JP4729522B2 - Navigation device, navigation method, and program - Google Patents

Description

  The present invention relates to a navigation device used by being mounted on a vehicle.

  2. Description of the Related Art Conventionally, a navigation device that assists driving by detecting a driver's line of sight, recognizing an object such as a building that the driver is viewing, and providing information related to the object is known (Patent Document 1). .

The navigation device described in Patent Literature 1 includes a line-of-sight detection camera for detecting the line of sight of the driver and a microphone that acquires the voice of the driver. The navigation device acquires the voice of the driver with a microphone. When the acquired voice is analyzed and a keyword that triggers such as “What is that?” Is recognized, the navigation device detects the driver's line of sight from the video captured by the camera, such as a building in the driver's line of sight, etc. Display the object on the screen. Next, the navigation device acquires correction commands such as “right” and “back” with a microphone, and selects an object in accordance with the correction commands. The navigation device performs route setting, guide information display, and the like based on the selected object.
JP-A-2005-24313

  In the navigation device described in Patent Literature 1 described above, on the assumption that the target cannot be specified only by the line-of-sight direction, the target is specified by receiving a voice trigger or a correction command.

  However, since the driver utters various words, there is a risk of erroneously recognizing a trigger or a correction command. Although it is possible to set a specific keyword as a trigger or correction command in advance, the convenience is reduced.

  Further, in the navigation device described in Patent Document 1, the driver has to look at the target object until the direction of the line of sight is specified at the same time as generating a trigger voice, which may hinder driving. .

  SUMMARY OF THE INVENTION In view of the above background, an object of the present invention is to provide a navigation device that can appropriately identify an object using the movement of a driver's line of sight.

  The navigation device of the present invention includes a line-of-sight pattern storage unit that stores a movement pattern of a driver's line of sight when gazing at an object inside and outside the vehicle as a line-of-sight pattern associated with the object, and line-of-sight detection that detects the line of sight of the driver And the movement of the driver's line of sight based on the line-of-sight information detected by the line-of-sight detection unit, and the object corresponding to the line-of-sight pattern matching the line-of-sight movement is read from the line-of-sight pattern storage unit A gaze target detection unit that detects the target as a gaze target.

  In this way, by determining the movement of the driver's line of sight and detecting the gaze target based on the movement of the line of sight, the driver can appropriately detect the gaze target even if the driver does not look at the target for a certain period of time. .

  In the navigation device of the present invention, the line-of-sight pattern storage unit stores a plurality of line-of-sight patterns different for each environment in association with the object, and the gaze target detection unit obtains the movement of the line of sight. The gaze pattern may be selected according to the environment, and the gaze object may be detected using the selected gaze pattern.

  As described above, by storing a plurality of line-of-sight patterns different for each environment in association with the object in the line-of-sight pattern storage unit, it is possible to appropriately cope with a case where the way of movement of the line of sight changes depending on the environment. Here, the environment includes, for example, the traveling speed of the vehicle, the driving experience of the driver, the external brightness, and the like. The navigation device acquires environmental information from various sensors. For example, information on the traveling speed of the vehicle is obtained from a speed sensor, and information on external brightness is obtained from an illuminance sensor. The driver's driving experience information is acquired from an input device that allows the driver to input the years of driving experience.

  The navigation device of the present invention includes a gaze behavior pattern storage unit that stores a plurality of gaze objects to be watched prior to a driver's driving behavior and a gaze order as a gaze behavior pattern in association with the driving behavior, and the gaze target. A gaze action is obtained based on information on the gaze object detected by the detection unit, a driving action corresponding to the gaze action pattern that matches the gaze action is read from the gaze action pattern storage unit, and the read driving action is watched A predicted driving behavior determining unit that determines the predicted driving behavior to be performed subsequently to the predicted driving behavior, and a driving support signal transmitting unit that transmits a driving support signal for supporting the predicted driving behavior may be provided.

  Thus, the driver's gaze behavior is obtained, and the driving support signal for supporting the predicted driving behavior determined based on the gaze behavior can be transmitted to appropriately support the driving based on the information obtained from the driver's line of sight. .

