Vehicle notification device

Abstract

To make it possible to properly keep a condition where a driver is ready to return to the driving operation within a specified time during automated driving without the necessity of constantly monitoring the driver during the automated driving, and to prompt the driver to improve his or her driving characteristics.SOLUTION: A vehicle notification device is adapted to an automated driving vehicle. The vehicle notification system includes one or more processors. The one or more processors are configured to execute a notification process of notifying a driver of at least eyes-on request out of hands-on request and the eyes-on request every time a notification interval elapses. The one or more processors are also configured to calculate the notification interval based on driving characteristic information of the driver.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a vehicle notification device applied to an autonomous vehicle.

Background Art

[0002] Patent Document 1 discloses a driver monitoring system. In this driver monitoring system, when the elapsed time since a hands-on request or a surrounding confirmation request is notified exceeds a determination reference time, the hands-on request or the surrounding confirmation request is notified again. Further, when a sign related to a decrease in the driver's attention is detected, a warning is given to the driver, and the determination reference time is set to be shorter than before the warning is given.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] According to autonomous driving that enables hands-off and eyes-off, a driver released from driving operations is allowed to perform a secondary activity, which is an act other than driving operations, while the vehicle is running. On the other hand, during the autonomous driving, the vehicle system always grasps that the driver can return to the driving operation before a specified time elapses after a driving handover request is issued, and is required to maintain a state (driving preparation state) in which the driver is ready to return to the driving operation as such.

[0005] If an attempt is made to maintain the above-described driving preparation state by constantly monitoring the driver during automated driving, it becomes necessary to restrict secondary activities or the cost required for monitoring increases. For this reason, it is desirable to be able to appropriately maintain the driving preparation state without constantly monitoring the driver during automated driving using a monitoring device such as a driver monitor. Further, it is desirable to be able to encourage the driver to improve driving characteristics by having the driver maintain the driving preparation state.

[0006] The present disclosure has been made in view of the above-described problems, and an object thereof is to provide a vehicle notification device that can appropriately continue a state in which a driver is prepared to return to a driving operation within a specified time during the automated driving without constantly monitoring the driver during the automated driving, and can encourage the driver to improve driving characteristics.

Means for Solving the Problems

[0007] The vehicle notification device according to the present disclosure is applied to an automated driving vehicle. The vehicle notification device includes one or more processors. The one or more processors execute a notification process of notifying the driver of at least an eyes-on request out of a hands-on request and an eyes-on request each time a notification interval elapses. Then, the one or more processors calculate the notification interval based on the driving characteristic information of the driver.

Effects of the Invention

[0008] According to the present disclosure, it is possible to appropriately continue a state in which a driver is prepared to return to a driving operation within a specified time during the automated driving without constantly monitoring the driver during the automated driving, and to encourage the driver to improve driving characteristics.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Mode for Carrying Out the Invention

[0010] 1. Configuration Example of an Autonomous Vehicle FIG. 1 is a diagram schematically showing an example of the configuration of an autonomous vehicle 1 to which a vehicle notification device according to an embodiment is applied. The autonomous vehicle 1 is configured to be capable of performing level 3 autonomous driving as defined by the Society of Automotive Engineers (SAE) in the United States. That is, the vehicle 1 is equipped with a hands-off function and an eyes-off function. In the following description, level 3 autonomous driving is simply also referred to as autonomous driving.

[0011] For example, the vehicle 1 includes an HMI (Human Machine Interface) device 10, a driver monitor 20, a sensor group 30, a traveling device 40, and an electronic control unit (ECU) 50.

[0012] The HMI device 10 provides information to the driver 2 of the vehicle 1 and also receives information from the driver 2. For example, the HMI device 10 includes an input device 12, a speaker 14, and a display device 16. The input device 12 is, for example, a touch panel or a switch. The driver 2 can use the input device 12 to input information such as a destination and switch the on / off of vehicle travel by level 3 autonomous driving. Each piece of information from the driver 2 via the input device 12 is transmitted to the ECU 50. The speaker 14 provides audio information to the driver 2. The display device 16 provides visual information to the driver 2. For example, the display device 16 is a HUD (Head Up Display), a MID (Multi Information Display), or a meter panel. Also, at least a part of the HMI device 10 does not necessarily have to be mounted on the vehicle 1 and may include, for example, a portable terminal (e.g., a smartphone or a tablet terminal) operated by the driver 2.

