120 transmits to the DMS 110, by the sign determination unit 14, a determination result indicating that there are signs of dozing off. As a result, the TCU 120 instructs the DMS 110 to execute a notification (step S25). In response to the instruction from the TCU 120, the DMS 110 outputs a sound including a message notifying the driver of signs of dozing off from the notification unit 15 based on the control of the notification control unit 118 (step S26). As a result, the DMS 110 uses the notification unit 15 to provide the driver with a notification corresponding to the signs of dozing off.

[0105] On the other hand, if the symptom determination unit 14 determines that the heart rate indicated by the heart rate data is not lower than the predicted value of the heart rate (step S24: NO), the process proceeds to step S27. Following step S24, the TCU 120 determines whether at least one of the vehicle position indicated by the current vehicle position data, the driving time indicated by the driving time data, and the driving date and time indicated by the date and time data is approaching the corresponding predicted value (step S27).

[0106] In step S27, the TCU 120 determines, using the symptom determination unit 14, whether the vehicle position indicated by the vehicle position data is approaching the predicted value of the vehicle position. Also, the TCU 120 determines, using the symptom determination unit 14, whether the driving time indicated by the driving time data is approaching the predicted value of the driving time. Furthermore, the TCU 120 determines, using the symptom determination unit 14, whether the driving date and time indicated by the date and time data is approaching the predicted value of the time period.

[0107] If the symptom determination unit 14 determines that at least one of the vehicle position, driving time, and driving date and time is approaching the corresponding predicted value (step S27: YES), the process proceeds to step S28. If the symptom determination unit 14 determines that all of the vehicle position, driving time, and driving date and time are not approaching the corresponding predicted value (step S27: NO), the TCU 120 ends the symptom determination process, and the process proceeds to step S31 in FIG.

[0108] Following step S27, the TCU 120 transmits to the DMS 110 the determination result, by the symptom determination unit 14, indicating that there is a symptom of at least one of inattentive driving and poor posture. As a result, the TCU 120 instructs the DMS 110 to execute a notification (step S28). In response to the instruction from the TCU 120, the DMS 110, under the control of the notification control unit 118, outputs a voice message from the notification unit 15 that includes a message notifying the driver of a symptom of at least one of inattentive driving and poor posture (step S29). As a result, the DMS 110 uses the notification unit 15 to notify the driver of the symptom of drowsiness. Thereafter, the TCU 120 ends the symptom determination process, and the process proceeds to step S31 in FIG. 7.

[0109] The audio message output by the notification unit 15 may have different content depending on the type of event, or may have the same content. An example of a message notifying of a sign that an event is about to occur is a message urging the driver to take a break.

[0110] Following step S27 or S29, the DMS 110 executes the event detection process shown in Fig. 7. Fig. 7 is a flowchart illustrating the event detection process. When the event detection process starts, the DMS 110 first determines, by the event detection unit 11, whether or not at least one type of event, namely, dozing, looking away, or poor posture, has occurred (step S31). In step S31, the event detection unit 11 determines, based on the face image data of the driver, whether or not an event has occurred to the driver. If the event detection unit 11 determines that an event has occurred to the driver (step S31: YES), the process proceeds to step S32.

[0111] Following step S31, the DMS 110 causes the event detection unit 11 to transmit event detection data indicating the type of event and the date and time when the event occurred to the TCU 120 (step S32). If the driver has been identified by personal authentication in the authentication unit 117, the event detection data transmitted by the DMS 110 is associated with the identifier of the identified driver.

[0112] Upon receiving the event detection data from the DMS 110, the TCU 120 transmits to the server 200 various factor data acquired when an event is detected, together with the sleep state data acquired at the start of driving (step S33). The various factor data transmitted by the TCU 120 here include heart rate data, vehicle position data, driving time data, and date and time data. If the driver has been identified by personal authentication in the authentication unit 117, the various data transmitted by the TCU 120 are associated with the identifier of the identified driver. Thereafter, the DMS 110 ends the event detection process, and the process proceeds to step S18 in FIG. 5.

[0113] On the other hand, if the event detection unit 11 determines that no event has occurred to the driver (step S31: NO), the DMS 110 ends the event detection process, and the process proceeds to step S18 in FIG. 5. Following step S31 or S33, the TCU 120 determines whether the IG switch has been turned off based on a signal received from the IG switch (step S18). When the TCU 120 receives an IG switch-off signal, it determines that the IG switch has been turned off (step S18: YES), and the series of processes for this cycle ends. On the other hand, when the TCU 120 does not receive an IG switch-off signal, it determines that the IG switch remains on (step S18: NO), and the process returns to step S16. The driving assistance system 10 repeats the processes of steps S16 and S17 until the IG switch is turned off.