  In the navigation device of the present invention, the gaze behavior pattern storage unit stores a plurality of gaze behavior patterns that are different for each environment in association with the driving behavior, and the predicted driving behavior determination unit detects the gaze behavior. The gaze behavior pattern may be selected according to the environment, and the predicted driving behavior may be determined using the selected gaze behavior pattern.

  In this way, by storing a plurality of gaze behavior patterns that are different for each environment in association with the driving behavior in the gaze behavior pattern storage unit, it is possible to appropriately cope with the case where the driving behavior changes depending on the environment. Here, the environment includes, for example, the traveling speed of the vehicle, the driving experience of the driver, the external brightness, and the like.

  In the navigation device of the present invention, the predicted driving behavior determination unit obtains a gaze behavior from information on the gaze target detected by the gaze target detection unit, and the gaze behavior is stored in the gaze behavior pattern storage unit. When a part of the gaze behavior pattern is matched, the driving behavior corresponding to the gaze behavior pattern may be read.

  As described above, when the driver's gaze behavior matches a part of the gaze behavior pattern, the driver's predicted driving behavior can be determined at an early stage by determining the driver's predicted driving behavior.

  The navigation device of the present invention obtains a gaze action performed before the driving action when the driving action detecting unit detects the driving action from the operation of the vehicle and the driving action detecting unit detects the driving action. And a gaze behavior pattern learning unit that updates information stored in the gaze behavior pattern storage unit using the gaze behavior.

  In this way, by learning the information stored in the gaze behavior pattern storage unit based on the actual driving behavior and the gaze behavior, the information in the gaze behavior pattern storage unit can be adapted to each driver.

  The navigation device of the present invention obtains a driver's line-of-sight movement based on the line-of-sight information detected by the line-of-sight detection unit, searches the line-of-sight pattern storage unit for a line-of-sight pattern most similar to the line-of-sight movement, and searches And a line-of-sight pattern learning unit that updates the line-of-sight pattern using the movement of the line of sight of the driver.

  Thus, by learning the information stored in the line-of-sight pattern storage unit using similar line-of-sight patterns detected from the actual line-of-sight pattern, the information in the line-of-sight pattern storage unit can be adapted to each driver. .

  The navigation method of the present invention includes a step of preparing a line-of-sight pattern storage unit that stores a movement pattern of a driver's line of sight when gazing at an object inside and outside the vehicle as a line-of-sight pattern, and stores the line-of-sight of the driver. A step of detecting, a step of obtaining a movement of the driver's line of sight based on the information of the detected line of sight, an object corresponding to the line of sight pattern matching the movement of the line of sight is read from the line of sight pattern storage unit, and the read object Detecting as a gaze target.

  Thus, by detecting the gaze object based on the movement of the line of sight, the gaze object can be appropriately detected even if the driver does not look at the object for a certain period of time. Note that various configurations of the navigation device of the present invention can be applied to the navigation method of the present invention.

  The program of the present invention includes a step of detecting a driver's line of sight in a navigation device having a function of detecting a driver's line of sight, a step of obtaining a movement of the driver's line of sight based on information of the detected line of sight, The object corresponding to the line-of-sight pattern matching the movement is read from the line-of-sight pattern storage unit stored in association with the object as the line-of-sight pattern of the driver's line-of-sight movement when gazing at the object inside and outside the vehicle, Detecting the read target object as a gaze target.

  Thus, by detecting the gaze object based on the movement of the line of sight, the gaze object can be appropriately detected even if the driver does not look at the object for a certain period of time. Various configurations of the navigation device of the present invention can be applied to the program of the present invention.

  According to the present invention, by detecting a gaze object based on the movement of the line of sight, the gaze object can be appropriately detected even if the driver does not look at the object for a certain period of time.