[0013] The driver monitor 20 includes, for example, a camera installed in the interior of the vehicle 1 so as to image the driver 2. By analyzing the image obtained by this camera, the driver 2 (individual) can be identified, and the state and actions of the driver 2 such as the face orientation, line of sight, and eye opening degree can be detected. Therefore, the driver monitor 20 can be used to detect whether the driver 2's eyes are on / off. Further, the driver monitor 20 includes, for example, a steering touch sensor installed on the steering wheel 3 operated by the driver 2. Thereby, it is possible to detect whether the driver 2 has hands on / hands off. The information acquired by the driver monitor 20 is transmitted to the ECU 50.

[0014] The sensor group 30 includes a recognition sensor, a vehicle state sensor, a position sensor, etc. The recognition sensor recognizes (detects) the situation around the vehicle 1. Examples of the recognition sensor include a camera, LIDAR (Laser Imaging Detection and Ranging), a radar, etc. The vehicle state sensor detects the state of the vehicle 1. The vehicle state sensor includes a speed sensor, an acceleration sensor, a yaw rate sensor, a steering angle sensor, an accelerator pedal sensor, a brake pedal sensor, etc. The position sensor detects the position and orientation of the vehicle 1. For example, the position sensor includes a GNSS (Global Navigation Satellite System) receiver. Further, the sensor group 30 includes, for example, a sensor for detecting the wearing of the seat belt of the driver 2 and a sensor for detecting the seating position and seating posture of the driver 2.

[0015] The traveling device 40 includes a steering device, a driving device, and a braking device. The steering device steers the wheels. For example, the steering device includes an electric power steering (EPS) device. The driving device is a power source that generates a driving force. For example, the driving device includes at least one of an internal combustion engine and an electric motor. The braking device generates a braking force.

[0016] The ECU 50 is a computer that controls the vehicle 1. The ECU 50 includes one or more processors 52 (hereinafter simply referred to as the processor 52) and one or more storage devices 54 (hereinafter simply referred to as the storage device 54). The processor 52 executes various processes. The various processes include processes related to autonomous driving control and the "notification process" described later. For example, the processor 52 includes a CPU (Central Processing Unit). The storage device 54 stores various information. Examples of the storage device 54 include a volatile memory, a non-volatile memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), and the like. Note that the ECU 50 may be configured by combining a plurality of ECUs.

[0017] By the processor 52 executing a computer program, various processes by the ECU 50 are realized. The computer program is stored in the storage device 54. Alternatively, the computer program may be recorded on a computer-readable recording medium.

[0018] The ECU 50 uses the sensor group 30 to acquire driving environment information indicating the driving environment of the vehicle 1. The driving environment information is stored in the storage device 54. Specifically, the driving environment information includes surrounding situation information indicating the recognition result by the recognition sensor. For example, the surrounding situation information includes an image captured by a camera and object information regarding an object around the vehicle 1. Further, the driving environment information includes vehicle state information indicating the vehicle state detected by a vehicle state sensor. Furthermore, the driving environment information includes vehicle position information indicating the position and orientation of the vehicle 1. The vehicle position information is obtained by, for example, a position sensor. Also, the driving environment information includes information obtained by the driver monitor 20.

[0019] The ECU 50 executes vehicle driving control for controlling the driving of the vehicle 1. The vehicle driving control includes steering control, drive control, and braking control. The ECU 50 executes vehicle driving control by controlling the driving device 40 (steering device, drive device, and braking device). Further, the ECU 50 performs autonomous driving control based on driving environment information. For example, the ECU 50 generates a driving plan for the vehicle 1 based on the driving environment information. Furthermore, the ECU 50 generates a target trajectory necessary for the vehicle 1 to travel according to the driving plan based on the driving environment information. The target trajectory includes a target position and a target speed. Then, the ECU 50 performs vehicle driving control so that the vehicle 1 follows the target trajectory.

[0020] 2. Notification Process According to level 3 autonomous driving, it is possible to take eyes off from the front of the vehicle together with hands off from the steering wheel 3. The driver 2 released from the driving operation by being able to take eyes off is allowed to perform a secondary activity (for example, operation of a mobile terminal or viewing of a video), which is an act other than the driving operation, during the driving of the vehicle 1. However, this is conditional on the driver 2 meeting the driving handover demand (TD: Transition Demand). More specifically, the driving handover demand is issued when the system of the vehicle 1 enters a situation that it cannot handle during autonomous driving (for example, outside the range of the operational design domain (ODD), bad weather, or system failure).

