Driving support device

Abstract

To provide a driving support device that correctly grasps a cognitive state of a driver regarding danger and enables appropriate driving support for the driver.SOLUTION: A driving support device that provides driving support to a driver of a vehicle, comprises: a driver operation sensor unit 110 that detects a driver operation while driving; a travel environmental sensor unit 120 for detecting a travel environment of the vehicle; a cognitive state estimation unit 131 that estimates a cognitive state of the driver with respect to the travel environment based on a relation between the driver operation and the travel environment; and a processing unit 132 that changes a content of the support information for the driving support according to the estimated cognitive state and outputs the content to the driver from an output unit 140, and / or changes an instruction content to a vehicle control unit 150 for controlling an automatic driving function of the vehicle and outputs it.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a driving assistance device. [Background technology]

[0002] A driving support device is known, for example, as described in Patent Document 1. The driving support device (vehicle warning device) of Patent Document 1 is configured to alert the driver to the possibility of a collision between the vehicle and an object to be warned by an alarm sound. Here, the degree of inattentiveness of the driver is determined by an inattentiveness degree determination means, and when the determined degree of inattentiveness increases, the timing of outputting the alarm sound to the driver is advanced. The degree of inattentiveness is determined, for example, based on the direction of the driver's line of sight detected by a camera.

[0003] Furthermore, if the brake depression speed exceeds a predetermined value after the alarm sound is output, the output timing of the alarm sound is corrected and set earlier.

[0004] This makes it possible to appropriately prevent the driver from suddenly braking due to inattentive driving or from approaching too close to an object that is the subject of a warning. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 4534789 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the above-mentioned Patent Document 1, a camera is simply used to determine the degree to which the driver is looking away from the vehicle from the direction of the driver's line of sight, so there is a possibility that the driver may receive a warning even if he or she recognizes the object to be warned, which may be annoying to the driver.

[0007] An object of the present disclosure is to provide a driving assistance device that correctly grasps a driver's awareness of danger and enables appropriate driving assistance to the driver. [Means for solving the problem]

[0008] In order to achieve the above object, the present disclosure employs the following technical means.

[0009] The present disclosure provides a driving assistance device that provides driving assistance to a driver of a vehicle, A driver action sensor unit (110) for detecting a driver action while driving; A driving environment sensor unit (120) for detecting a driving environment of a vehicle; A cognitive state estimation unit (131) that estimates a cognitive state of the driver with respect to the driving environment based on a relationship between the driver's action and the driving environment; The device is equipped with a processing unit (132) that changes the content of assistance information for driving assistance according to the estimated cognitive state and outputs the changed information to the driver from an output unit (140), and / or changes and outputs instructions to a vehicle control unit (150) that controls the vehicle's automatic driving function.

[0010] According to the present disclosure, the cognitive state estimation unit estimates the cognitive state of the driver with respect to the driving environment (awareness level of dangers and obstacles) from the relationship between the driver's actions and the driving environment, and therefore can estimate an accurate cognitive state. Then, the processing unit changes the content of the support information for driving support according to the estimated cognitive state and outputs it to the driver, and / or changes the content of instructions to the vehicle control unit, so that it is possible to provide appropriate driving support to the driver.

[0011] The reference characters in parentheses for each of the above means indicate the corresponding relationship with the specific means described in the embodiments below. [Brief description of the drawings]

[0012] [Figure 1] 1 is an explanatory diagram showing an overall configuration of a driving support device; [Diagram 2]1 is an explanatory diagram conceptually showing a method for estimating a driver's cognitive state; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, a number of embodiments for carrying out the present disclosure will be described with reference to the drawings. In each embodiment, the same reference numerals may be used to designate parts corresponding to matters described in the preceding embodiment, and duplicated descriptions may be omitted. In each embodiment, when only a part of the configuration is described, other embodiments described previously may be applied to other parts of the configuration. In addition to combinations of parts that are specifically indicated as being possible in each embodiment, it is also possible to partially combine embodiments even if not indicated, as long as there is no particular problem with the combination.