Hereinafter, a navigation device according to an embodiment of the present invention will be described with reference to the drawings.
(First embodiment)
FIG. 1 is a diagram illustrating a configuration of a navigation device 1 according to the first embodiment. The navigation device 1 includes a camera 10 that captures a driver, a line-of-sight detection unit 12 that detects the line of sight of the driver based on video captured by the camera 10, and a gaze target based on the line of sight detected by the line-of-sight detection unit 12. Based on information on the gaze target detected by the gaze target detection unit 14 and the gaze target detected by the gaze target detection unit 14, the driving behavior predicted to be performed by the driver (referred to as “predicted driving behavior”) is determined. A predictive driving action determining unit 16 and a driving support signal transmitting part 18 for transmitting a driving support signal for supporting the driving action determined by the predictive driving action determining unit 16 to the vehicle are provided. Although not shown in FIG. 1, the navigation device 1 according to the first embodiment includes a map information storage unit that stores map information necessary for navigation, a vehicle position detection unit that detects a vehicle position, and map information. It also has a configuration provided in a normal navigation device, such as a display for display and a speaker for outputting a navigation message.

  Hereinafter, each structure of the navigation apparatus 1 shown in FIG. 1 is demonstrated. The camera 10 is installed at a position where the driver's face can be photographed, such as a dashboard or a room mirror. As the camera 10, for example, a CCD camera 10 can be used.

  The line-of-sight detection unit 12 detects the driver's line of sight by determining the position and orientation of the driver's head and the position of the eyeball relative to the head based on the video captured by the camera 10.

  The gaze target detection unit 14 has a function of detecting the target watched by the driver based on the driver's gaze information detected by the gaze detection unit 12 and the information stored in the gaze pattern storage unit 20.

  FIG. 2 is a diagram illustrating an example of information stored in the line-of-sight pattern storage unit 20. The line-of-sight pattern storage unit 20 stores, for example, information on line-of-sight patterns at the time of gazing at the gaze target for each gaze target such as “roadside” and “navigation”. The line-of-sight pattern includes a horizontal line-of-sight movement pattern, a vertical line-of-sight movement pattern, or a horizontal and vertical line-of-sight movement pattern, depending on the gaze target. FIG. 2 shows an example in which only the horizontal line-of-sight movement pattern is stored. For example, when the gaze target is a “high-rise building”, the vertical line-of-sight movement pattern is also associated. Remember. In the example illustrated in FIG. 2, the line-of-sight pattern storage unit stores a temporal change in the line-of-sight angle θ as viewed from the traveling direction of the vehicle as a line-of-sight pattern.

  FIG. 3 is a diagram illustrating in detail an example of information stored in the line-of-sight pattern storage unit 20. FIG. 3 shows an example of a horizontal line-of-sight pattern when the gaze target is “roadside”. As shown in FIG. 3, the line-of-sight pattern storage unit 20 stores a plurality of different line-of-sight patterns according to the speed of the vehicle. In the example shown in FIG. 3, when the vehicle speed is “high speed”, the number of times that the line of sight is moved beyond the threshold value X is T = D4 times or more in 4 seconds, and the total time is larger than F seconds. In this case, it is detected that the “roadside” is watched. When the speed of the vehicle is “medium speed”, the line-of-sight pattern is different from the case of “high speed” in that the time T used for determination is T = D3 seconds. In the example shown in FIG. 3, the time T used for determination has a relationship of D1 ≧ D2 ≧ D3 ≧ D4. Therefore, in order to detect the “roadside” gaze, it is necessary that the line-of-sight pattern appears in a shorter time as the vehicle speed increases. As described above, by storing a plurality of different line-of-sight patterns according to the speed of the vehicle, it is possible to appropriately detect the gaze target even when the movement of the line of sight varies according to the speed of the vehicle.

Note that the line-of-sight pattern for detecting the gaze object shown in FIGS. 2 and 3 is an example, and an appropriate determination formula can be used depending on the gaze object. For example, when the gaze target is an object outside the vehicle such as a building or a restaurant, the shift angle θ of the line of sight with respect to the traveling direction of the vehicle is θ _n <θ _{n + 1} (where n Satisfying the numerical value indicating the number of times) may be a condition for detecting the gaze target. Since the angle θ of the line of sight when looking at an object outside the vehicle from the traveling vehicle increases as the vehicle advances, the object outside the vehicle can be selected by using θ _n <θ _{n + 1} as a condition. Appropriately detect what you are looking at.