[0021] The system of the vehicle 1 always grasps that the driver 2 can return to the driving operation within a specified time (for example, the specified time T4 described later) after the driving handover demand is issued, and maintains a state (driving preparation state) in which the driver 2 is prepared to return to the driving operation in such a manner.

[0022] In order to always be able to monitor the state and operation of the driver 2 using the driver monitor 20 during automatic driving to maintain the above-mentioned driving readiness state, it may be necessary to limit secondary activities that are difficult to monitor (for example, reading, operating a smartphone, eating, or using a personal computer (PC)). Conversely, if an attempt is made to eliminate blind spots for the expansion of secondary activities, it may be necessary to acquire camera images from various angles for driver monitoring. This leads to an increase in vehicle costs.

[0023] Also, it is desirable to be able to encourage the driver 2 to improve driving characteristics by having the driver 2 maintain the driving readiness state.

[0024] In view of the above problems, in this embodiment, the ECU 50 (processor 52) executes the following "notification process". That is, in the notification process, the ECU 50 notifies the driver 2 of the hands-on request and the eyes-on request each time the notification interval (more specifically, the notification time interval) T1 elapses. More specifically, this notification requests the driver 2 to perform hands-on and eyes-on within a specified time (for example, the specified time T4). And this notification interval T1 is calculated based on the driving characteristic information of the driver 2. More specifically, the notification interval T1 is changed according to the driving characteristic information.

[0025] FIG. 2 is a flowchart showing an example of a process related to the notification process according to the embodiment. The process of this flowchart starts, for example, when the execution conditions for level 3 automatic driving control are satisfied.

[0026] In step S100, the ECU 50 (processor 52) executes a personal authentication process using the driver monitor 20 to identify the driver 2. Further, the ECU 50 sets (calculates) the above notification interval T1. The notification interval T1 corresponds to a time interval for repeatedly confirming that the driving preparation state of the driver 2 is maintained during autonomous driving. More specifically, the notification interval T1 corresponds to an expected value (estimated value) of the time from the start of hands-off and eyes-off until the driving preparation state begins to decline.

[0027] The calculation of the notification interval T1 in step S100 is executed as follows, for example. That is, the notification interval T1 is calculated by multiplying the basic value T1b by the correction coefficient K as shown in Equation (1). As the basic value T1b, for example, a pre-determined value (e.g., 3 minutes) is used. The correction coefficient K is determined based on the response time T2 of the driver 2 and the driver information DI of the driver 2. More specifically, the correction coefficient K is, for example, the product of the correction coefficient K1 based on the response time T2 and the correction coefficient K2 based on the driver information DI as shown in Equation (2). T1 = T1b × K ···(1) K = K1 × K2 ···(2)

[0028] The response time T2 corresponds to the time from the start of the hands-on and eyes-on requests until the detection of both hands-on and eyes-on. For the calculation of the correction coefficient K1, for example, the average response time T2m is used. The average response time T2m is the average value of the response times T2 measured a predetermined number of times in the past for the driver 2. More specifically, the average response time T2m may be, for example, the moving average value of the response times T2 measured a predetermined number of times most recently.

[0029] The correction coefficient K1 is calculated, for example, as a value corresponding to the reciprocal of the average response time T2m. More specifically, when the average response time T2m is equal to a specified time T4 described later, the correction coefficient K1 is set to 1. When the average response time T2m is shorter than the specified time T4, the correction coefficient K1 is calculated within a range greater than 1 such that it increases as the average response time T2m becomes shorter. On the other hand, when the average response time T2m is longer than the specified time T4, the correction coefficient K1 is calculated within a range of 0 or more and less than 1 such that it decreases as the average response time T2m becomes longer. Note that, for the calculation of the correction coefficient K1, instead of the average response time T2m, for example, the value of the response time T2 measured most recently may be used.

[0030] The above driver information DI regarding the calculation of the correction coefficient K2 includes the driving characteristic information of driver 2. The driving characteristic information includes, for example, information regarding the dangerous driving of driver 2. This information includes, for example, the number of times of actuation of the emergency brake automatically performed by the braking device of vehicle 1. Further, the information includes, for example, the number of times of each of sudden acceleration, sudden deceleration, sudden steering, serpentine driving, and driving on the shoulder by driver 2. Furthermore, the information includes the number of violations regarding the driving support function during vehicle travel when the level 3 automated driving function is not being used. This violation is, for example, that driver 2 releases their hand from the steering wheel 3 during the use of a driving support function such as Adaptive Cruise Control (ACC). Such various driving characteristic information is acquired using, for example, the sensor group 30 and the driver monitor 20 during vehicle travel when the level 3 automated driving function is not being used, and is stored in the storage device 54.