[0014] (First embodiment) A driving support device 100 according to a first embodiment is shown in Fig. 1 and Fig. 2. The driving support device 100 is a device that provides driving support to a driver who drives a vehicle. Driving support may be, for example, a device that uses a display, a voice, a tactile sense (vibration, operating force), or the like output from an output unit 140, or a device that uses a change in the content of automatic driving of the vehicle. As shown in Fig. 1, the driving support device 100 includes a driver operation sensor unit 110, a driving environment sensor unit 120, and a control unit 130 (a cognitive state estimation unit 131 and a processing unit 132). The control unit 130 (processing unit 132) is connected to the output unit 140 and / or a vehicle control unit 150.

[0015] The driver motion sensor unit 110 is a detection unit that detects the driver's motion while driving. Hereinafter, the driver motion sensor unit 110 will be referred to as the "motion sensor unit 110." Examples of the motion sensor unit 110 include a steering wheel sensor that detects the driver's operation state with respect to the steering wheel, a pedal sensor that detects the state of depression of the accelerator pedal and the brake pedal, and an in-car camera that obtains an image of the driver and detects the line of sight, facial orientation, facial expression (satisfied, dissatisfied, nervous), posture, and the like. Examples of the motion sensor unit 110 include an audio vibration sensor that detects the driver's voice and body vibrations (movement), and a biosensor that detects the driver's vital signs such as breathing, body temperature, blood pressure, and pulse.

[0016] In addition, the in-vehicle camera is not limited to a camera dedicated to the driving assistance device 100, but may also be a camera set up in a DMS (driver monitoring system) that detects the driver's inattentiveness, drowsiness, etc., and provides driving assistance.

[0017] Data on the driver's actions detected by the action sensor unit 110 is output to a cognitive state estimation unit 131 of the control unit 130.

[0018] The driving environment sensor unit 120 is a detection unit that detects the driving environment of the vehicle. Hereinafter, the driving environment sensor unit 120 will be referred to as the "environment sensor unit 120." For example, an external camera provided outside the vehicle, a millimeter wave radar, a LiDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging), a sonar, etc. are used as the environment sensor unit 120. The environment sensor unit 120 detects the driving environment, such as the shape of the road on which the vehicle is traveling (straight, curved, uphill, downhill, width, etc.), the distance between the vehicle and other vehicles traveling around (ahead), the presence or absence of pedestrians, and driving regulation conditions such as traffic lights, signs, etc.

[0019] Data on the driving environment detected by the environmental sensor unit 120 is output to a cognitive state estimation unit 131 of the control unit 130.

[0020] When performing driving assistance, the control unit 130 is a part that controls the output unit 140 and / or issues instructions to the vehicle control unit 150. The control unit 130 includes a cognitive state estimation unit 131, a processing unit 132, and the like.

[0021] The cognitive state estimation unit 131 is a part that estimates the cognitive state of the driver with respect to the driving environment from the acquired data of the driver's operation and the data of the driving environment. The cognitive state of the driver with respect to the driving environment here means the driver's recognition level (whether or not the driver recognizes) of dangers, obstacles, etc. while driving.

[0022] The cognitive state estimation unit 131 stores a relationship map obtained by pre-learning the relationship between the driving environment and the driver's actions in a large-scale database, as shown in Fig. 2. The cognitive state estimation unit 131 calculates (estimates) the cognitive state of the driver by sequentially substituting various driving environment data obtained during driving and driver action data into this relationship map.

[0023] In Fig. 2, the lower curve indicates the driving environment (characteristics), and the upper curve indicates the driver's behavior (characteristics). The characteristics of the driving environment include road shape (straight, curved, etc.), road width (wide, narrow), vehicle distance (wide, narrow), pedestrians (presence or absence), traffic lights (presence or absence), etc. The driving environment is related to the load on the driving operation (ease or difficulty of driving).