  Next, the process of the gaze target detection unit 14 will be described. The gaze target detection unit 14 first acquires information on the speed of the vehicle from a speed sensor (not shown) in the vehicle speed information acquisition unit 24, and selects a line-of-sight pattern used for detection of the gaze target based on the vehicle speed information. For example, when the vehicle is traveling at high speed, the line-of-sight pattern corresponding to “high speed” shown in FIG. 3 is selected. Although FIG. 3 shows a line-of-sight pattern when the gaze target is “roadside”, the gaze target detection unit 14 also uses other gaze targets such as “navigation” and “room mirror”, for example, A line-of-sight pattern for “high speed” is selected from a plurality of line-of-sight patterns.

  After selecting the line-of-sight pattern according to the vehicle speed, the gaze target detection unit 14 temporarily accumulates information on the line of sight detected by the line-of-sight detection unit 12, and uses the accumulated line-of-sight information to move the driver's line of sight Ask for. Subsequently, the gaze target detection unit 14 determines whether or not the movement of the driver's line of sight matches any of the selected line-of-sight patterns. If the line-of-sight pattern matches, the target corresponding to the line-of-sight pattern is determined. An object is detected as a gaze target.

  Next, the predicted driving action determination unit 16 will be described. The predicted driving behavior determination unit 16 is predicted to be performed following the gaze behavior based on the information on the gaze target detected by the gaze target detection unit 14 and the information stored in the gaze behavior pattern storage unit 22. It has a function to determine the predicted driving behavior.

  FIG. 4 is a diagram illustrating an example of information stored in the gaze behavior pattern storage unit 22. The gaze action pattern storage unit 22 stores a driver's gaze action pattern performed prior to the driving action. In the example shown in FIG. 4, the gaze behavior pattern “(1) front → (2) room mirror → (3) front → (4) side mirror → (5) front” ”is associated with the driving behavior“ lane change ”. It has been. Therefore, it is predicted that “lane change” will be performed following this gaze behavior pattern.

  Next, the process of the predicted driving action determination unit 16 will be described. The predicted driving action determination unit 16 temporarily accumulates information on the gaze object detected by the gaze object detection unit 14 and obtains the driver's gaze action using the accumulated gaze object information. The gaze behavior is a gaze object that the driver gazes at and the gaze order thereof. For example, the gaze behavior is information such as gaze object A → gaze object B → gaze object A → gaze object C.

  Subsequently, the predictive driving behavior determination unit 16 determines whether or not the driver's gaze behavior matches any one of the gaze behavior patterns stored in the gaze behavior pattern storage unit 22, and if it matches the gaze behavior pattern Then, the driving behavior corresponding to the gaze behavior pattern is read to determine the driving behavior.

  The driving support signal transmission unit 18 transmits a driving support signal for supporting the driving behavior determined by the predicted driving behavior determination unit 16 to the vehicle. For example, when the driving action is “left turn”, the driving support signal transmission unit 18 transmits a driving support signal that instructs the vehicle to take the left blinker.

  Next, operation | movement of the navigation apparatus 1 of this Embodiment is demonstrated. The navigation device 1 according to the present embodiment specifies a gaze object based on the movement of the driver's line of sight in addition to the route search to the destination and the navigation operation according to the set route. Driving assistance based on information. Hereinafter, the operation of driving assistance by the navigation device 1 will be described.

  FIG. 5 is a diagram illustrating a driving support operation by the navigation device 1 according to the present embodiment. First, the line-of-sight detection unit 12 of the navigation device 1 detects the line of sight of the driver using the video captured by the camera 10 (S10). The gaze target detection unit 14 of the navigation device 1 temporarily accumulates the information of the line of sight detected by the line of sight detection unit 12, and obtains the movement of the line of sight based on the accumulated information of the line of sight (S12). The gaze target detection unit 14 selects a line-of-sight pattern according to the traveling speed of the vehicle (S14). The gaze target detection unit 14 determines whether or not the driver's line-of-sight movement matches the selected line-of-sight pattern, and when the driver's line-of-sight movement matches any line-of-sight pattern, the line-of-sight pattern The object corresponding to is detected as a gaze object (S16).

