JP2009251860A - Parking support device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a parking support device for precisely estimating a parking area intended by a driver. <P>SOLUTION: A parking support device is provided with: an imaging means for imaging a driver; a direction detection means for detecting the direction of at least one of the directions of the face and the direction of the line of sight of the driver; a non-linear state detection means for detecting such a non-linear state that a steering angle is larger than a prescribed value; a duration integration means for integrating a duration in which the driver visually observes a parking area for each parking area; and a parking support means for determining the parking area whose duration is the longest or the parking area whose duration exceeds a prescribed time as a target parking area, and for supporting parking to the target parking area. The duration integration means stops or resets integration processing of the duration when the non-linear state is detected by the non-linear state detection means. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a parking support apparatus that supports parking.

Conventionally, the driver estimates the parking space intended for parking based on the orientation information regarding the driver's face orientation and / or gaze direction, and advances to the parking start position based on the estimated parking space. A parking assistance device that performs assistance at a stage and assistance at a stage of setting a target parking position is known (see, for example, Patent Document 1).
JP 2007-253819 A

  Incidentally, the direction of the driver's face and / or the direction of the line of sight is useful information for estimating a parking area intended by the driver, as described in Patent Document 1 above. However, since the driver generally does not always look at the intended parking area, the driver appropriately determines at which timing the detected face direction and / or gaze direction is used for estimation. This is important for improving the estimation accuracy.

  Therefore, the present invention aims to provide a parking assistance device that can accurately estimate a parking area intended by a driver by using face orientation and / or gaze orientation information selected under appropriate conditions. And

In order to achieve the above object, a first invention provides a parking assistance device for assisting parking,
Imaging means for imaging the driver;
A direction detection unit that performs image processing on a captured image of the imaging unit and detects at least one of the direction of the driver's face and the direction of the line of sight;
Non-straight running state detection means for detecting a non-straight running state in which the steering angle is greater than a predetermined value based on the detection result of the steering angle of the steering;
Based on the detection result of the orientation detection means, duration integration means for integrating the duration for which the driver visually observes the parking area for each parking area;
Parking that assists parking in the target parking area by determining a parking area having the longest duration or a parking area having a duration exceeding a predetermined time based on the integration result of the duration integration means as a target parking area With support means,
The duration integration unit stops or resets the duration integration process when the non-straight running state is detected by the non-straight running state detection unit.

2nd invention is the parking assistance apparatus which concerns on 1st invention,
A non-low speed state detecting means for detecting a non-low speed state in which the vehicle speed is greater than a predetermined value based on the detection result of the vehicle speed;
The duration integration unit stops or resets the duration integration process even when a non-low-speed state is detected by the non-low-speed state detection unit.

3rd invention is the parking assistance apparatus which concerns on 1st or 2nd invention,
The vehicle further comprises stop state detection means for detecting a stop state for a predetermined time or more,
The duration integration unit stops or resets the integration process of the duration even when the stop state is detected by the stop state detection unit.

4th invention is the parking assistance apparatus which assists parking,
Imaging means for imaging the driver;
A direction detection unit that performs image processing on a captured image of the imaging unit and detects at least one of the direction of the driver's face and the direction of the line of sight;
At least one of a non-straight traveling state in which the steering angle is greater than a predetermined value based on the detection result of the steering angle of the steering wheel, a non-low speed state in which the vehicle speed is greater than the predetermined value based on the detection result of the vehicle speed, Predetermined state detecting means for detecting any one state;
Based on the detection result of the orientation detection means, duration integration means for integrating the duration for which the driver visually observes the parking area for each parking area;
Parking that assists parking in the target parking area by determining a parking area having the longest duration or a parking area having a duration exceeding a predetermined time based on the integration result of the duration integration means as a target parking area With support means,
The duration integration unit stops or resets the integration process of the duration when the at least one state is detected by the predetermined state detection unit.

5th invention is the parking assistance apparatus which assists parking,
Imaging means for imaging the driver;
A direction detection unit that performs image processing on a captured image of the imaging unit and detects at least one of the direction of the driver's face and the direction of the line of sight;
A straight traveling state detection means for detecting a straight traveling state in which the steering angle is equal to or less than a predetermined value based on the detection result of the steering angle of the steering;
Of the detection results of the direction detection means, the duration for which the driver visually observes the parking area is determined for each parking area using only the detection result of the direction detection means in the straight traveling state detected by the straight traveling state detection means. A duration integration means for integrating;
Parking that assists parking in the target parking area by determining a parking area having the longest duration or a parking area having a duration exceeding a predetermined time based on the integration result of the duration integration means as a target parking area And a support means.