[0031] For example, the correction coefficient K2 is set to 1 or a predetermined value greater than 1 when there is no history of the above-mentioned dangerous driving. Then, as the number of each of the above-mentioned dangerous driving acts increases, the correction coefficient K2 is calculated to become smaller toward 0. Note that, depending on the content of each dangerous driving act, the amount of decrease in the correction coefficient K2 for one dangerous driving may be changed.

[0032] In addition, the driver information DI may include driving-related information such as the number of times of looking aside and the number of times of closing eyes of the driver 2 together with the above-described vehicle characteristic information. And, for example, the correction coefficient K2 may be calculated so as to become smaller toward 0 as the number of times of looking aside and closing eyes increases. Further, the driver information DI may include personal information (for example, age) of the driver 2. And, for example, the correction coefficient K2 may be calculated so as to increase as the age is younger. Furthermore, the driver information DI may include, as a history of the response time T2, the probability that the response time T2 falls within the specified time T4. And when the probability is higher than the threshold value, the correction coefficient K2 may be corrected so as to increase within a range greater than 1.

[0033] The notification interval T1 calculated as described above in this step S100 is stored in the storage device 54. In addition, when data regarding the response time T2 and the driver information DI are not stored in the storage device 54, an initial value determined in advance may be used as the notification interval T1.

[0034] In step S102 following step S100, the ECU 50 determines whether or not automatic driving (automatic driving control) is being executed. As a result, if automatic driving is being executed (step S102; Yes), the process proceeds to step S104. On the other hand, when automatic driving is not being executed (step S102; No), the process of this flowchart ends.

[0035] In step S104, the ECU 50 determines whether or not the notification interval T1 starting from the start time of automatic driving or the elapsed time of the previous notification interval T1 has elapsed. As a result, when the notification interval T1 has elapsed (step S104; Yes), the process proceeds to step S106. On the other hand, when the notification interval T1 has not elapsed (step S104; No), the process returns to step S102.

[0036] In step S106, the ECU 50 uses the HMI device 10 to notify the driver 2 of the hands-on request and the eyes-on request. That is, the ECU 50 executes the above-described notification process. As methods for requesting hands-on and eyes-on, for example, requests by voice using the speaker 14, displays of requests on the screen of the display device 16, requests by vibration of the seat belt, or requests by sound and messages using the driver 2's portable terminal are applicable.

[0037] In step S108 following step S106, the ECU 50 determines whether both hands-on and eyes-on of the driver 2 corresponding to the hands-on request and the eyes-on request are detected. As a result, when both hands-on and eyes-on are detected (step S108; Yes), the process proceeds to step S110. Note that if one or both of hands-on and eyes-on are not detected until a predetermined time longer than the upper limit time T3 described later elapses from the start of the hands-on and eyes-on requests, the process may proceed to step S114.

[0038] In step S110, the ECU 50 calculates the response time T2 for the hands-on and eyes-on requests by the process of step S106. The response time T2 is calculated, for example, as the above-described average response time T2m. The calculated response time T2 is stored in the storage device 54.

[0039] In step S112 following step S110, the ECU 50 determines whether the calculated response time T2 is longer than the upper limit time T3 (for example, 10 seconds). As a result, when the response time T2 exceeds the upper limit time T3 (step S112; Yes), the process proceeds to step S114.

[0040] In step S114, the ECU 50 terminates the automatic driving control. That is, as a penalty for the driver 2 not performing hands-on and eyes-on within the upper limit time T3, the use of the level 3 automatic driving function is restricted. More specifically, for example, the driver 2 is notified in advance via the HMI device 10 that the automatic driving control will be terminated. Then, the automatic driving control is terminated. Alternatively, first, a notification (warning) may be given after the elapse of the upper limit time T3 to inform that the automatic driving control will be terminated if hands-on and eyes-on are not performed. And even with such a warning, the automatic driving control may be terminated if hands-on and eyes-on are not performed. The cancellation of the termination of the automatic driving control is performed, for example, when the system of the vehicle 1 is turned off (i.e., when the ignition is off).