[0024] In addition, the characteristics of the driver's actions include steering operation (fixed, sudden operation), pedal operation (fixed, sudden operation), gaze direction (looking forward, looking aside, falling asleep), facial expression (relaxed, tense, dissatisfied), posture (presence or absence of unnecessary movements), voice (satisfied, dissatisfied), etc. The driver's actions are related to the driver's driving operation ability (driving response ability).

[0025] In the relationship map, the longer the distance between the two corresponding characteristics (upward arrow in Figure 2) in response to changes in the driving environment and changes in the driver's actions, the lower the risk of driving (the greater the degree of driving composure).On the other hand, the shorter the distance between the two corresponding characteristics, the higher the risk of driving (the smaller the degree of driving composure) (star in Figure 2).

[0026] The cognitive state estimation unit 131 substitutes the acquired driving environment data and driver operation data into the above-mentioned relationship map to calculate (estimate) the driver's cognitive state, and in particular, when the two characteristics are close to each other at the same time as shown by the star in Fig. 2, it estimates the driver's cognitive state (there is awareness of dangers and obstacles).In this way, the cognitive state estimation unit 131 estimates the driver's cognitive state by learning the complex relationships in the relationship map using AI (Artificial Intelligence).

[0027] Specific examples of targets (dangers or obstacles) that affect the driver's cognitive state include pedestrians suddenly running out into the road, delays in driving operations due to looking away or falling asleep, the sudden approach of other vehicles, and branching and merging points on expressways.

[0028] The cognitive state estimation unit 131 outputs data on the calculated (estimated) cognitive state of the driver to the processing unit 132.

[0029] The processing unit 132 is a part that changes the content of the assistance information for driving assistance depending on the estimated cognitive state of the driver and outputs it to the driver from the output unit 140, and / or changes and outputs the instruction content to the vehicle control unit 150 that controls the vehicle's automatic driving function.

[0030] The output unit 140 is configured with, for example, a display (meter display, center display, etc.), a speaker, an actuator for generating a tactile sensation, etc., and is a part that outputs support information for driving support to the driver.

[0031] For example, the display may display warnings, cautions (advice), and the like in conversational text to help avoid danger as support information.

[0032] In addition, the speaker conveys support information such as warnings, cautions (advice), etc. to the driver by voice using a large language model. By using a large language model as support information, it is possible to provide support information that is easy for the driver to understand and does not distract the driver's eyes while driving.

[0033] In addition, the actuator vibrates the seat, steering wheel, pedals, etc., or changes (heavier) the force required to operate the steering wheel, pedals, etc., to convey support information, such as a warning or caution (advice), to the driver.

[0034] The vehicle control unit 150 is a part that controls various driving function units that enable automatic driving of the vehicle. Examples of the automatic driving function include an automatic braking function, a lane keeping function, an automatic overtaking function, and an automatic following driving function.

[0035] In this embodiment, both the output unit 140 and the vehicle control unit 150 are assumed to receive instructions from the processing unit 132. However, depending on the driving assistance required, the case may include only the output unit 140 or only the vehicle control unit 150.

[0036] The driving assistance device 100 of this embodiment is configured as described above, and its operation, functions and effects will be described below.

[0037] While the vehicle is traveling, data on the driver's actions detected by the action sensor unit 110 and data on the traveling environment detected by the environmental sensor unit 120 are output to the cognitive state estimation unit 131.

[0038] The cognitive state estimation unit 131 estimates the driver's cognitive state with respect to the driving environment (hazards and extraneous objects) from the driving environment characteristics of the relationship map and the driver's operation characteristics.

[0039] Then, the processing unit 132 changes the support information for driving support (warning, advice, etc.) so as to correspond to the cognitive state of the driver, and outputs it to the output unit 140. For example, the processing unit 132 displays text such as "There are pedestrians ahead. Please be careful" on the display. Alternatively, the processing unit 132 outputs audio such as "There are pedestrians ahead. Please be careful" from a speaker. Alternatively, the processing unit 132 activates an actuator to vibrate the seat or steering wheel to warn the driver.