6th invention is the parking assistance apparatus which concerns on 5th invention,
A low speed state detecting means for detecting a low speed state in which the vehicle speed is equal to or lower than a predetermined value based on a detection result of the vehicle speed;
The duration integration means uses only the detection result of the direction detection means in the straight traveling state and the low speed state detected by the straight traveling state detection means and the low speed state detection means among the detection results of the direction detection means, The duration time is integrated.

7th invention is the parking assistance apparatus which concerns on 5th or 6th invention,
The vehicle further comprises stop state detection means for detecting a stop state for a predetermined time or more,
The duration integration unit stops or resets the duration integration process when the stop state is detected by the stop state detection unit.

The eighth invention is the parking assist device according to any one of the first to seventh inventions,
The parking assist means automatically turns on a hazard lamp when a vehicle approaches the target parking area.

A ninth invention is the parking assist apparatus according to the eighth invention,
The parking assist means automatically turns on a hazard lamp when a vehicle approaches the target parking area and a subsequent vehicle is detected.

  ADVANTAGE OF THE INVENTION According to this invention, the parking assistance apparatus which can estimate the parking area which a driver | operator intends accurately is obtained.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  FIG. 1 is a system configuration diagram showing a basic configuration of a parking assistance apparatus 10 according to the present invention. As shown in FIG. 1, the parking assistance device 10 is configured around an electronic control unit 12 (hereinafter referred to as “parking assistance ECU 12”). Parking assistance ECU12 is comprised as a microcomputer which consists of CPU, ROM, RAM, etc. which were mutually connected via the bus | bath which is not shown in figure. The ROM stores programs and data executed by the CPU.

  The parking assist ECU 12 detects a steering angle sensor 16 that detects a steering angle of a steering wheel (not shown) and a vehicle speed via an appropriate bus such as a CAN (Controller Area Network) or a high-speed communication bus. A vehicle speed sensor 18 is connected. The vehicle speed sensor 18 may be a wheel speed sensor that is disposed on each wheel and generates a pulse signal at a cycle according to the wheel speed.

  A face orientation monitoring camera 74 is connected to the parking assist ECU 12. The face-facing monitoring camera 74 is, for example, a camera including a color or infrared-sensitive CCD (charge-coupled device) sensor array. For example, it is provided at an appropriate location of the vehicle so that the front surface of the driver (for example, the face portion from the front) can be captured. For example, the face-facing monitoring camera 74 is installed at an appropriate location such as a vehicle rearview mirror, a dashboard of an instrument panel, or a steering column. The face orientation monitoring camera 74 may acquire an image of the driver in real time and supply it to the parking assist ECU 12 in a stream format with a predetermined frame period (for example, 30 fps).

  The front LKA camera 20 is connected to the parking assist ECU 12. The front LKA camera 20 is mounted at an appropriate location (for example, the front side of a room mirror) at the front of the vehicle so as to capture a landscape in a predetermined angle region in front of the vehicle. The front LKA camera 20 may be a camera that includes a CCD, a CMOS, or the like in the imaging element, like the face orientation monitoring camera 74. The front LKA camera 20 is preferably a lane shooting camera used in Lane Keep Assist (LKA), but may be another camera. The front LKA camera 20 may acquire an environment image in front of the vehicle in real time and supply it to the parking assist ECU 12 in a stream format with a predetermined frame period (for example, 30 fps).

  In addition, a navigation device 32 including a GPS receiver, a hazard lamp 22 and a speaker 24 are connected to the parking assist ECU 12. The hazard lamp 22 is an emergency flashing lamp and is installed at a specified location. The speaker 24 may be provided at any place in the vehicle interior.

  Next, an example of main processing realized by the parking assistance ECU 12 of the present embodiment will be described with reference to the flowcharts of FIGS. 2, 4, and 7 and the explanatory diagrams of FIGS. 3, 5, and 6.

  FIG. 2 is a flowchart illustrating an example of a parking area candidate determination process realized by the parking assist ECU 12 according to the present embodiment.

  In step 100, parking area information (map information) from the navigation device 32 is acquired.