[0114] When the series of processes in the driving assistance system 10 ends in step S18, the processes of the driving assistance system 10 shown in FIG. 5 are started again when the IG switch is turned on next time, and the driving assistance system 10 executes the processes of steps S11 to S18.

[0115] In the processing of the driving assistance system 10 in response to the next turning on of the IG switch, in step S15, the TCU 120 receives the latest various predicted values ​​stored in the predicted value storage unit 212 from the server 200. Then, the TCU 120 executes the symptom determination processing in step S16 using the various updated predicted values.

[0116] Following step S33, server 200, which has received various data from TCU 120, performs a predicted value calculation process shown in Fig. 8 to calculate a predicted value. Fig. 8 is a flowchart illustrating the predicted value calculation process. When a driver is identified through personal authentication in authentication unit 117, server 200 stores various data for each driver in data storage unit 211 based on the driver's identifier, and executes the predicted value calculation process. This makes it possible to determine with even higher accuracy whether there are signs of an event occurring for each driver.

[0117] First, the server 200 classifies various data received from the TCU 120 by event type and stores the data in the data storage unit 211 (step S41). In step S41, the server 200 stores, for example, heart rate data in an area corresponding to "dozing" in the data storage unit 211. Also in step S41, the server 200 stores, for example, sleeping state data, vehicle position data, driving time data, and date and time data in an area corresponding to "distraction and poor posture" in the data storage unit 211.

[0118] Next, the server 200 calculates, by the predicted value calculation unit 13, a predicted value of the heart rate for determining whether or not there is a sign of falling asleep, based on the history of the accumulated heart rate data (step S42).

[0119] Next, the server 200 causes the predicted value calculation unit 13 to classify the accumulated driving time data, date and time data, and vehicle position data based on the sleep state data (step S43). Here, the predicted value calculation unit 13 is set with four sleep state categories, for example, based on whether the sleep time is "6 hours or more" or "less than 6 hours" and whether the sleep quality is "good" or "bad". Then, the predicted value calculation unit 13 classifies the accumulated driving time data, date and time data, and vehicle position data into a sleep state category selected from the four sleep state categories based on the sleep state data.

[0120] Next, the server 200 calculates a predicted value of the driving time based on the history of the accumulated driving time data using the predicted value calculation unit 13 to determine whether there are signs of inattentiveness and poor posture (step S44).

[0121] In step S44, the predicted value calculation unit 13 weights the driving time data so that, for example, the shorter the driver's driving time, the higher the score. Then, the predicted value calculation unit 13 classifies the driving time data into one of a plurality of arbitrarily set driving time segments according to the driving time indicated by the driving time data. Thereafter, the predicted value calculation unit 13 multiplies the number of occurrences of events (the number of data points in the driving time data) tallied for each driving time segment by the weight (score), and calculates the driving time of the driving time segment with the highest evaluation score as the predicted value of the driving time.

[0122] Fig. 9 is a diagram illustrating an example of a method for calculating a predicted value of driving time. In the example shown in Fig. 9, driving time is divided into 4 hours or more, 3 to 4 hours, 1 to 2 hours, and 0 to 1 hour, and 4 hours or more is set to 1 point, 3 to 4 hours to 2 points, 2 to 3 hours to 3 points, 1 to 2 hours to 4 points, and 0 to 1 hour to 5 points.

[0123] For example, if the driving time data indicates that at least one of inattentive driving and poor driving posture is detected when the driving time is three hours, the number of occurrences for the 3-4 hour driving time segment is incremented by 1. Therefore, the number of occurrences for that driving time segment is updated from 3 to 4. With this increase in the number of occurrences, the evaluation score for the 3-4 hour driving time segment is updated from 6 to 8. In the example shown in FIG. 9, the evaluation score for the 3-4 hour driving time segment is the highest out of the five driving time segments, so three hours is ultimately calculated as the predicted driving time.

[0124] Next, the server 200 calculates, using the predicted value calculation unit 13, a predicted value for a time period for determining whether there are signs of at least one of inattentiveness and poor posture occurring, based on the accumulated date and time data (step S45).

[0125] In step S45, the predicted value calculation unit 13 weights the date and time data so that, for example, the closer the data is to a time period when the driver's concentration is likely to decrease, the higher the score.The predicted value calculation unit 13 then classifies the date and time data into one of a plurality of arbitrarily set time period segments according to the date and time indicated by the date and time data.The predicted value calculation unit 13 then multiplies the number of occurrences of events (the number of data points in the date and time data) tallied for each time period segment by the weight (score), and calculates the time period of the time period segment with the highest evaluation score as the predicted value for the time period.

