JP2016207001A - Driving support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driving support device which can ensure traveling safety and prevent hindrance to interaction between occupants when there are a plurality of occupants.SOLUTION: A driving support device 1 comprises: a deviation determination part 22 for determining reduction in safety of a vehicle on the basis of a traveling state of the vehicle; and an alarming device 4 for alarming a driver when a determination value of the deviation determination part 22 has exceeded a determination threshold value. It comprises: a voice interaction device 10 which acquires voice of interaction between the driver and an occupant: a determination threshold value setting part 23 which makes a determination threshold value when the driver is speaking to an occupant greater than a determination threshold value when the driver is listening to an occupant; and an enthusiasm level determination part 24 for determining an enthusiasm level of the driver in the interaction on the basis of words that are included in the voice of the driver and the occupant who are engaged in the interaction. It reduces the determination threshold value when it is determined that the enthusiasm level of the driver in the interaction is high by the enthusiasm level determination part 24.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a driving support device that issues a warning to a driver when a determination value of a safety deterioration determination unit exceeds a determination threshold.

2. Description of the Related Art Conventionally, there is known a driving support device that warns a driver of a decrease in vehicle safety based on detection information of a running state (behavior) of a vehicle. For example, when the vehicle travels, the left and right lanes are captured by a camera to recognize the position of the host vehicle, and when it is determined that there is a possibility of lane departure, a lane departure warning is issued to alert the driver. Such a lane departure warning is activated even when the driver consciously travels in the end area of the center line or roadside line when traveling in a curve section, and the lane departure warning is activated, causing the occupant to feel uncomfortable or annoying. I may feel it.
In view of this, a technique has been proposed that restricts the operation of unnecessary alarms based on whether or not the driver is in a random state based on the reflection time by the driver.

The driving support device of Patent Document 1 estimates the driving state of the vehicle or the driver's state, and when this estimated value exceeds a determination threshold value regarding whether or not the driving requires attention, the driving support device is notified. A driving support device that informs the driver, and the driver's state is a normal state or a casual state based on a reaction time from when information is acquired through the information acquisition means until a predetermined stop preparation operation is performed The determination threshold is changed to a value smaller than the normal value when it is determined that the state is indiscriminate.
As a result, when the driver is in an ambiguous state, the driver can be notified that the driving is in a state requiring attention earlier than usual, and in the normal state where the driver is concentrated, the notification timing is intentionally delayed. This makes it difficult to be notified of intentional actions.

JP 2008-204056 A

The driving support device of Patent Literature 1 can achieve safety and troublesomeness when the driver drives the vehicle alone.
However, in addition to the situation where the vehicle is driven alone, the driver frequently has a situation where one or more passengers are in addition to the driver.
When a plurality of occupants are in the vehicle and the driver is interacting with the passenger, the driver's awareness is naturally activated. In addition, even when the driver is not interacting with the passenger, since the driver cares about the passenger's condition, the driver's consciousness is expected to be considerably activated.
In other words, in a driving situation where a passenger is present, it can be said that the possibility that the driver is in a state of ambiguity is extremely low.
In other words, as disclosed in Patent Document 1, when it is simply determined whether or not the state is indiscriminate, it is possible to solve the problem of ensuring safety and eliminating bothersomeness on a single ride, but it is troublesome to ensure safety sufficiently on boarding a plurality of passengers. There is a possibility that it is impossible to achieve both resolution.

Therefore, in order to clarify the mutual relationship between the interaction between the passengers and the operation characteristics of the driver, a verification experiment by simulation was performed.
First, the contents of the verification experiment will be described.
Compare the behavior of the vehicle between normal driving in which the driver does not interact with the voice interaction device and interactive driving in which the driver interacts with the voice interaction device by giving the subject a driving task that runs on the circuit in the drive simulator. did.
As experimental conditions, six subjects (men and women in their 30s and 50s), running tasks were set to maintain a target speed (160 km / h) on an oval course of about 2,100 m. The driving course has five sections, a straight line before dialogue, a first dialogue curve, a dialogue line, a second dialogue curve, and a straight line after dialogue, and three zones, a middle first dialogue curve, dialogue straight line, and second dialogue curve. Was set as the dialogue interval. Each subject traveled 9 laps in the traveling course, and average data from all subjects was acquired.