  In step 102, based on the parking area information and the own vehicle position information from the navigation device 32, it is determined whether or not the current own vehicle position is within the parking area defined in the map data. If the current vehicle position is within the parking area, the process proceeds to step 104. If the current vehicle position is outside the parking area, the process returns to step 100.

  In step 104, the captured image of the front LKA camera 20 is subjected to image processing, and a parking empty area detection process in front of the vehicle is executed. There are various parking space detection methods, and any appropriate method may be adopted. For example, if a parking frame line (white line) is image-recognized and there is no edge representing an obstacle in the area surrounded by the parking frame line, the area may be detected as a parking empty area. Moreover, the detection of a parking empty area may be implement | achieved using an ultrasonic sensor. Moreover, the detection of a parking vacant area may be accompanied by the detection of a plurality of parking vacant areas.

  In step 106, it is determined whether or not a parking empty area is detected in front of the vehicle as a result of the processing in step 104 (including the result before the previous cycle). If a parking empty area is detected, the process proceeds to step 108. If a parking empty area is not detected, the process returns to step 104.

  In step 108, the detected parking vacant areas are designated as parking area candidates B1, B2, B3,. . . , Bn, and information representing parking area candidates is generated. This parking area candidate information may be generated in association with the position information of the detected parking empty area, for example. FIG. 3 is a plan view showing an example of a situation in the parking lot area. In the example shown in FIG. 3, parking area candidates B1, B2, B3, and B4 are determined. When the process of step 108 is completed, the process proceeds to the process of FIG.

  FIG. 4 is a flowchart illustrating an example of the visual direction evaluation process realized by the parking assist ECU 12 of the present embodiment.

In step 110, it is determined whether a reset condition is satisfied. The reset condition is satisfied when any of the following conditions (1) to (3) is satisfied. That is, as the reset condition, the following conditions (1) to (3) are used as the OR condition.
Condition (1) Based on the detection result of the steering angle sensor 16, a non-straight running state in which the steering angle is greater than the predetermined value X [deg] is detected. The predetermined value X corresponds to the maximum value of the range that can be taken by the steering angle in the substantially straight traveling state, and is a suitable value, but may be, for example, 10 degrees.
Condition (2) Based on the detection result of the vehicle speed sensor 18, a non-low speed state in which the vehicle speed is greater than the predetermined value V1 [km / h] is detected. The predetermined value V1 corresponds to the maximum value of the range that the vehicle speed can take in the low speed state, and is a conforming value, but may be, for example, 10 [km / h].
Condition (3) A stop state of a predetermined time Ta [sec] or more is detected. The predetermined time Ta is a suitable value that eliminates an instantaneous stop, but may be 3 [sec], for example. The stop state may be detected when the vehicle speed is smaller than a predetermined value V2 (for example, 2 [km / h]) based on the detection result of the vehicle speed sensor 18.

  In step 110, if the reset condition is satisfied, that is, if any of the conditions (1) to (3) is satisfied, the process proceeds to step 112, and if the reset condition is not satisfied, that is, the condition (1) If none of (3) to (3) is established, the process proceeds to step 114.

  In step 112, durations A1, A2, A3,. . . , An (initial value zero) is reset, and the process returns to step 104 in FIG. That is, durations A1, A2, A3,. . . , An are all reset to zero, and the processing after step 104 in FIG. 2 is repeated again from the beginning. Face durations A1, A2, A3,. . . , An will be described later in detail. In an alternative embodiment, in this step 119, durations A1, A2, A3,. . . , An may be held without being reset, and the processing of the current cycle may be terminated and the process may return to step 104 in FIG. That is, the durations A1, A2, A3,. . . , An may be maintained, and the process may return to step 104 in FIG. 2 without performing integration of the current cycle.

  In step 114, the driver's face orientation is recognized based on the captured image of the face orientation monitoring camera 74. That is, by the image processing on the image acquired by the face orientation monitoring camera 74, the driver's face orientation is recognized, and information regarding the driver's face orientation is generated. For example, first, representative points (feature points) of the face are extracted from the image acquired by the face orientation monitoring camera 74. This feature extraction method may be any appropriate method, and for example, a technique based on Active Appearance Model (AAM) may be used. An appropriate edge detection algorithm (eg, Sobel's edge detection algorithm) is then applied to extract the boundary between the face and facial features. Next, using the edges and feature points extracted as described above, the facial parts are separated as shown in FIG. 5, and the shape of the facial parts is extracted. FIG. 5 shows only a part of the face as a representative, and shows an image in which the eyebrows and eyes are extracted as parts. Next, the current posture (face orientation) is detected by comparing the degree of matching between the position or orientation of the extracted facial part and the position or orientation of the same part stored in advance. The orientation of the face may be represented by, for example, a rotation angle in a horizontal plane with respect to the front direction of the face when in a normal posture.