[0040] Alternatively, the processing unit 132 changes the content of the instruction to the vehicle control unit 150 in accordance with the driver's cognitive state to change the content of the automatic driving. For example, the processing unit 132 lowers the set vehicle speed to enable safer driving operation.

[0041] As described above, in this embodiment, the cognitive state estimation unit 131 estimates the cognitive state of the driver with respect to the driving environment (awareness level of dangers and obstacles) from the relationship between the driver's actions and the driving environment, and can estimate (grasp) the cognitive state accurately. Then, the processing unit 132 changes the content of the support information for driving support according to the estimated cognitive state and outputs it from the output unit 140 to the driver, and / or changes the instruction content to the vehicle control unit 150, making it possible to provide appropriate driving support to the driver.

[0042] In other words, if the driver is unaware of dangers or obstacles, assistance information is output (or the content of the automated driving is changed), but if the driver is aware of them, assistance information is not output (or the content of the automated driving is not changed), so the driver is not inconvenienced.

[0043] This type of driving assistance not only helps avoid risks while driving, but also leads to changes in the driver's daily driving behavior through repeated control.

[0044] Furthermore, this embodiment can resolve dissatisfaction with driving assistance information, such as the fact that it is a simple, standard message, making it difficult to understand why it has been pointed out, the fact that it is pointed out suddenly without taking into account the time, place, and occasion, and the fact that it is pointed out even when the driver is paying proper attention.

[0045] Other embodiments including specific details will be described below.

[0046] Second embodiment In the second embodiment, the cognitive state estimation unit 131 estimates the cognitive state based on the driver's reaction to a predetermined event in the driving environment among the driver's actions. The predetermined event is a dangerous scene predicted in advance, such as a pedestrian is predicted to be present at a crosswalk without traffic lights, or the distance between the vehicle and the vehicle ahead is equal to or smaller than a predetermined distance.

[0047] In this way, by utilizing the driver's reaction to a specific event, it becomes possible to more easily estimate the driver's cognitive state.

[0048] Third embodiment In the third embodiment, the cognitive state estimation unit 131 determines that the driver has responded if the driver's response occurs a predetermined time before the occurrence of a predicted predetermined event.

[0049] Normally, a driver tends to predict that a certain event may occur a certain time before the event (for example, about 4 seconds before the event occurs). Therefore, by determining that the driver has responded when the driver has responded a certain time before the predicted occurrence of the certain event, the estimation accuracy of the cognitive state can be improved.

[0050] (Fourth embodiment) In the fourth embodiment, when a predetermined event does not occur, the processing unit 132 outputs, as support information, advice information for the next time to the output unit 140. For example, when no pedestrian jumps out at the crosswalk, the processing unit 132 outputs support information such as "No pedestrians jumped out this time, but please be careful next time" at the output unit 140.

[0051] This makes it possible to make the driver aware of the possibility of such a situation and to pay attention to it at all times.

[0052] Fifth embodiment In the fifth embodiment, the processing unit 132 causes the output unit 140 to vibrate the seat to allow the driver to simulate the occurrence of a predetermined event, and displays a mark indicating the predetermined event on the road. For example, the output unit 140 vibrates the seat using an actuator to allow the driver to simulate the experience of vehicle vibration caused by a bump (hump) on the road, and further displays the bump on the road using a head-up display device (or a floodlight) or the like.

[0053] This makes it possible to make the driver aware of the possibility of a collision, as in the fourth embodiment.

[0054] Sixth embodiment In the sixth embodiment, when the ignition switch of the vehicle is turned on, the processing unit 132 displays a character image representing a driving instructor on the output unit 140 (display). The character image conveys support information (warnings and advice) to the driver as a driving instructor.

[0055] This allows the driver to understand that the driving assistance system is provided with a driving instructor, and allows the driver to receive effective driving assistance.