Next, the verification result will be described.
FIG. 7 shows the first verification result.
As indicated by the average speed (solid line) during normal travel, the vehicle decelerates from before the first and second dialogue curves and accelerates from the second half of the first and second dialogue curves.
As shown in FIG. 7, the average speed (dotted line) at the time of interactive driving has a larger deviation rate and speed variation from the target speed than the average speed at the time of normal driving. It was found that it has an influence on the speed characteristics.
FIG. 8 shows the second verification result.
As indicated by the average steering acceleration during normal travel (solid line), a large steering angle operation is performed when the first and second dialogue curves enter and exit.
As shown in FIG. 8, the average steering acceleration (dotted line) at the time of interactive travel has more variation in the steering acceleration due to the corrected steering than the average steering acceleration at the time of normal travel. It has been found that it has an influence on the characteristics).

FIG. 9 shows a third verification result.
As for the deceleration start timing when entering a curve, the characteristics are shown for normal driving (solid line), driver utterance (dashed line), and driver listening (dashed line).
As shown in FIG. 9, during normal driving in which the driver does not interact with the voice interaction device, the driver starts decelerating earliest before the curve, and then during the utterance that the driver himself is speaking. The deceleration start timing is early. Then, there is a tendency that the deceleration start timing at the time when the driver is listening is more delayed than the deceleration start timing at the time of speaking.
As a result, utterance has less influence on cognitive (visual / auditory) and operation characteristics, whereas listening is presumed to have more influence on cognitive (visual) and operational characteristics than utterance.

From the above verification results, it was possible to obtain the knowledge that the operation characteristics during the interactive travel are more varied than the operation characteristics during the normal travel although the driver's consciousness is activated.
Therefore, in order to limit unnecessary warnings when the driver's consciousness is activated, the normal driving level in which the driver does not interact with the passenger at the alarm level when the driver interacts with the passenger. It is possible to make it lower than the alarm level at the time.
However, if the driver's level of warning during dialogue driving where the driver interacts with the passenger is lowered uniformly (threshold is increased), the driver's cognitive and operating characteristics in the listening state are reduced. Even when alerting is required, the alarm is delayed, which may cause a problem in driving safety.
On the other hand, when the warning level during dialogue driving where the driver talks with the passenger is limited (threshold is reduced), the driver's consciousness in the utterance state is activated, but the operation within the expected range Even if the characteristics vary, the warning is advanced, which may hinder the interaction between the driver and the passenger.

  An object of the present invention is to provide a driving support device or the like that can ensure both driving safety when a plurality of occupants get on and prevent dialogue inhibition between the occupants.

  According to the first aspect of the present invention, there is provided a safety decrease determination means for determining a decrease in vehicle safety based on a running state of the vehicle, and a driver when a determination value of the safety decrease determination means exceeds a determination threshold value. In a driving support device having a warning means for performing a warning, a voice acquisition means for acquiring a voice of dialogue between the driver and the passenger, and a warning level by the warning means when the driver is speaking to the passenger Warning level changing means for lowering the warning level by the warning means when the driver is listening to the passenger's utterance.

  Since this driving support device includes voice acquisition means for acquiring the voice of the dialogue between the driver and the passenger, whether or not the passenger is present and whether the driver in the conversation is in a speech state or listening state Can be clearly distinguished. There is an alarm level changing means for lowering the alarm level by the alarm means when the driver is speaking to the passenger to be lower than the alarm level by the alarm means when the driver is listening to the passenger's speech. When the driver is in an utterance state where the attention level (cognitiveness) is higher than the listening state, the warning level can be lowered to prioritize the dialogue between the passengers, and the driver has a higher attention level than the utterance state. In a low listening state, priority can be given to ensuring driving safety by raising the alarm level.

According to a second aspect of the present invention, in the first aspect of the invention, the warning level changing means is configured such that the driver listens to the passenger's utterance with respect to the determination threshold value when the driver is speaking to the passenger. It is characterized by being larger than the determination threshold value.
According to this configuration, it is possible to achieve both driving safety and prevention of interference between passengers with a simple configuration of adjusting the determination threshold value of the safety decrease determination unit.

According to a third aspect of the present invention, in the second aspect of the invention, the alarm level changing means determines when the driver is not interacting with the fellow passenger as to a judgment threshold value when the driver is speaking to the fellow passenger. It is characterized by being larger than the threshold value.
According to this configuration, it is possible to further reduce the troublesomeness caused by the alarm.