  Note that there are various face orientation detection methods by image recognition processing, and the present invention is not limited to the above-described one. For example, the outline of the eyeball may be extracted as a facial part, and the direction of the driver's line of sight may be detected based on the position of the eyeball in the outline of the eye. The driver's gaze direction and driver's face direction are parameters that are effective in estimating the driver's intention, so they can be used alone or in combination. is there. Hereinafter, parking assistance using the direction of the driver's face will be described. However, instead of or in addition to the face direction, the direction of the driver's line of sight can also be used.

  In step 116, either the left or right area direction is selected from the continuous distribution of face orientations based on the detection result in step 114. If the duration of the face orientation to the left is longer than the duration of the face orientation to the right, the left area direction is selected and the duration of the face orientation to the right is If it is longer than the duration of the face orientation in the direction, the right area direction is selected. Note that the processing of step 116 uses the detection result of step 114 over a predetermined time (for example, the detection result for a predetermined period before the current period including the current period) in order to evaluate the continuous distribution in the face direction. .

  In step 118, the process of integrating the face orientation duration in the selected area direction is executed. For example, when the right area direction is selected in step 116, the angle ranges a1, a2, a3,. . . , An, the durations A1, A2, A3,. . . , An are updated. For example, when the face orientation detected in the current cycle is within the angle range a1, the duration A1 related to the angle range a1 is integrated for a predetermined time (a time corresponding to one cycle). Each angle range a1, a2, a3,. . . , An are determined in association with the parking area candidates detected as described above. That is, each angular range a1, a2, a3,. . . , An are not constant, and change according to the positional relationship between the vehicle position and the parking area candidate that change as the vehicle moves. Accordingly, the durations A1, A2, A3,. . . , An are parking area candidates B1, B2, B3,. . . , Bn corresponding to the face orientation duration time. In the example shown in FIG. 6, each angle range a1, a2, a3 linked | related with parking area candidate B1, B2, B3 is shown.

  In this step 118, the integrated value of the duration of the face direction may be expressed (evaluated) using the number of cycles, the frequency, or the like in which the face direction directed to each angle range is detected.

  In step 119, the durations A1, A2, A3,. . . , An, it is determined whether the total duration has exceeded a predetermined time. The predetermined time is longer than the general viewing time when the driver visually observes the parking area intended for parking, and there is a significant difference in continuation for each face to determine the parking area intended for parking. Time A1, A2, A3,. . . , An corresponds to the time that appears between An. When the total duration exceeds the predetermined time, the process proceeds to the process of FIG. 7 described later. When the total duration is equal to or shorter than the predetermined time, the process returns to step 104 of FIG. 2 and the total duration exceeds the predetermined time. The processes after step 104 in FIG.

  FIG. 7 is a flowchart illustrating an example of the target parking area determination process and the parking support process realized by the parking support ECU 12 of the present embodiment. The processing routine shown in FIG. 7 is started when the parking area candidate and the face direction duration time are determined and calculated as described above.

  In step 120, matching between the parking area candidate and the face direction duration is evaluated. That is, it is determined whether there is a parking area candidate to which the driver's face is directed among the parking area candidates. At this time, preferably, the face orientation duration time is also considered. That is, among the parking area candidates, it is determined whether or not there is a parking area candidate for which the face direction duration time equal to or longer than the predetermined time Tk is calculated. The predetermined time Tk corresponds to a general viewing time when the driver visually observes a parking area intended for parking, and is a suitable value, but may be, for example, 3 seconds. When there is a parking area candidate to which the driver's face is directed, the parking area candidate is determined as the target parking area, and the process proceeds to step 126. If there are a plurality of parking area candidates to which the driver's face is directed, the parking area candidate with the longest face direction continuation time (highest priority) is determined as the target parking area. move on. On the other hand, when the parking area candidate does not match the face duration time (for example, when there is no parking area candidate where the driver's face is directed for a duration time equal to or longer than the predetermined time Tk), the driver Determines that it has also decided the parking area for parking, and proceeds to step 122.