[0126] Therefore, Fig. 10 is a diagram illustrating an example of a method for calculating predicted values ​​for time periods. In the example shown in Fig. 10, the time period from 2 to 3 p.m., when it is thought that a driver's concentration is most likely to decline, is set to 5 points. In addition, in the example shown in Fig. 10, the time periods from 1 to 2 p.m. and 3 to 4 p.m. are each set to 4 points, the time periods from 1 to 2 p.m. and 4 to 5 p.m. are each set to 3 points, and the time periods from 12 to 1 p.m. and 4 to 5 p.m. are each set to 2 points. Furthermore, in the example shown in Fig. 10, the time periods before 12 p.m. and the time periods after 5 p.m. are each set to 1 point.

[0127] For example, if the driving date and time indicated by the date and time data is 12:30 and at least one of inattentive driving and poor posture is detected, the number of occurrences for the 12:00 to 1:00 pm time slot is incremented by 1. Therefore, the number of occurrences for that time slot is updated from 2 to 3. With this increase in the number of occurrences, the evaluation score for the 12:00 to 1:00 pm time slot is updated from 4 to 6. In the example shown in FIG. 10, the evaluation score for the 14:00 to 1:00 pm time slot is finally the highest out of the five time slots, so 14:00 to 1:00 pm is calculated as the predicted value for the time slot.

[0128] Next, the server 200 calculates a predicted value of the vehicle position based on the history of the accumulated vehicle position data using the predicted value calculation unit 13 to determine whether there are signs of at least one of inattentive driving and poor posture (step S46).

[0129] Then, the server 200 stores the calculated various predicted values ​​(predicted heart rate value, predicted driving time value, predicted time zone value, and predicted vehicle position value) in the predicted value storage unit 212 (step S47). When the driver is identified by personal authentication in the authentication unit 117, the server 200 stores the various predicted values ​​for each driver in the predicted value storage unit 212 based on the driver's identifier. In this way, the server 200 calculates and stores various predicted values ​​used in the symptom determination process of the driving assistance system 10 when the IG switch is next turned on, through the predicted value calculation process of steps S41 to S47. Then, the server 200 can supply these predicted values ​​in their latest state to the TCU 120 in response to a signal requesting the various predicted values.

[0130] As described above, in this embodiment, similar to the first embodiment, whether or not there is a sign of an event occurring is determined using a predicted value of the occurrence factor obtained by learning the tendency of the occurrence factor when the event occurs from a history that accumulates factor data acquired when the occurrence of the event is detected. Furthermore, in this embodiment, various data including various factor data are collected, and whether or not there is a sign of various events occurring is determined using a predicted value calculated from the various collected data. This makes it possible to determine with high accuracy whether or not there is a sign of an event occurring for each event. Furthermore, in this embodiment, it is possible to notify the driver of the presence of a sign of various events occurring. Therefore, according to this embodiment, by notifying the driver of the presence of a sign of a specific event occurring before the event occurs, it is possible to prevent accidents from occurring.

[0131] Note that there are no particular limitations on the entity or device that performs each process. The driving assistance system 10 may not include a server 200 provided outside the vehicle 100, and an in-vehicle device such as the TCU 120 may have functions similar to those of the control unit 202 and storage unit 203 of the server 200. In this case, the driving assistance system 10 is implemented within the vehicle 100. Furthermore, processing may be performed by a processor or the like incorporated in an in-vehicle device other than the TCU 120. However, from the viewpoint of reducing the processing load, etc., on the in-vehicle device, it is preferable to include the server 200.

[0132] Other embodiments Although the above-described embodiment has been described as a hardware configuration, the present disclosure is not limited to this. Any processing in the present disclosure can also be realized by causing a CPU to execute a computer program.

[0133] In the above example, the program can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic storage media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical storage media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, DVDs (Digital Versatile Discs), and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can be supplied to a computer via wired communication paths such as electrical wires and optical fibers, or via wireless communication paths.

[0134] The present disclosure is not limited to the above-described embodiments, and may be modified as appropriate without departing from the spirit and scope of the present disclosure. In addition, the present disclosure may be implemented by appropriately combining the respective embodiments.

