DE102012221036A1 - Method for automatic switching on of parking- and/or maneuvering assistance systems e.g. park distance control system in motor car, involves switching on parking- and/or maneuvering assistance systems based on evaluation of driver behavior - Google Patents

Abstract

The method involves determining a view behavior of a driver (100), and evaluating the view behavior. Parking- and/or maneuvering assistance systems are switched-on (160) based on the evaluation of the view behavior. Parking- and/or maneuvering intentions of a driver are recognized based on the behavior of the driver. A probability is intended for the parking- and/or ranking intentions based on the behavior, and a threshold value of probability is compared (150) with a threshold value. A determined view behavior is compared with deposited profiles of the view behavior by a correlation process.

Description

The invention relates to a method for automatically switching on a parking and / or maneuvering assistance system in a motor vehicle.

Various parking and / or maneuvering assistance systems of driver assistance systems are known, which support the driver during parking and / or maneuvering operations.

For example, in a simple PDC system (Park Distance Control) the distance to objects and obstacles is typically determined by means of an ultrasonic sensor system and the driver is visually and / or acoustically informed of the distance. In a reversing camera system, the environment behind the vehicle is recorded by means of a camera and the camera image is displayed to the driver. In a top-view system, the image of several cameras in different directions is combined to a bird's-eye view and displayed to the driver.

In semi-automatic or fully automatic parking assistance systems with parking space measurement, longitudinal and / or transverse parking spaces are measured while driving past below or up to a certain limit speed (for example 40 km / h), without the parking assistance system typically having to be switched on for this. After switching on the parking assistance system by the driver, the driver is informed whether the respective parking space is suitable because of their dimensions as a parking lot. In parking assistance systems with automatic lateral guidance, the steering of the vehicle is taken over by the system during the parking process. The driver must accelerate during the parking process and brakes (longitudinal guidance). In parking assistance systems with automatic lateral and longitudinal guidance, this task is also taken over by the parking assistant.

The switching on of the parking and / or maneuvering assistance system is typically carried out by an active driver's action. In a PDC system, for example, the system is manually turned on by pressing a specific operation button or by engaging reverse gear while the engine is running. A semi-automatic or fully automatic parking assistance system with parking space measurement is turned on, for example, by pressing a specific button or engaging a reverse gear, whereupon the user the current status of the parking space search is displayed.

Furthermore, it may be that, in addition to the active driver's action, one or more different boundary conditions must be fulfilled in order for the assistance system to be switched on; for example, the speed of the own vehicle must be less than or equal to a certain speed threshold in order to limit the assistance function to a certain speed range.

The disadvantage of the manual operation by the driver via a control button is that the driver can activate the assistance system only by additional effort. He must independently press the control button. As a result, the comfort is impaired because the driver has to perform the switching on of the assistance system independently by searching and pressing the control button. In addition, the driver loses time by manually switching on and may not find the control button in stressful situations.

Furthermore, it is known to automatically (ie without dedicated driver action) automatically turn on an ultrasound-based PDC system without active driver action in the case of cumulative presence of a plurality of boundary conditions: for example, if a speed below a defined speed threshold (eg below 10 km / h) is present certain gear stage (eg forward gear), presence of a certain engine condition (eg engine started) and falling below a threshold value (eg 1 m) for the distance value to an obstacle determined by means of ultrasonic sensors.

In such methods for automatically switching on the PDC system can lead to false triggering, whereby a system switching can occur outside of parking and maneuvering situations. For example, most known systems are designed for rather lower speeds up to, for example, 10 km / h, whereby the PDC system automatically switches on up to this threshold value and the presence of the other boundary conditions and becomes acoustically and / or optically active. An automatic switch-on is thus for example also in a traffic jam situation, slow-moving traffic or when approaching a stationary vehicle at a traffic light. This is perceived by the driver as disturbing and can lead to irritation of the driver who then has to switch off the parking and / or maneuvering manually.

At the same time, reducing the speed threshold or distance threshold to avoid false tripping outside of parking and / or maneuvering situations reduces the benefit of automatically turning on the assistance system, as parking and maneuvering situations may often occur in which the system is not automatically turned on. This is disadvantageous because the driver automatically turns on the system hoped for and after the failure to turn on the assistance system manually.

Typically, the parking space measurement of parking assistance systems takes place up to a speed of about 40 km / h without the parking assistance system has to be turned on for this purpose. Methods for automatically switching on a parking assistance system are limited due to the above reasons to low vehicle speeds of, for example, a maximum of 10 km / h. Thus, the driver must manually engage the park assist system above this maximum vehicle speed of, for example, 10 km / h (eg, via the control button), even if the parking assist system has detected a suitable parking space at significantly higher speeds.

The publication DE 10 2006 058 213 A1 describes a vehicle with a parking assistance system which is arranged such that the parking assistance system is automatically activated as soon as the measured vehicle speed is within a predetermined speed range.

From the publication DE 10 2006 052 575 A1 It is known to activate the automatic parking mode of a Parklenkassistenten when the driver performs an action that always performs the driver in a parking operation, such as the operation of a direction selection element.

In the publication DE 10 2006 032 171 A1 an automatically activatable distance warning system is described, wherein the steering angle of the motor vehicle is detected and when a predetermined steering angle is exceeded below a certain vehicle speed, the feedback for the transmission of distance information on the front vehicle surrounding area is activated.