[0041] On the other hand, when the response time T2 is within the upper limit time T3 (step S112; No), the process proceeds to step S116. In step S116, the ECU 50 determines whether the response time T2 calculated in step S110 is longer than the specified time T4 (judgment threshold value). The driver 2 is required by the system of the vehicle 1 to perform hands-on and eyes-on so that the response time T2 is equal to or less than the specified time T4 (for example, 4 seconds).

[0042] When the response time T2 exceeds the specified time T4 (step S116; Yes), the process proceeds to step S118. In step S118, the ECU 50 shortens and updates the notification interval T1. More specifically, the notification interval T1 is shortened, for example, by subtracting a predetermined correction time ΔT1 from the current value. The notification interval T1 updated by step S118 is stored in the storage device 54.

[0043] On the other hand, when the response time T2 is shorter than the specified time T4 (step S116; No), the process proceeds to step S120. In step S120, the ECU 50 extends and updates the notification interval T1. More specifically, the notification interval T1 is extended, for example, by adding a predetermined correction time ΔT2 to the current value. The notification interval T1 updated by step S120 is stored in the storage device 54. The correction time ΔT1 and the correction time ΔT2 may be the same or different. Additionally, when the response time T2 is equal to the specified time T4, the notification interval T1 is maintained at the current value. Also, even when the response time T2 is substantially equal to the specified time T4, the notification interval T1 may be maintained at the current value.

[0044] The processing of steps S118 and S120 is further supplemented. When the process first proceeds to step S118 or S120 after the start of the processing of the flowchart shown in FIG. 2, the notification interval T1 set in step S100 corresponds to the "current value" of the notification interval T1 as referred to here. On the other hand, when the process proceeds to step S118 or S120 for the second time or later after the start of the processing of this flowchart, the value of the notification interval T1 updated by the processing of the most recent step S118 or S120 (that is, the previous value) corresponds to the "current value".

[0045] 3. Effects As described above, according to the present embodiment, the hands-on request and the eyes-on request are notified to the driver 2 each time the notification interval T1 elapses. And the notification interval T1 is calculated based on the driving characteristic information of the driver 2. Thereby, for example, it becomes possible to appropriately continue the driving preparation state using the notification interval T1 based on the driving characteristic information without the need to constantly monitor the driver 2 during autonomous driving using a monitoring device such as the driver monitor 20.

[0046] In addition, the driver 2 is aware in advance that the notification interval T1 is calculated based on its own driving characteristic information. Therefore, the driver 2 can understand that improving driving characteristics such as the execution of safe driving leads to an extension of the notification interval T1, that is, an improvement in the convenience of autonomous driving (for example, an increase in the freedom of secondary activities and an increase in the eyes-off time). Therefore, by changing the notification interval T1 according to the driving characteristic information, it becomes possible to prompt the driver 2 to improve the driving characteristics for the purpose of extending the notification interval T1.

[0047] When the driving preparation state can be appropriately maintained by the method of this embodiment, even when a sudden driving handover request occurs, the driver 2 can quickly return to the driving operation. This contributes to safe vehicle driving by level 3 autonomous driving. In addition, it leads to the ability to provide the driver 2 with various secondary activities (for example, reading, operating a smartphone, eating, or using a PC). Furthermore, since it is not necessary to constantly monitor the driver 2 during autonomous driving, the vehicle cost required for monitoring can be reduced.

[0048] Also, in this embodiment, the driving characteristic information includes information regarding the dangerous driving of the driver 2. When there is a history of dangerous driving, the ECU 50 shortens the notification interval T1 compared to the case where there is no such history (see step S100). As a result, while prompting the driver 2 to improve the driving characteristics so that the driver 2 pays more attention to safer driving, it becomes possible to appropriately maintain the driving preparation state.