  The predicted driving behavior determination unit 16 of the navigation device 1 temporarily accumulates information on the gaze target detected by the gaze target detection unit 14 and detects the driver's gaze behavior from the accumulated gaze target information ( S18). The predicted driving action determination unit 16 searches the gaze action pattern storage unit 22 for a gaze action pattern that matches the driver's gaze action. When the driver's gaze behavior matches any gaze behavior pattern stored in the gaze behavior pattern storage unit 22 (YES in S20), the predicted driving behavior determination unit 16 drives the driving behavior corresponding to the gaze behavior pattern. And predicting driving behavior is determined (S22). And the driving assistance signal transmission part 18 of the navigation apparatus 1 transmits the driving assistance signal for assisting the prediction driving action determined in the prediction driving action determination part 16 with respect to a vehicle (S24).

  If the driver's gaze behavior does not match any gaze behavior pattern stored in the gaze behavior pattern storage unit 22 (NO in S20), the process returns to step S16 to detect the gaze behavior, and the driver's gaze behavior is again performed. Detect (S18). The configuration and operation of the navigation device 1 according to the first embodiment have been described above.

  Since the navigation device 1 according to the present embodiment obtains the movement of the driver's line of sight and detects the gaze object based on the movement of the line of sight, the navigation apparatus 1 appropriately selects the gaze object even if the driver does not look at the object for a certain period of time. Can be detected. Thereby, it is possible to detect the gaze object without hindering driving.

  Further, in the navigation device 1 of the present embodiment, the line-of-sight pattern storage unit 20 stores a plurality of line-of-sight patterns that differ according to the traveling speed of the vehicle, and detects a gaze object according to the traveling speed of the vehicle. Since the line-of-sight pattern to be changed is changed, it is possible to appropriately detect the gaze target corresponding to the movement of the line of sight different depending on the traveling speed of the vehicle.

  Further, the navigation device 1 according to the present embodiment obtains the driver's gaze behavior based on the gaze target information detected by the gaze target detection unit 14, and is predicted to be performed following the driver's gaze behavior. Determine the expected driving behavior. And since the driving assistance signal which supports a prediction driving action is transmitted to a vehicle, even if a driver does not give directions consciously by voice etc., driving can be supported appropriately.

(Second Embodiment)
FIG. 6 is a diagram illustrating a configuration of the navigation device 2 according to the second embodiment of this invention. The basic configuration of the navigation device 2 of the second embodiment is the same as that of the navigation device 1 of the first embodiment, but in addition to the configuration of the navigation device 1 of the first embodiment, the gaze behavior As a configuration for learning the information stored in the pattern storage unit 22, a vehicle motion information acquisition unit 26, a driving behavior detection unit 28, and a gaze behavior learning unit 30 are provided. Hereinafter, these configurations will be described.

  The vehicle operation information acquisition unit 26 has a function of acquiring vehicle operation information from the vehicle. The vehicle operation information includes, for example, operation information of various devices such as steering wheel movement, blinker, air conditioner, audio, and opening / closing of windows.

  The driving behavior detection unit 28 has a function for obtaining driving behavior based on the vehicle information acquired by the vehicle operation information acquisition unit 26. For example, when it is detected that the steering wheel has been turned to the left after the left winker is taken out, a left-turn driving action is detected.

  The gaze behavior learning unit 30 learns the gaze behavior pattern based on the detected driving behavior and the gaze behavior performed prior to the driving behavior when the driving behavior detection unit 28 detects the driving behavior. Do.