[0135] Some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes. (Appendix A1) a first in-vehicle device including an event detection unit that detects a specific event that occurs to a driver while driving a vehicle; a data collection unit that acquires, as factor data, at least one of physical condition data indicating a physical condition of the driver that is a cause of the event and driving situation data indicating a driving situation of the driver that is a cause of the event; and a server connected to the second in-vehicle device via a network; Equipped with the second in-vehicle device transmits the acquired cause data to the server via the network when the event is detected; the server includes a predicted value calculation unit that calculates a predicted value for determining whether or not there is a sign that the event will occur based on a history of accumulated cause data received from the second in-vehicle device; the second in-vehicle device includes a symptom determination unit that compares the current factor data with the predicted value received from the server via the network to determine whether or not there is a symptom of the event occurring; The first in-vehicle device is a driving assistance system including a notification unit that, when it is determined that there is a sign that the event will occur, notifies the driver of the determination result. (Appendix A2) the event is falling asleep, The driving assistance system according to Appendix A1, wherein the physical condition data is heart rate data. (Appendix A3) the predicted value calculation unit calculates a value that is larger by a predetermined value than an average value of the heart rate indicated by the accumulated heart rate data as the predicted value of the heart rate; The driving assistance system according to appendix A2, wherein the symptom determination unit determines that there is a symptom of drowsiness occurring when the heart rate indicated by the current heart rate data is lower than the predicted heart rate value. (Appendix A4) the event is at least one of inattentiveness and poor posture, The driving assistance system according to Appendix A1, wherein the driving situation data is at least one of driving time data, date and time data, and vehicle position data. (Appendix A5) When the factor data is the driving time data, the predicted value calculation unit calculates the predicted value of the driving time based on a weight assigned to the accumulated driving time data and the number of data items of the driving time data; the sign determination unit determines that there is a sign of at least one of the inattentive driving and the poor posture occurring when the driving time indicated by the current driving time data is approaching the predicted value of the driving time, When the factor data is the date and time data, the predicted value calculation unit calculates the predicted value of the time period based on a weight assigned to the accumulated date and time data and the number of data items in the date and time data; the sign determination unit determines that there is a sign of at least one of the inattentive driving and the poor posture occurring when the driving date and time indicated by the current date and time data is close to a predicted value for the time period; When the factor data is the vehicle position data, the predicted value calculation unit extracts points where at least one of the inattentive driving and the poor posture is likely to occur based on the accumulated vehicle position data and map data, and calculates a value indicating the position of the point as the predicted value of the vehicle position; The driving assistance system described in Appendix A4, wherein the symptom determination unit determines that there is a symptom of at least one of the inattentiveness and the poor posture occurring when the vehicle position indicated by the current vehicle position data is approaching a predicted value of the vehicle position. (Appendix A6) the data collection unit acquires activity state data indicating an activity state of the driver before driving that is a cause of the event; The driving assistance system according to Appendix A1, wherein the predicted value calculation unit calculates the predicted value using the factor data classified based on the activity state data. (Appendix A7) The driving assistance system according to Appendix A6, wherein the activity state data is sleep state data. (Appendix A8) The driving assistance system according to Appendix A1, wherein the event detection unit detects the event based on facial image data of the driver. (Appendix B1) Detecting a specific event occurring to a driver while driving a vehicle; acquiring, as factor data, at least one of physical condition data indicating a physical condition of the driver that is a factor causing the event and driving situation data indicating a driving situation of the driver that is a factor causing the event; calculating a predicted value for determining whether or not there is a sign that the event will occur based on a history of accumulated cause data acquired when the event is detected; a step of comparing the current factor data with the predicted value to determine whether there is a sign that the event will occur; a step of notifying the driver of a result of the determination when it is determined that there is a sign that the event will occur; A driving assistance method comprising: (Appendix C1) A process of detecting a specific event occurring to a driver while driving a vehicle; a process of acquiring, as factor data, at least one of physical condition data indicating a physical condition of the driver that is a factor causing the event and driving situation data indicating a driving situation of the driver that is a factor causing the event; a process of calculating a predicted value for determining whether or not there is a sign that the event will occur based on a history of accumulated cause data acquired when the event is detected; a process of comparing the current factor data with the predicted value to determine whether there is a sign that the event will occur; a process of notifying the driver of a result of the determination when it is determined that there is a sign that the event will occur; A program that causes a computer to execute the following. (Appendix D1) an event detection unit that detects a specific event that occurs to a driver while driving a vehicle; a data collection unit that acquires, as factor data, at least one of physical condition data indicating a physical condition of the driver that is a factor causing the event and driving situation data indicating a driving situation of the driver that is a factor causing the event; a predicted value calculation unit that calculates a predicted value for determining whether or not there is a sign that the event will occur based on a history that accumulates the cause data acquired when the event is detected; and a sign determination unit that compares the current factor data with the predicted value to determine whether or not there is a sign that the event will occur; a notification unit that notifies the driver of a determination result when it is determined that there is a sign that the event will occur; A driving assistance device comprising: [Explanation of symbols]