According to a fourth aspect of the present invention, in the second or third aspect of the invention, there is provided a degree of heat determination means for determining a degree of heat for the driver's dialogue based on words included in the voices of the driver and the passenger during the dialogue. The determination threshold value is reduced when it is determined by the heat degree determination means that the heat degree with respect to the driver's dialogue is high.
According to this configuration, it is possible to determine a decrease in the driver's attention level based on the degree of heat, and it is possible to ensure the driving safety of the driver who is enthusiastic about the conversation.

  According to the driving support device of the present invention, it is possible to achieve both driving safety when a plurality of passengers get on and prevention of dialogue inhibition between passengers.

1 is a functional block diagram of a driving support apparatus according to Embodiment 1. FIG. It is a functional block diagram of a voice interactive apparatus. It is a figure which shows an example of an interest degree table. It is a figure which shows an example of a physical condition table. It is a flowchart of a driving assistance process. It is a flowchart of a determination threshold value setting process. It is a graph which shows the relationship between a dialog and a speed characteristic. It is a graph which shows the relationship between a dialog and a steering characteristic. It is a graph which shows each operation characteristic at the time of normal time without speech, at the time of speech, and listening.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
The following description exemplifies a case where the present invention is applied to a vehicle, and does not limit the present invention, its application, or its use.

Embodiment 1 of the present invention will be described below with reference to FIGS.
The driving support device 1 determines a decrease in the safety of the vehicle based on the traveling state of the vehicle, and when this determination value exceeds the determination threshold T, for example, a white line ( When the degree of approach between the lane marking) and the host vehicle approaches a predetermined determination threshold T, the driver is warned.

As shown in FIG. 1, the driving support device 1 includes a CCD (Charge Coupled Device) camera 2 as a running state detection unit, a driver's biological information acquisition unit 3, a voice interaction device 10 (voice acquisition unit), An ECU (Electronic Control Unit) 20 and an alarm device 4 are provided.
The CCD camera 2 is disposed on a roof front end (not shown) or left and right door mirrors (not shown) in the vehicle interior. The CCD camera 2 images a white line on a traveling lane in which the host vehicle is traveling. A CMOS (Complementary Metal Oxide Semiconductor) camera may be used in place of the CCD camera 2, and any means can be used as long as it can convert the light incident by the photoelectric conversion element into a digital image of luminance gradation according to the luminance. is there.

The biometric information acquisition unit 3 is configured to be able to detect the driver's emotion and fatigue based on the driver's biometric information. The biometric information acquisition unit 3 is detected by, for example, the degree of sweating of the driver's hand detected by a sweat meter (not shown) provided on the steering wheel, and a heart rate sensor (not shown) built in the driver's seat. The driver's heart rate, the driver's pupil diameter, line-of-sight direction, and the like captured by an indoor camera (not shown) disposed in the passenger compartment are acquired.
Since the degree of sweating and heart rate correlate with the driver's activity, and the pupil diameter and line-of-sight direction correlate with the driver's pleasure, it is possible to estimate the mental state such as the driver's stress. .
The correlation between these detection results and the driver's physical condition (good, normal, bad) is prepared in advance by experiments or the like.

Next, the voice interaction device 10 will be described.
The voice interaction device 10 is configured to acquire voices generated by the driver and one or more passengers (hereinafter collectively referred to as passengers) and to control each passenger in response to voice input from the passengers. Has been. In one example, the voice interactive apparatus 10 is a portable terminal device such as a smartphone, and in another example, the voice interactive apparatus 10 is an in-vehicle PC having a communication function mounted on a vehicle (may include a navigation function or the like). . Further, when the portable terminal device and the in-vehicle PC perform cooperative control, the portable terminal device and the in-vehicle PC correspond to the voice interaction device 10.

As shown in FIG. 2, the voice interaction device 10 is electrically connected to a microphone 5 and a speaker 6 and configured to be able to transmit and receive various information to and from the server 7.
The microphone 5 is used to input the voice of each occupant and is installed at a predetermined position in the passenger compartment. The microphone 5 can acquire the passenger's voice and the passenger's voice from the start of voice input from any of the passengers.
The speaker 6 outputs the sound production data converted by the speech synthesizer 13.