  In step 122, for example, an empty parking area close to the current vehicle position is guided by voice via the speaker 24, for example. Note that video guidance may be executed via a display (not shown) instead of or in addition to voice guidance.

  In step 124, the hazard lamp 22 is turned on in response to the driver's operation. That is, when the driver turns on the hazard switch, the hazard lamp 22 for informing the rear vehicle of the parking intention is turned on. Alternatively, the hazard lamp 22 may be automatically turned on when approaching a parking area guided by the own vehicle.

  In step 126, the vehicle is guided to the matching parking area candidate (that is, the target parking area). This vehicle guidance may be realized by voice guidance through the speaker 24 or may be realized by video guidance through a display (not shown). Alternatively, it may be realized by automatic guidance control using an automatic steering system or the like.

  In step 128, based on the relationship between the position information of the own vehicle and the position information of the target parking area, the hazard lamp 22 is automatically turned on as an intention to park when the vehicle approaches the target parking area. . For example, the hazard lamp 22 is automatically turned on when the vehicle reaches a predetermined distance (for example, 1 m) before the target parking area. This eliminates the need for the user to turn on the hazard switch. It should be noted that the automatic lighting of the hazard lamp 22 may be realized only when a subsequent vehicle is detected by, for example, a rear monitoring camera or a radar.

  According to the parking assistance apparatus 10 of the present embodiment described above, the following excellent effects are achieved, among others.

  According to the present embodiment, as described above, when the non-straight running state in which the steering angle is greater than the predetermined value X is detected, the calculation processing of the duration of the face direction is reset or stopped, so the face in the non-straight running state The parking area intended by the driver can be estimated without considering the direction. Here, the driver typically tends to search for a desired parking area while going straight, and the non-straight driving state typically means that the driver does not find the intended parking area in the parking lot. It is formed when moving to another place (section), or when the intended parking area is found and the steering wheel is turned to reach the parking start position suitable for the parking area. Therefore, according to the present embodiment, it is possible to accurately estimate the parking area intended by the driver by estimating the parking area intended by the driver without considering the face orientation in the non-straight running state. .

  In addition, according to the present embodiment, as described above, when the non-low speed state in which the vehicle speed is greater than the predetermined value V1 is detected, the calculation process for the duration of the face direction is reset or stopped. The parking area intended by the driver can be estimated without considering the face orientation. Here, the driver typically tends to search for the desired parking area at low speed, and the non-low speed condition typically results in the driver not finding the intended parking area in the parking lot. It is formed when moving to another location (section). Therefore, according to the present embodiment, it is possible to accurately estimate the parking area intended by the driver by estimating the parking area intended by the driver without considering the face orientation in the non-low speed state. .

  Further, according to the present embodiment, as described above, when the stop state of the predetermined time Ta or more is detected, the calculation process of the face duration duration is reset or stopped, so the stop state of the predetermined time Ta or more The parking area intended by the driver can be estimated without taking into account the face orientation. Here, the driver typically has a tendency to search for a desired parking area in a low-speed state, and a stopped state of a predetermined time Ta or more typically indicates a parking area that the driver intends in the parking lot. It is formed when you are not sure where to park without being found. In such a case, the driver typically tends to look around. Therefore, according to the present embodiment, the parking area intended by the driver is accurately estimated by estimating the parking area intended by the driver without considering the face orientation in the stopped state for the predetermined time Ta or longer. can do.

  The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above-described embodiments, and various modifications and substitutions can be made to the above-described embodiments without departing from the scope of the present invention. Can be added.

  For example, in the embodiment described above, if a negative determination is made in step 120, the process proceeds to step 122. However, the process after step 122 may be omitted and the process may return to step 104 in FIG. In this case, the process of step 119 may also be omitted, and the process of step 120 may be executed after the process of step 118.

1 is a system configuration diagram showing a basic configuration of a parking assistance device 10 according to the present invention. It is a flowchart which shows an example of the parking area candidate determination process implement | achieved by parking assistance ECU12 of a present Example. It is a top view which shows an example of the condition in a parking lot area. It is a flowchart which shows an example of the visual direction evaluation process implement | achieved by parking assistance ECU12 of a present Example. It is explanatory drawing of the image recognition method of a viewing direction, and is a figure which shows the image from which the eyebrows and eyes of the face were extracted as parts. It is a top view which shows the correspondence of parking area candidate B1, B2, B3, B4 in a parking lot area, and each angle range a1, a2, a3, a4. It is a flowchart which shows an example of the target parking area determination process and parking assistance process which are implement | achieved by parking assistance ECU12 of a present Example.