[0056] Seventh embodiment In the seventh embodiment, a notification unit for vehicle exterior notification is provided outside the vehicle (outside the vehicle). The notification unit is, for example, an exterior speaker or an exterior indicator lamp (color-coded indicator lamp), etc. Then, the processing unit 132 notifies the outside of the vehicle of information related to the evaluation result of the recognition state by the notification unit. The information related to the evaluation result of the recognition state is, for example, information when the driver predicts that there may be a pedestrian on the crosswalk. The processing unit 132 notifies the pedestrian outside the vehicle of information such as "The driver has noticed you and is about to stop" by the exterior speaker or the exterior indicator lamp, etc., or turns on a green lamp.

[0057] The outside-of-vehicle notification is performed when a pedestrian is present on the crosswalk. In other words, when a pedestrian is present and the processing unit 132 determines that the vehicle is capable of driving while predicting the pedestrian in advance, the processing unit 132 issues an outside-of-vehicle notification.

[0058] This provides a sense of security to pedestrians outside the vehicle.

[0059] Eighth embodiment In the eighth embodiment, the processing unit 132 determines whether or not the driver has made advance preparations for switching from automatic driving to manual driving associated with the automatic driving function based on the driver's cognitive state. Advance preparations for switching to manual driving include, for example, looking ahead, gripping the steering wheel, and placing the foot on the accelerator pedal.

[0060] Furthermore, when the driver is prepared in advance, the processing unit 132 reduces notifications related to the driver's handover and increases information for advice on safe driving in the output unit 140. In other words, when the driver smoothly takes over the driver's handover action, the processing unit 132 reduces notifications urging the driver to take over the driver's handover and increases better advice for the driver.

[0061] In addition, the processing unit 132 grasps the driver's level of tension from the driver's cognitive state, and when advance preparations for a driver handover have been made and the level of tension is higher than a predetermined value, performs processing to output a dialogue using artificial intelligence as support information.

[0062] This enables driving support for the driver when switching from automatic driving to manual driving. Also, when the driver is prepared in advance, the output unit 140 reduces notifications related to the driver switching and increases information for advice on safe driving, thereby enabling useful driving support without imposing a psychological burden on the driver. Also, when the driver's level of tension is higher than a predetermined value at the time of the driver switching, output processing is performed through dialogue to ease the driver's mind.

[0063] Ninth embodiment In the ninth embodiment, the processing unit 132 grasps the degree of tension of the driver from the driver's cognitive state, and when the degree of tension is higher than a predetermined value, changes the transfer content of the vehicle control related to the automatic driving function. For example, even in the case of switching from automatic driving to manual driving, the change of the transfer content means stopping the switch to manual driving and continuing automatic driving. This makes it possible to reduce the burden on the driver associated with tension.

[0064] Tenth embodiment In the tenth embodiment, the processing unit 132 reduces the output of the support information when the cognitive state estimation unit 131 estimates that the driver is dissatisfied from the detected driver's operation by using feedback on the output of support information to the driver. The cognitive state estimation unit 131 grasps the dissatisfied state from, for example, the words and facial expressions of the driver.

[0065] As a result, if the driver feels dissatisfied, the output of the assistance information is reduced (suppressed), so that the driver is not made further dissatisfied and is given time to calm down.

[0066] Eleventh embodiment In the eleventh embodiment, the processing unit 132 judges whether the estimation accuracy of the cognitive state of the driver is good or bad by using feedback on the output of assistance information to the driver. The processing unit 132 judges whether the estimation accuracy of the cognitive state is good or bad by, for example, the dissatisfied state of the driver described in the above tenth embodiment. When the driver is dissatisfied, the processing unit 132 judges that the estimation accuracy of the cognitive state is relatively low.

[0067] If the estimation accuracy is lower than a predetermined value, the processing unit 132 changes the control setting value related to the automatic driving function of the vehicle to the safe side. Examples of changes to the safe side include relatively lowering the speed setting value in automatic driving, relatively advancing the timing of automatic braking, or relatively increasing the inter-vehicle distance in automatic following driving.