The server 7 is disposed in a separately installed information center (not shown).
The server 7 is formed by a large computer having a large amount of data processing capability and a large amount of data capacity, and transmits and receives various kinds of information to and from the voice interaction device 10.
Here, the information center will be briefly described.
The information center is configured to be able to generate a text for responding to an operation task in which the extracted concept cannot be processed in the vehicle, a text for interrupting a conversation between passengers, and the like. The information center includes a server 7, a communication unit (not shown), and the like, and is configured to be communicable with the voice interaction apparatus 10.
The server 7 includes a dialogue management unit, an information database, a vocabulary grammar database, and the like (all not shown). The dialogue management unit is configured to search the information database and the Internet, and to generate a response text incorporating the search result based on the vocabulary grammar database. The generated response text or the like is converted into a predetermined format and transmitted from the communication unit to the voice interaction apparatus 10.

As shown in FIG. 2, the voice interaction device 10 includes a voice recognition unit 11, a dialogue processing unit 12, a voice synthesis unit 13, and the like.
The voice recognition unit 11 takes in the driver's voice input to the microphone 4 as a digital signal by an A / D converter, converts this digital signal into waveform data, and then performs frequency analysis to extract a feature vector. The speech recognition unit 11 calculates the degree of coincidence between the language model stored in the storage unit 15 and the occupant's speech, recognizes the input speech, and then sends it to the dialogue processing unit 12 as text expressed in a word string. Output.

Next, the dialogue processing unit 12 will be described.
The dialogue processing unit 12 is configured to be able to generate a response text relating to a completion report for an operation task from the occupant (driver) to the voice dialogue device 10 and responses to various inquiries.
Further, the dialog processing unit 12 is configured to display the presence / absence of a dialog between the driver and the passenger, the subject of the current dialog (driver or passenger), and driving for the word included in the current dialog. The user's interests f1 and f2 can be output to the ECU 20.

As shown in FIG. 2, the dialogue processing unit 12 includes a control unit 14, a storage unit 15, and the like.
The control unit 14 cooperatively controls each function unit including on-vehicle devices (not shown) so that a dialogue is established using a dialogue scenario describing a response rule between the occupant and the voice dialogue device 10.
The control unit 14 extracts a concept (intention to speak) from the text recognized by the speech recognition unit 11, outputs the concept in a format that can be handled by the voice interaction device 10, and outputs the intention understanding function. A dialog control function for determining a concept to which the voice interaction device 10 responds based on the result, a response generation function for generating a response text based on the concept output from the dialog control function, and a timing for uttering the response text are set. Utterance timing setting function, and is configured to be able to manage mainly the content of dialogue with the driver and the overall dialogue timing.

In the case where the extracted concept is an operation task that can be processed in the vehicle like an operation of the in-vehicle device by the occupant (driver), the control unit 14 does not communicate with the server 7 for the predetermined in-vehicle device. To output an operation command and generate a response text in the vehicle.
When the concept extracted from the utterance is a task that cannot be processed in the vehicle or a dialogue between passengers, the control unit 14 formats the URL based on the text based on the voice of the passenger expressed as a word string as an information search command. To be sent to the information center.

The storage unit 15 stores a language model database, an intention understanding model database, a problem solving knowledge database, a dialogue scenario database, an interest determination word database (all not shown), an interest degree table 31, and the like.
The language model database stores a word model whose degree of coincidence is compared with the input speech. The intention understanding model database stores the correlation between each word, the expression pattern, and the extracted concept.
The problem solving knowledge database is used to create a concept when a solution to the problem is proposed when the extracted concept is a statement of the problem. The dialog scenario database stores response text creation rules.

The interest determination word database classifies and stores each target included in the dialogue between the passengers into three groups of high interest, middle interest, and low interest based on the interest (preference) of the driver.
As shown in FIG. 3, the interest level table 31 indicates that the driver includes a dialogue subject (driver) so that the words included in the dialogue between passengers correspond to the groups classified using the interest determination word database. The interest level f1 of the driver at the time of utterance) and the interest level f2 of the driver when the passenger is the conversation subject (driver listening) are preset.

Next, the speech synthesizer 13 will be described.
The speech synthesis unit 13 normalizes the response text determined by the dialogue processing unit 12 or the response text received from the information center into an expression suitable for speech output, and each word of the normalized text and the storage unit 15. It is converted into pronunciation data using the language model stored inside. The voice synthesizer 13 converts the sound generation data into waveform data based on a digital signal by a D / A converter and outputs the waveform data to the speaker 6.