Explanation of symbols

10 Parking assistance device 12 Parking assistance ECU
16 Steering angle sensor 18 Vehicle speed sensor 20 Front LKA camera 22 Hazard lamp 24 Speaker 32 Navigation device 74 Face orientation monitoring camera

Claims (9)

In a parking assistance device that supports parking,
Imaging means for imaging the driver;
A direction detection unit that performs image processing on a captured image of the imaging unit and detects at least one of the direction of the driver's face and the direction of the line of sight;
Non-straight running state detection means for detecting a non-straight running state in which the steering angle is greater than a predetermined value based on the detection result of the steering angle of the steering;
Based on the detection result of the orientation detection means, duration integration means for integrating the duration for which the driver visually observes the parking area for each parking area;
Parking that assists parking in the target parking area by determining a parking area having the longest duration or a parking area having a duration exceeding a predetermined time based on the integration result of the duration integration means as a target parking area With support means,
The parking assistance device according to claim 1, wherein the duration integration unit stops or resets the integration process of the duration when a non-straight running state is detected by the non-straight running state detection unit. A non-low speed state detecting means for detecting a non-low speed state in which the vehicle speed is greater than a predetermined value based on the detection result of the vehicle speed;
The parking assistance device according to claim 1, wherein the duration integration unit stops or resets the integration process of the duration even when a non-low speed state is detected by the non-low speed state detection unit. The vehicle further comprises stop state detection means for detecting a stop state for a predetermined time or more,
The parking assistance device according to claim 1 or 2, wherein the duration integration unit stops or resets the duration integration processing even when the stop state is detected by the stop state detection unit. In a parking assistance device that supports parking,
Imaging means for imaging the driver;
A direction detection unit that performs image processing on a captured image of the imaging unit and detects at least one of the direction of the driver's face and the direction of the line of sight;
At least one of a non-straight traveling state in which the steering angle is greater than a predetermined value based on the detection result of the steering angle of the steering wheel, a non-low speed state in which the vehicle speed is greater than the predetermined value based on the detection result of the vehicle speed, Predetermined state detecting means for detecting any one state;
Based on the detection result of the orientation detection means, duration integration means for integrating the duration for which the driver visually observes the parking area for each parking area;
Parking that assists parking in the target parking area by determining a parking area having the longest duration or a parking area having a duration exceeding a predetermined time based on the integration result of the duration integration means as a target parking area With support means,
The parking assistance device, wherein the duration integration unit stops or resets the integration processing of the duration when the at least one state is detected by the predetermined state detection unit. In a parking assistance device that supports parking,
Imaging means for imaging the driver;
A direction detection unit that performs image processing on a captured image of the imaging unit and detects at least one of the direction of the driver's face and the direction of the line of sight;
A straight traveling state detection means for detecting a straight traveling state in which the steering angle is equal to or less than a predetermined value based on the detection result of the steering angle of the steering;
Of the detection results of the direction detection means, the duration for which the driver visually observes the parking area is determined for each parking area using only the detection result of the direction detection means in the straight traveling state detected by the straight traveling state detection means. A duration integration means for integrating;
Parking that assists parking in the target parking area by determining a parking area having the longest duration or a parking area having a duration exceeding a predetermined time based on the integration result of the duration integration means as a target parking area A parking support device comprising a support means. A low speed state detecting means for detecting a low speed state in which the vehicle speed is equal to or lower than a predetermined value based on a detection result of the vehicle speed;
The duration integration means uses only the detection result of the direction detection means in the straight traveling state and the low speed state detected by the straight traveling state detection means and the low speed state detection means among the detection results of the direction detection means, The parking assistance device according to claim 5, wherein the duration time is integrated. The vehicle further comprises stop state detection means for detecting a stop state for a predetermined time or more,
The parking assistance device according to claim 5 or 6, wherein the duration integration unit stops or resets the integration process of the duration when the stop state is detected by the stop state detection unit.   The parking support device according to any one of claims 1 to 7, wherein the parking support means automatically turns on a hazard lamp when a vehicle approaches the target parking area.   The parking support device according to claim 8, wherein the parking support means automatically turns on a hazard lamp when a vehicle approaches the vicinity of the target parking area and a subsequent vehicle is detected.

Images