[0068] This makes it possible to drive more safely by changing the conditions for autonomous driving, even if the estimation accuracy of the driver's cognitive state is relatively low.

[0069] Twelfth embodiment In the twelfth embodiment, when the output unit 140 outputs a voice language as assistance information, the processing unit 132 adjusts the output time of the assistance information so that the assistance information can be completed before the autonomous driving function is started. For example, if the time until the autonomous driving function is started is relatively short, the processing unit 132 sets the output time from the output unit 140 to be relatively short, and provides concise assistance information.

[0070] After the autonomous driving function is implemented, there is substantially no need for assistance in spoken language to the driver. Therefore, by ending assistance information in spoken language before the autonomous driving function is started, reliable driving assistance can be provided during manual driving.

[0071] Thirteenth embodiment In the thirteenth embodiment, for example, a speech notification using voice and a sensory notification imparting a tactile sense are prepared as assistance information for the driver.

[0072] Then, the processing unit 132 changes the output content of the support information in the output unit 140 according to the implementation level of the preparatory movement for avoiding danger by the driver as the driver's movement. The preparatory movement is, for example, a movement of a pre-reaction to a danger, such as a driver quickly releasing the accelerator pedal slightly or lightly depressing the brake pedal when the driver recognizes a danger.

[0073] Furthermore, the lower the implementation level of the preparatory movement, the higher the priority the processing unit 132 gives in outputting a sensory notification over a vocal linguistic notification.

[0074] Furthermore, when the execution level of the preparatory action for a driver who is inexperienced in driving is lower than a predetermined level, the processing unit 132 makes the time of the voice language notification using voice as the support information longer than usual. Examples of "drivers who are inexperienced in driving" include beginners who have obtained a driver's license within one year, paper drivers who drive infrequently, and elderly drivers. Note that the processing unit 132 can identify inexperienced drivers by having the driver themselves pre-register (input) the driver's information in the control unit 130 and authenticate them while driving.

[0075] This allows the driver to receive support in line with his / her driving skills by changing the output content of the support information according to the driver's preparatory movement. In addition, the lower the level of the preparatory movement, the more effective the driving support can be provided by giving priority to the output of the sensory notification. Furthermore, for a driver who is inexperienced in driving, when the level of the preparatory movement is lower than a predetermined level, the time of the voice notification is extended longer than usual, allowing for polite verbal support.

[0076] (Other embodiments) The disclosure in this specification and drawings, etc. is not limited to the exemplified embodiments. The disclosure includes the exemplified embodiments and modifications by those skilled in the art based thereon. For example, the disclosure is not limited to the combination of parts and / or elements shown in the embodiments. The disclosure can be implemented by various combinations. The disclosure can have additional parts that can be added to the embodiments. The disclosure includes the omission of parts and / or elements of the embodiments. The disclosure includes the substitution or combination of parts and / or elements between one embodiment and another embodiment. The disclosed technical scope is not limited to the description of the embodiments. Some disclosed technical scopes are indicated by the description of the claims, and should be interpreted as including all modifications within the meaning and scope equivalent to the description of the claims. [Explanation of symbols]

[0077] 100 Driving support device 110 Driver motion sensor unit (motion sensor unit) 120 Driving environment sensor unit (environmental sensor unit) 131 Cognitive State Estimation Unit 132 Processing section 140 Output section 150 Vehicle control unit

Claims

1. A driving assistance device that provides driving assistance to a driver of a vehicle, A driver action sensor unit (110) for detecting a driver action while driving; A driving environment sensor unit (120) that detects a driving environment of the vehicle; A cognitive state estimation unit (131) that estimates a cognitive state of the driver with respect to the driving environment based on a relationship between the driver's action and the driving environment; A driving assistance device comprising: a processing unit (132) that changes the content of assistance information for driving assistance depending on the estimated cognitive state and outputs the information to the driver from an output unit (140), and / or changes and outputs instruction content to a vehicle control unit (150) that controls the automatic driving function of the vehicle.