Next, the ECU 20 will be described.
The ECU 20 includes a CPU that executes programs, a ROM that stores programs, a RAM that temporarily stores data and programs, an EEPROM (electronically erasable and programmable read only memory), an input / output interface that inputs and outputs data, and the like. A communication unit that communicates with the ECU is configured as a microcomputer connected via a bus. As for this ECU20, the warning which calls a passenger | crew (driver | operator) attention is implement | achieved because CPU runs a program.

As shown in FIG. 1, the ECU 20 includes a travel lane recognition unit 21, a departure determination unit 22 (safety reduction determination unit), a determination threshold setting unit 23 (alarm level change unit), and a heat degree determination unit 24 (hot). Degree determination means) and the like.
The traveling lane recognition unit 21 differentiates the luminance of the front image captured by the CCD camera 2 in the horizontal direction to generate edges that become high frequency components at both ends of the white line, thereby utilizing the white line portion of the traveling lane. The white line is extracted from the estimated white line portion based on the threshold value determined from the brightness and the contrast with the road surface, the threshold value of the white line width, and the like.
The traveling lane recognition unit 21 calculates the extracted white line information and the separation distance between the white line and the host vehicle.

Based on the calculation result by the traveling lane recognition unit 21, the departure determination unit 22 may deviate from the traveling lane when the degree of approach between the white line and the own vehicle approaches the determination threshold T. The operation command can be output to the alarm device 4 by determining.
When the determination threshold value T is large, the departure determination unit 22 is allowed to approach the white line by the own vehicle, and thus is not determined to have a possibility of departure until the separation distance between the white line and the own vehicle becomes small. In addition, when the determination threshold T is small, the approach to the white line by the own vehicle is limited, and therefore it is determined that there is a possibility of departure even if the separation distance between the white line and the own vehicle is a certain large value.
Note that the determination threshold T when the driver is not interacting with the passenger is set in advance as a reference determination threshold T1 that is normally used.

The determination threshold value setting unit 23 is configured to be able to change / set the determination threshold value T of the departure determination unit 22 based on the driving situation of the driver, so-called physical condition, presence / absence of dialogue, degree of interest in dialogue, and the like.
The determination threshold value setting unit 23 sets a determination threshold value T when the driver is speaking to the passenger when the driver is in a predetermined state (the driver is the subject of the dialogue). It is larger than the determination threshold value T when listening (the passenger is the subject of the dialogue).
This is because the driver's cognition (attention level) when the driver is speaking to the passenger is higher than the driver's cognition when the driver is listening to the passenger's speech.

In addition, the determination threshold setting unit 23 is configured to determine the determination threshold T when the driver is speaking to the passenger except when the driver's physical condition is in a poor state when the driver is not interacting with the passenger. It is configured to be larger than the determination threshold T (reference determination threshold T1).
This is because the consciousness when the driver speaks to the passenger is basically activated.
When the determination threshold T when the driver is speaking to the passenger is the first determination threshold T2, and the determination threshold T when the driver is listening to the passenger's utterance is the second determination threshold T3, The first and second determination threshold values T2 and T3 can be expressed by the following equations, respectively.
T2 = T1 + k1 × α
T3 = T1 + k2 × β (1)
Here, k1 and k2 are a heat intensity coefficient when the driver is speaking to the passenger and a heat intensity coefficient when the driver is listening to the passenger's speech, and α and β are 0, respectively. <Β <α is satisfied.

The heat degree determination unit 24 is configured to determine the driver's physical condition g1, g2 based on the detection result of the biometric information acquisition unit 3, and to determine the heat degree coefficients k1, k2 corresponding to the heat degree with respect to the driver's dialogue. Has been.
The heat degree determination unit 24 is provided with a physical condition table 32.
As shown in FIG. 4, the physical condition table 32 indicates that the physical condition g1 of the driver when the driver is a subject of dialogue (driver's utterance) so as to correspond to the physical condition (good, normal, bad) of the driver. And the physical condition g2 of the driver when the passenger is the main subject of dialogue (listening to the driver).

The degree of enthusiasm for the driver's dialogue is that the driver speaks to the passenger using the interests f1, f2 for the driver's dialogue based on the words included in the dialogue and the driver's physical condition g1, g2. The heat intensity coefficient k1 when the driver is listening and the heat intensity coefficient k2 when the driver is listening to the passenger's speech can be expressed respectively.
This is because even if the conversation has the same degree of interest for the driver, when the physical condition is low, the degree of heat is lower than when the physical condition is high.