  FIGS. 7A and 7B are diagrams for explaining learning of a gaze behavior pattern. FIG. 7A is a diagram illustrating an example of the driving behavior detected by the driving behavior detector 28 and the gaze behavior performed before the driving behavior. The information shown in FIG. 7A is a so-called “teacher signal” used for learning information in the gaze behavior pattern storage unit 22. FIG. 7B is a diagram illustrating an example of information stored in the gaze behavior pattern storage unit 22. The gaze behavior pattern storage unit 22 stores information on “driving behavior”, “gaze behavior”, and “weighting”. “Weighting” is information for determining a predicted driving action when the same gaze action pattern is stored for different driving actions, such as the gaze action pattern P2 and the gaze action pattern P3. The predictive driving action determination unit 16 determines that the weighting is greater than a predetermined threshold (for example, “4”) or more, and determines that it is the predictive driving action with the larger weighting. The predicted driving behavior is not determined even when there is a gaze behavior pattern that matches the gaze behavior.

  When the teacher signal shown in FIG. 7A is input, the gaze behavior learning unit 30 increases the weighting of the gaze behavior pattern P2 corresponding to the teacher signal as shown by underlining in FIG. 7B. . With this learning, the weighting difference between the gaze action pattern P2 and the gaze action pattern P3 becomes “4”. For example, when the threshold for determining the predicted driving behavior among a plurality of driving behaviors associated with the same gaze behavior pattern is “4”, when a gaze behavior that matches the gaze behavior pattern P2 is detected, It can be determined that the driving action is “lane change”. Here, an example in which the same gaze behavior as the gaze behavior pattern already stored in the gaze behavior pattern storage unit 22 is input as the teacher signal has been described. However, any gaze behavior pattern stored in the gaze behavior pattern storage unit 22 is described. If a gaze action that does not match both is input as a teacher signal, it is registered as a new gaze action pattern. The navigation device 2 according to the second embodiment has been described above.

  Since the navigation device 2 according to the second embodiment performs learning in the gaze behavior learning unit 30 and updates the information stored in the gaze behavior pattern storage unit 22, the navigation device 2 responds to the driver's habit and the like. Therefore, the predicted driving behavior can be determined appropriately.

(Third embodiment)
FIG. 8 is a diagram showing a configuration of the navigation apparatus 3 according to the third embodiment of the present invention. The basic configuration of the navigation device 3 of the third embodiment is the same as that of the navigation device 2 of the second embodiment, but in addition to the configuration of the navigation device 2 of the second embodiment, the line-of-sight pattern A line-of-sight pattern learning unit 32 is provided for learning information stored in the storage unit 20.

  The line-of-sight pattern learning unit 32 obtains the movement of the line of sight based on the information on the line of sight of the driver detected by the line-of-sight detection unit 12. Next, the line-of-sight pattern learning unit 32 searches the line-of-sight pattern storage unit 20 for a line-of-sight pattern similar to the obtained line-of-sight movement. The line-of-sight pattern learning unit 32 learns similar line-of-sight patterns using the driver's line-of-sight movement as a teacher signal.

  FIG. 9A is a diagram illustrating an example of a line-of-sight pattern stored in the line-of-sight pattern storage unit 20, and FIG. 9B is a diagram illustrating an example of the obtained line-of-sight movement. Note that the line-of-sight pattern illustrated in FIG. 9A is a line-of-sight pattern when the gaze target is “roadside”. As shown in FIG. 9A, the line-of-sight pattern stored in the line-of-sight pattern storage unit 20 is such that the total time tn when the line-of-sight angle θ exceeds the threshold value x is greater than F seconds, whereas the movement of the driver's line of sight is As shown in FIG. 9B, the total time tn was F ′ seconds smaller than F seconds. In this case, the line-of-sight pattern learning unit 32 updates the line-of-sight pattern shown in FIG. 9A using the detected movement of the line of sight of the driver (see FIG. 9B). Specifically, for example, an update method of updating the time of F seconds used for determination of the total time to F ′ seconds or updating to (F + F ′) / 2 seconds is conceivable. The navigation device 3 according to the third embodiment has been described above.

  The navigation device 3 according to the third embodiment performs learning in the line-of-sight pattern learning unit 32 and updates the information stored in the line-of-sight pattern storage unit 20. The line-of-sight pattern when gazing at the object can be adapted to each driver.