[0136] 1 Driving assistance devices 10. Driving assistance systems 11 Event detection unit 12 Data collection unit 13 Prediction value calculation unit 14 Symptom Judgment Section 15 Notification Section 100 vehicles 110 DMS 111 monitoring sensor 113 in-vehicle communication unit 114 control unit 115 memory unit 116 image data acquisition unit 117 authentication unit 118 notification control unit 120 TCU 121 Input / output unit 122 Timekeeping unit 123 Positioning unit 124 External communication unit 125 Internal communication unit 126 Short-range communication unit 127 control unit 128 memory unit 130 mobile devices 131 Biometric sensor 132 Activity management unit 133 Short-range communication unit 134 control unit 135 memory unit 136 Physical condition data acquisition unit 137 Activity condition data acquisition unit 200 servers 201 Server communication unit 202 Control unit 203 Storage unit 211 Data storage unit 212 Predicted value storage unit BS wireless base station NW network

Claims

1. a first in-vehicle device including an event detection unit that detects a specific event that occurs to a driver while driving a vehicle; a second in-vehicle device including a data collection unit that acquires, as factor data, at least one of physical condition data indicating a physical condition of the driver that is a cause of the event and driving situation data indicating a driving situation of the driver that is a cause of the event; a server connected to the second in-vehicle device via a network; Equipped with the second in-vehicle device transmits the acquired cause data to the server via the network when the event is detected; the server includes a predicted value calculation unit that calculates a predicted value for determining whether or not there is a sign of the event occurring based on a history of accumulated cause data received from the second in-vehicle device; the second in-vehicle device includes a symptom determination unit that compares the current factor data with the predicted value received from the server via the network to determine whether or not there is a symptom of the event occurring; the first in-vehicle device includes a notification unit that, when it is determined that there is a sign that the event will occur, notifies the driver of the determination result; the data collection unit acquires activity state data indicating an activity state of the driver before driving that is a cause of the event; The predicted value calculation unit calculates the predicted value according to the activity state.

2. the event is falling asleep, The driving assistance system according to claim 1 , wherein the physical condition data is heart rate data.

3. the predicted value calculation unit calculates a value that is larger by a predetermined value than an average value of the heart rate indicated by the accumulated heart rate data as the predicted value of the heart rate; The driving assistance system according to claim 2 , wherein the symptom determination unit determines that there is a symptom of drowsiness when the heart rate indicated by the current heart rate data is lower than the predicted heart rate value.

4. the event is at least one of inattentiveness and poor posture, The driving assistance system according to claim 1 , wherein the driving situation data is at least one of driving time data, date and time data, and vehicle position data.

5. When the factor data is the driving time data, the predicted value calculation unit calculates the predicted value of the driving time based on a weight assigned to the accumulated driving time data and the number of data items of the driving time data; the sign determination unit determines that there is a sign of at least one of the inattentive driving and the poor posture occurring when the driving time indicated by the current driving time data is approaching the predicted value of the driving time, When the factor data is the date and time data, the predicted value calculation unit calculates the predicted value of the time period based on a weight assigned to the accumulated date and time data and the number of data items in the date and time data; the sign determination unit determines that there is a sign of at least one of the inattentive driving and the poor posture occurring when the driving date and time indicated by the current date and time data is close to a predicted value for the time period; When the factor data is the vehicle position data, the predicted value calculation unit extracts points where at least one of the inattentive driving and the poor posture is likely to occur based on the accumulated vehicle position data and map data, and calculates a value indicating the position of the point as the predicted value of the vehicle position; 5. The driving assistance system according to claim 4, wherein the symptom determination unit determines that there is a symptom of at least one of the inattentive driving and the poor posture occurring when the vehicle position indicated by the current vehicle position data is approaching a predicted value of the vehicle position.

6. 2. The driving assistance system according to claim 1, wherein the predicted value calculation unit classifies the factor data accumulated based on the activity state indicated by the activity state data into one of a plurality of activity state categories divided based on the degree of likelihood of the event occurring, and calculates the predicted value using the classified factor data.

7. The driving assistance system according to claim 6 , wherein the activity status data is sleep status data.

8. The driving assistance system according to claim 1 , wherein the event detection unit detects the event based on facial image data of the driver.

Images