The thermal intermediate coefficients k1 and k2 can be expressed by the following equations, respectively.
k1 = f1 + g1
k2 = f2 + g2 (2)
The thermal intensity coefficients k1 and k2 obtained by the above equation (2) are output to the determination threshold value setting unit 23 and then substituted into the equation (1), respectively, to obtain the first and second determination threshold values T2 and T3. .

The warning device 4 alerts the driver based on the operation command from the departure determination unit 22 when it is determined that the host vehicle may deviate from the traveling lane.
In the alarm device 4, for example, at least one of a speaker, a buzzer, and a display corresponds to the alarm means.

Next, a driving support processing procedure by the driving support device 1 will be described based on the flowcharts of FIGS.
Si (i = 1, 2,...) Indicates steps for each process.
First, various information is read (S1), and the process proceeds to S2.
In S2, a determination threshold value setting process for setting a determination threshold value T for comparison with the degree of approach is performed.

After the determination threshold T is set in S2, the process proceeds to S3, where it is determined whether or not the degree of approach between the host vehicle and the white line exceeds the determination threshold T.
As a result of the determination in S3, if the degree of approach exceeds the determination threshold value T, the process proceeds to S4 in order to alert the driver, the alarm device 4 is activated, and then the process returns.
As a result of the determination in S3, when the degree of approach does not exceed the determination threshold T, the safety is not deteriorated, and the process returns without operating the alarm device 4.

Next, the determination threshold value setting process performed in S2 will be described in detail.
As shown in FIG. 6, first, it is determined based on information from the voice interaction device 10 whether or not a conversation between passengers is being performed (S11).
As a result of the determination in S11, when a conversation between the passengers is performed, the process proceeds to S12, and it is determined whether or not the conversation between the passengers is a conversation between the driver and the passenger.
As a result of the determination in S12, when the dialogue between the passengers is a dialogue between the driver and the passenger, the process proceeds to S13, and it is determined whether or not the driver is the subject of the dialogue, so-called speech.

If the driver is speaking as a result of the determination in S13, the driver is the subject of the dialogue, so the process proceeds to S14, and the first determination threshold T2 is set.
In S14, the heat degree coefficient k1 is calculated by Expression (2) based on the degree of interest and the physical condition, and the first determination threshold value T2 is calculated by substituting the heat degree coefficient k1 into Expression (1).
For example, when the driver's state is low interest level and good physical condition, the heat intensity coefficient k1 = 1.2, so T2 = T1 + 1.2α, and when the interest level is normal and the physical condition is normal, the heat intensity coefficient Since k1 = 1.0, T2 = T1 + 1.0α. When the degree of interest is high and the physical condition is poor, the thermal intermediate coefficient k1 = −0.3, so T2 = T1−0.3α.
The first determination threshold T2 calculated in S14 is set as the determination threshold T (S15), and the process ends.

As a result of the determination in S13, if the driver is not speaking, the passenger is the subject of the conversation and the driver is in the listening state, so the process proceeds to S16, and the second determination threshold T3 is set.
In S16, the heat degree coefficient k2 is calculated by Expression (2) based on the degree of interest and physical condition, and the second determination threshold value T3 is calculated by substituting the heat degree coefficient k2 into Expression (1).
For example, when the driver's state is low interest level and good physical condition, the heat intensity coefficient k2 = 1.0, so T3 = T1 + 1.0β. Since k2 = −0.3, T3 = T1−0.3β. When the degree of interest is high and the physical condition is poor, the thermal intermediate coefficient k2 = −0.5, so T3 = T1−0.5β. Become.
The second determination threshold T3 calculated in S16 is set as the determination threshold T (S17), and the process ends.

  As a result of the determination of S11, when the conversation between the passengers is not performed, and as a result of the determination of S12, when the conversation between the passengers is not a conversation between the driver and the passenger (a conversation between the passengers) Since the driver maintains the normal attention level and the normal operation characteristics, the reference determination threshold T1 is set to the determination threshold T (S18), and the process ends.