  The navigation device of the present invention has been described in detail with reference to the embodiment. However, the navigation device of the present invention is not limited to the above-described embodiment.

  Although the navigation apparatus and the navigation method have been described in the above-described embodiment, the present invention includes a program for executing each step of the navigation method of the above-described embodiment.

  In the above-described embodiment, the example in which the driver's line of sight is detected from the video captured by the camera 10 has been described as the configuration for detecting the driver's line of sight. For example, the driver's line of sight is detected using the eye mark camera 10. May be.

  In the above-described embodiment, when the driver's gaze behavior matches one of the gaze behavior patterns stored in the gaze behavior pattern storage unit 22, the predicted driving behavior determination unit 16 corresponds to the gaze behavior pattern. Although the example of determining the driving behavior as the predicted driving behavior has been described, the predicted driving behavior may be determined even when the driver's gaze behavior does not completely match the gaze behavior pattern. For example, when the driver's gaze behavior matches a part of the gaze behavior pattern, the driving behavior corresponding to the gaze behavior pattern may be determined as the predicted driving behavior. That is, for example, when a gaze action of a gaze object A → gaze object B → gaze object C → gaze object B is detected, the gaze object A → gaze object B → gaze object C → gaze object Since it matches a part of the gaze behavior pattern of the object B → the gaze object D, the driving behavior associated with the gaze behavior pattern can be set as the predicted driving behavior. As a result, the predicted driving behavior can be determined before detecting the gaze of the gaze target D, so that driving assistance can be performed at an earlier stage.

  Note that when this configuration is adopted, the driver's gaze behavior may match some of the plurality of gaze behavior patterns, but based on the weighting parameters described in the second embodiment described above, One gaze behavior pattern may be selected from a plurality of gaze behavior patterns.

  In the above-described embodiment, an example in which a plurality of line-of-sight patterns different according to the vehicle speed is stored in the line-of-sight pattern storage unit 20 is described. However, a plurality of line-of-sight patterns are stored according to environmental parameters other than the vehicle speed. Also good. As environmental parameters, for example, parameters such as the years of driving experience of the driver and whether the driving time is day or night can be used.

  In the above-described embodiment, an example of changing the line-of-sight pattern for detecting a gaze target according to the environment has been described. However, depending on the environment, the gaze action pattern used for obtaining the predicted driving action is determined. It may be changed. That is, the navigation device stores a plurality of gaze behavior patterns that are different for each environment in association with the driving behavior in the gaze behavior pattern storage unit. Then, the navigation device selects a gaze behavior pattern corresponding to the environment from the gaze behavior patterns stored in the gaze behavior pattern storage unit, and determines the predicted driving behavior using the selected gaze behavior pattern.

  The present invention has an effect that, by detecting a gaze object based on the movement of the line of sight, the driver can appropriately detect the gaze object even if the driver does not look at the object for a certain period of time. It is useful as a navigation device or the like having a function of performing driving support based on the information of the object being watched.

It is a figure which shows the structure of the navigation apparatus of 1st Embodiment. It is a figure which shows the example of the information memorize | stored in the gaze pattern memory | storage part. It is a figure which shows in detail the example of the data memorize | stored in the gaze pattern memory | storage part. It is a figure which shows the example of the information memorize | stored in the gaze action pattern memory | storage part. It is a figure which shows the operation | movement of the driving assistance by the navigation apparatus 1 of this Embodiment. It is a figure which shows the structure of the navigation apparatus of 2nd Embodiment. (A) It is a figure which shows what is called a "teacher signal" used for learning of a gaze action pattern. (B) It is a figure which shows the example of the information memorize | stored in a gaze action pattern memory | storage part. It is a figure which shows the structure of the navigation apparatus of 3rd Embodiment. (A) It is a figure which shows the example of the gaze pattern memorize | stored in the gaze pattern memory | storage part. (B) It is a figure which shows the example of the motion of the calculated | required eyes | visual_axis.