[0049] Furthermore, in the present embodiment, when the response time T2 (for example, the average response time T2m) is longer than the specified time T4 (determination threshold value), the notification interval T1 is shortened, and when the response time T2 is shorter than the specified time T4, the notification interval T1 is extended (see steps S100 and S116 - S120). The driver 2 is aware in advance that the notification interval T1 is thus changed according to the specified time T4. Therefore, the driver 2 can understand that the shortening of the response time T2 leads to the extension of the notification interval T1, that is, an improvement in the convenience of autonomous driving. Thus, according to the present embodiment, by increasing or decreasing the notification interval T1 according to the response time T2 of the driver 2 in this way, it is possible to make the driver 2 aware of always returning to the driving operation within the specified time T4. For this reason, even if the driver 2 is not constantly monitored during autonomous driving, the driving preparation state can be maintained well. Additionally, restricting the use of autonomous driving when the response time T2 exceeds the upper limit time T3 (see steps S112 and S114) also leads to making the driver 2 aware of promptly returning to the driving operation in response to the hands - on and eyes - on requirements.

[0050] 4. Other Examples of Notification Processing In the notification processing according to the above - described embodiment, both the hands - on requirement and the eyes - on requirement are notified to the driver 2 to maintain the driving preparation state of the driver 2. Instead of such an example, the notification processing may be executed as follows.

[0051] FIG. 3 is a flowchart showing another example of the processing related to the notification processing according to the embodiment. Hereinafter, the differences between the processing of this flowchart and the processing of the flowchart shown in FIG. 2 will be described.

[0052] In FIG. 3, the processing proceeds to step S200 after step S104. In step S200, the ECU 50 notifies only the eyes - on requirement to the driver 2 using the HMI device 10.

[0053] In step S202 following step S200, the ECU 50 determines whether the eyes-on of driver 2 in response to the eyes-on request is detected. As a result, if eyes-on is detected (step S202; Yes), the ECU 50 calculates, in step S204, the response time T2e for the eyes-on request. The calculated response time T2e is stored in the storage device 54. Thereafter, the process proceeds to step S112. In this case, the response time T2e is used in the determinations of steps S112 and S116.

[0054] On the other hand, if the eyes-on request is not detected (step S202; No), the ECU 50 determines, in step S206, whether a predetermined time T5 has elapsed since the eyes-on request. This predetermined time T5 is set, for example, to be longer than the specified time T4 and shorter than the upper limit time T3. If this predetermined time T5 has not elapsed (step S206; No), the process returns to step S202.

[0055] On the other hand, when the predetermined time T5 elapses without eyes-on being detected (step S206; Yes), the process proceeds to step S208. In step S208, the ECU 50 uses the HMI device 10 to notify driver 2 of the hands-on request. Thereafter, the process proceeds to step S108.

[0056] According to the process shown in FIG. 3 described above, if it is confirmed that driver 2 is conscious due to eyes-on being performed in response to the notification of the eyes-on request, the hands-on request is not made. Thereby, while improving the convenience of driver 2, the driving preparation state can be appropriately maintained.

Description of Reference Numerals

[0057] 1 Autonomous vehicle, 2 Driver, 3 Steering wheel, 10 HMI device, 20 Driver monitor, 30 Sensor group, 40 Travel device, Electronic control unit (ECU), 52 Processor, 54 Storage device

Claims

1. A vehicle notification device applied to an autonomous vehicle, comprising: one or more processors; wherein the one or more processors perform a notification process of notifying a driver of at least the eyes-on request among a hands-on request and an eyes-on request each time a notification interval elapses; calculate the notification interval based on the driving characteristic information of the driver; wherein the driving characteristic information includes information regarding dangerous driving of the driver; when there is a history of the dangerous driving, the one or more processors shorten the notification interval compared to the case where there is no such history; the information regarding the dangerous driving includes the number of times of emergency braking automatically performed by a braking device of the autonomous vehicle, the number of times of rapid acceleration, the number of times of rapid deceleration, the number of times of rapid steering, the number of times of serpentine driving, or the number of times of shoulder driving; a vehicle notification device.

2. In the notification process, the one or more processors do not notify the hands-on request when the eyes-on of the driver is detected after notifying the eyes-on request; notify the hands-on request when the eyes-on of the driver is not detected after notifying the eyes-on request. The vehicle notification device according to claim 1.

3. The one or more processors shorten the notification interval when the time from the notification of at least the eyes-on request among the hands-on request and the eyes-on request to the response of at least the eyes-on among the hands-on and eyes-on of the driver corresponding to the notification is longer than a determination threshold. The vehicle notification device according to claim 1.

4. The one or more processors extend the notification interval when the time from the notification of at least the eyes-on request among the hands-on request and the eyes-on request to the response of at least the eyes-on among the hands-on and eyes-on of the driver corresponding to the notification is shorter than a determination threshold. The vehicle notification device according to claim 1.

Images