Next, the operation and effect of the driving support device 1 will be described.
According to the present driving support device 1, since the voice dialogue device 10 that acquires the voice of the dialogue between the driver and the passenger is provided, the presence / absence of the passenger and whether the driver in the dialogue is in an utterance state is listened to. It is possible to clearly determine whether the state is present. A determination threshold setting unit 23 is provided for lowering the alarm level by the alarm device 4 when the driver is speaking to the passenger to be lower than the alarm level by the alarm device 4 when the driver is listening to the passenger's speech. Therefore, when the driver is in an utterance state where the attention level (cognitiveness) is higher than the listening state, the warning level can be lowered to give priority to dialogue between passengers, and the driver is more careful than the utterance state. When listening at a low power level, priority can be given to ensuring driving safety by raising the alarm level.

  The determination threshold setting unit 23 sets the determination threshold T (T2) when the driver is speaking to the passenger to be larger than the determination threshold T (T3) when the driver is listening to the passenger's speech. For this reason, it is possible to achieve both driving safety and prevention of interference between passengers with a simple configuration of adjusting the determination threshold T of the departure determination unit 22.

  Since the determination threshold value setting unit 23 makes the determination threshold value T (T2) when the driver is speaking to the fellow passenger larger than the determination threshold value T (T1) when the driver is not interacting with the fellow passenger, The troublesomeness caused by the lane departure warning can be further reduced.

  A heat degree determination unit 24 for determining a heat degree (heat degree coefficient k1, k2) for the driver's conversation based on words included in the voices of the driver and the passengers in the conversation is provided. When it is determined that the degree of enthusiasm for the driver's dialogue is high, in order to reduce the determination threshold T (T2, T3), a decrease in the driver's attention level can be determined by the degree of enthusiasm, and the enthusiasm for the dialogue The driving safety of the driver can be ensured.

Next, a modified example in which the embodiment is partially changed will be described.
1) In the above-described embodiment, the example in which the departure determination unit that performs the departure determination of the driving lane by the vehicle is used as the prerequisite technology has been described as the safety decrease determination unit. An overspeed determination unit for the speed limit and a reaction time determination unit for a change in the traffic light may be used as a prerequisite technology.

2) In the above embodiment, the example in which the approach level determination threshold is changed based on the driver's utterance state and listening state to control the alarm level has been described, but the approach degree determination threshold value is maintained constant. The intensity of the alarm level may be controlled based on the driver's utterance state and listening state.
Specifically, the volume when the driver is speaking is lower than the volume when the driver is listening, or the brightness and color of the display when the driver is speaking are It is also possible to suppress more than the brightness and color at the time.

3] In the above-described embodiment, an example of a voice interactive device capable of interacting with an occupant has been described as a voice acquisition unit. A dedicated device such as a directional microphone may be used, or a passenger detection device and a driver dedicated microphone may be provided.

4] In the above-described embodiment, an example has been described in which the heat intensity coefficient is set by adding the degree of interest and the degree of physical condition. However, it may be set only by the degree of interest or only by the degree of physical condition.
It is also possible to provide a map in which a determination threshold is defined and extract the determination threshold from the map as needed based on determination elements such as speech, listening, interest, and physical condition.

5] In addition, those skilled in the art can implement the present invention with various modifications added without departing from the spirit of the present invention, and the present invention includes such modifications. is there.

DESCRIPTION OF SYMBOLS 1 Driving assistance device 4 Alarm device 10 Voice dialogue device 22 Deviation determination device 23 Determination threshold value setting part 24 Heat degree determination part

Claims (4)

Safety reduction determination means for determining a reduction in vehicle safety based on the running state of the vehicle, and alarm means for warning the driver when the determination value of the safety reduction determination means exceeds a determination threshold value In the driving support device provided,
Voice acquisition means for acquiring voice of dialogue between the driver and passenger,
An alarm level changing means for lowering an alarm level by the alarm means when the driver is speaking to the passenger lower than an alarm level by the alarm means when the driver is listening to the passenger's speech; and
A driving support apparatus comprising:   The warning level changing means makes the determination threshold when the driver is speaking to the passenger larger than the determination threshold when the driver is listening to the passenger's speech. Item 2. The driving support device according to Item 1.   The warning level changing means makes the determination threshold value when the driver is speaking to the passenger larger than the determination threshold value when the driver is not interacting with the passenger. The driving assistance apparatus as described. A degree of heat determination means for determining a degree of heat for the driver's dialogue based on words included in the voices of the driver and the passenger during the conversation,
4. The driving support apparatus according to claim 2, wherein the determination threshold value is reduced when it is determined by the heat degree determination unit that the heat degree with respect to the driver's dialogue is high.