Explanation of symbols

1-3 Navigation device 10 Camera 12 Gaze detection unit 14 Gaze target detection unit 16 Predictive driving behavior determination unit 18 Driving assistance signal transmission unit 20 Gaze pattern storage unit 22 Gaze behavior pattern storage unit 24 Vehicle speed information acquisition unit 26 Vehicle information acquisition unit 28 Driving Behavior Detection Unit 30 Gaze Behavior Learning Unit 32 Gaze Pattern Learning Unit

Claims (9)

A line-of-sight pattern storage unit that stores a pattern of movement of the line of sight of the driver when gazing at an object inside and outside the vehicle as a line-of-sight pattern and stored in association with the object;
A line-of-sight detector that detects the line of sight of the driver;
Based on the line-of-sight information detected by the line-of-sight detection unit, the movement of the line of sight of the driver is obtained, the object corresponding to the line-of-sight pattern matching the movement of the line of sight is read from the line-of-sight pattern storage unit, and the read object A gaze target detection unit that detects the gaze target;
A navigation device comprising: The line-of-sight pattern storage unit stores a plurality of line-of-sight patterns different for each environment in association with the object,
The navigation device according to claim 1, wherein the gaze target detection unit selects the gaze pattern according to an environment for which the movement of the gaze is obtained, and detects the gaze target using the selected gaze pattern. A gaze behavior pattern storage unit that stores a plurality of gaze objects to be watched prior to the driver's driving behavior and the gaze order in association with the driving behavior as a gaze behavior pattern;
The gaze behavior is obtained based on the gaze target information detected by the gaze target detection unit, and the driving behavior corresponding to the gaze behavior pattern that matches the gaze behavior is read from the gaze behavior pattern storage unit and read out. A predictive driving action determination unit that determines the action as a predictive driving action performed following the gaze action;
A driving support signal transmitter for transmitting a driving support signal for supporting the predicted driving behavior;
A navigation device according to claim 1 or 2. The gaze behavior pattern storage unit stores a plurality of gaze behavior patterns that are different for each environment in association with the driving behavior,
The navigation device according to claim 3, wherein the predicted driving behavior determination unit selects the gaze behavior pattern according to an environment in which the gaze behavior is detected, and determines the predicted driving behavior using the selected gaze behavior pattern. .   The predictive driving behavior determination unit obtains a gaze behavior from information on the gaze target detected by the gaze target detection unit, and the gaze behavior is part of a gaze behavior pattern stored in the gaze behavior pattern storage unit. The navigation device according to claim 4, wherein when the two match, the driving behavior corresponding to the gaze behavior pattern is read out. A driving behavior detector for detecting driving behavior from the movement of the vehicle;
When the driving behavior is detected by the driving behavior detection unit, the gaze behavior performed before the driving behavior is obtained and the information stored in the gaze behavior pattern storage unit is updated using the gaze behavior. A behavior pattern learning unit;
A navigation device according to any one of claims 3 to 5. Based on the line-of-sight information detected by the line-of-sight detection unit, the movement of the driver's line of sight is obtained, the line-of-sight pattern storage unit that searches for the line-of-sight pattern most similar to the line-of-sight movement is searched for, A line-of-sight pattern learning unit that updates using the line-of-sight movement,
A navigation device according to any one of claims 1 to 6. Preparing a line-of-sight pattern storage unit that stores a pattern of movement of the line of sight of a driver when gazing at an object inside and outside the vehicle as a line-of-sight pattern and associated with the object;
Detecting the driver's line of sight;
Determining the movement of the driver's line of sight based on the detected line of sight information;
Reading an object corresponding to a line-of-sight pattern matching the movement of the line of sight from the line-of-sight pattern storage unit, and detecting the read object as a gaze object;
A navigation method comprising: In a navigation device with a function to detect the driver's line of sight,
Detecting the driver's line of sight;
Determining the movement of the driver's line of sight based on the detected line of sight information;
The line-of-sight pattern storage unit that stores an object corresponding to the line-of-sight pattern that matches the line-of-sight movement as a line-of-sight pattern of a driver's line-of-sight movement when gazing at an object inside or outside the vehicle. And detecting the read object as a gaze object,
A program that executes

Images