2. The driving support device according to claim 1 , wherein the cognitive state estimation unit estimates the cognitive state based on a reaction of the driver to a predetermined event in the driving environment, among the driver's actions.

3. The driving support device according to claim 2 , wherein the cognitive state estimation unit determines that the driver has responded when the driver's response occurs a predetermined time before the occurrence of the predicted predetermined event.

4. The driving support device according to claim 2 , wherein, when the predetermined event has not occurred, the processing unit causes the output unit to output, as the support information, information for advice to be provided for a next time.

5. The driving support device according to claim 2 , wherein the processing unit causes the output unit to vibrate the seat so as to allow the driver to simulate an occurrence of the predetermined event, and displays a mark on the road indicating the predetermined event.

6. The driving assistance device according to claim 1 , wherein the processing unit uses a large-scale language model when outputting the assistance information by voice from the output unit.

7. 2. The driving assistance device according to claim 1, wherein the processing unit displays a character image representing a driving instructor on the output unit when an ignition switch of the vehicle is turned on.

8. The driving support device according to claim 1 , wherein the processing unit notifies information related to the assessment result of the cognitive state to an outside of the vehicle by an annunciation unit provided outside the vehicle.

9. The driving support device according to claim 8 , wherein the processing unit issues the outside-vehicle notification when a pedestrian is present on a pedestrian crossing.

10. The driving support device according to claim 9 , wherein the processing unit issues the outside-vehicle alert when it is determined that the driving is performed while predicting the pedestrian in advance in the case where the pedestrian is present.

11. The driving assistance device according to claim 1 , wherein the processing unit determines whether or not advance preparations for switching from automatic driving to manual driving associated with the automatic driving function have been completed based on the cognitive state.

12. The driving support device according to claim 11 , wherein, when the driver is prepared in advance, the processing unit reduces notifications related to the driving changeover and increases information for advice on safe driving for the driver at the output unit.

13. The driving support device according to claim 11, wherein the processing unit grasps the degree of tension of the driver from the cognitive state, and when the advance preparation for the driving change is completed and the degree of tension is higher than a predetermined value, performs the output process of a dialogue using artificial intelligence as the support information.

14. The driving assistance device according to claim 1 , wherein the processing unit grasps a degree of tension of the driver from the cognitive state, and when the degree of tension is higher than a predetermined value, changes the content of the transfer of vehicle control related to the automatic driving function.

15. 2. The driving assistance device according to claim 1, wherein the processing unit reduces the output of the assistance information when the cognitive state estimation unit estimates that the driver is dissatisfied from the detected driver's behavior using feedback on the output of the assistance information to the driver.

16. The driving assistance device according to claim 1 , wherein the processing unit determines whether the estimation accuracy of the cognitive state is good or bad by using feedback on an output of the assistance information to the driver.

17. The driving support device according to claim 16 , wherein the processing unit changes a setting value of control related to the autonomous driving function of the vehicle to a safer side when the estimation accuracy is lower than a predetermined value.

18. The driving assistance device according to claim 1 , wherein when the processing unit outputs speech language as the assistance information, the processing unit adjusts an output time of the assistance information so that the output is completed before the autonomous driving function is started.

19. The driving support device according to claim 1 , wherein the processing unit changes an output content of the support information from the output unit in accordance with a level of execution of a preparatory action for avoiding a danger by the driver as the driver's action.

20. As the support information, a speech notification using voice and a sensory notification imparting a tactile sense are prepared, The driving assistance device according to claim 19 , wherein the processing unit is configured to give a higher priority to outputting the sensory notification as the execution level of the preparatory movement is lower.

21. 20. The driving assistance device according to claim 19, wherein the processing unit, when the driver is inexperienced at driving and the level of the preparatory movement is lower than a predetermined level, sets a time for providing a voice-based audio notification as the assistance information to be longer than usual.

Images