Furthermore, in the document DE 10 2008 040 149 A1 a release of an automatic guidance by a semi or fully automatic parking assistance system described. Here, the field of view of the driver is compared with the direction of travel in the automatic guidance and the automatic guidance released only when the direction of travel of the vehicle is located in the automatic guidance in the field of view of the driver. However, the driver's field of view evaluation is not used to turn on the park assist system. Thus, after an initialization step, the detection of a parking space is started and a parking trajectory is calculated. Subsequently, the driver is informed that now takes place an automatic guidance of the vehicle. After manual confirmation, the automatic parking procedure is started. Here, it is first checked whether the destination has been reached, and only then checked whether the route of the vehicle is in the field of vision of the driver. If this is the case, the vehicle is guided along the calculated route.

It is an object of the invention to provide an improved method for automatically switching on a parking assistance system.

The object is solved by the features of independent claim 1. Advantageous embodiments are described in the dependent claims.

In the method according to the invention for automatically switching on a parking and / or maneuvering assistance system (for example a semi-automatic or fully automatic parking assistance system with parking space measurement or a PDC system) in a motor vehicle, the driver's gaze behavior is determined, for example by means of an interior camera. The gaze behavior is evaluated; In the evaluation of the gaze behavior, clear gaze characteristics with respect to a parking and / or shunting intension can be detected. Depending on the evaluation of the gaze behavior, the parking and / or maneuvering assistance system is switched on. In addition to the evaluation of the gaze behavior, one or more further influencing factors, such as the vehicle speed, are optional for the decision to switch on the assistance system.

As part of the evaluation of the gaze behavior, a parking and / or maneuvering intention of the driver is preferably detected based on the driver's gaze behavior. Depending on the parking and / or maneuvering intent detected, the parking and / or maneuvering assistance system is automatically switched on. If, for example, a parking and / or maneuvering intension has been detected, the parking and / or maneuvering assistance system is automatically switched on. If necessary, to switch on the assistance system, in addition to the parking and / or maneuvering intension, the presence of one or more further prerequisites is checked, for example the presence of a sufficiently low vehicle speed.

By evaluating the gaze behavior, in particular detection of parking and / or maneuvering based on gaze behavior, an automatic switching on of a parking and / or maneuvering assistance system can also be made possible at higher speeds without dedicated action on the part of the driver (eg automatic activation of a display via recognized, suitable parking spaces when driving past parking spaces without driver's action). Furthermore, by evaluating the gaze behavior, the probability of a disturbing false triggering of the parking and / or maneuvering assistance system can be reduced in certain traffic situations such as congestion or when approaching a vehicle standing at a traffic light.

The invention is based on the recognition that, based on the gaze behavior of the driver alone or in conjunction with other parameters, it can be established that the driver wants to park or maneuver, and therefore the parking and / or maneuvering assistance system is expediently automatically switched on to assist the driver should. So it is from the dissertation "System ergonomic analysis of display and operating concepts to support the parking process", Christian Doisl, Department of Ergonomics of the Technical University of Munich, 22.11.2007 It is known that the gaze behavior of a test person when looking for a parking space differs significantly from the gaze behavior when driving freely (see in particular chapter 3.2.3). There it is described that the view avertion from the street escape point is much more pronounced when looking for a parking space than when driving freely. Thus, the eye relief time per eye-sighting when looking for a parking space is significantly higher both in terms of the average and in terms of the maximum value than in the case of free travel. Also, the percentage Blickabwendungszeit from the street escape point is significantly higher in a search for a parking space in terms of both the average and in terms of the maximum value than the free ride. The disclosure content contained in the dissertation for distinguishing the gaze behavior in a search for a parking space in comparison to the gaze behavior when driving freely is hereby incorporated by reference into the disclosure content of this application.

On the basis of this, in the evaluation of the gaze behavior according to the invention, one or more parameters can be determined which relate to the gaze averted from the pathway ahead of the vehicle (and / or the gaze attraction to one or more surrounding areas outside this travel path), for example the duration of viewing giveaways (FIG. Fixations) outside the road ahead and / or their frequency.

By means of an evaluation of the gaze behavior of the driver, not only the switching on of a parking and / or maneuvering assistance system can be controlled, but also the deactivation of an already activated parking and / or maneuvering assistance system can be prevented, if, for example, it is recognized from the evaluation of the gaze behavior that the Driver still has a parking or Rangierintention.

By means of an interior camera system, it is possible to capture glances and hints of the driver over a certain period of time and thus to make a statement as to whether the driver is looking for a parking space or not. If, for example, the head movement of the vehicle driver is markedly left or right over a certain period of time, the presence of a parking intention can be inferred.

The parking and / or maneuvering assistance system is turned on, for example, when the one or more characteristics indicate that gazing away from the forward pathway and / or viewing gaze from one or more vehicle surrounding areas outside of that driving path exceeds a certain level.

By way of example, a probability for a parking and / or maneuvering intention of the driver is determined by evaluating the gaze behavior, and a threshold value comparison of the probability with a threshold value is carried out. If the probability exceeds the threshold value, for example, it is assumed that the driver is parked and / or shunting.

By means of a vehicle interior camera system and additional image processing, the field of vision of the driver can be determined. The field of view of the driver can be assigned to predefined areas in the vehicle (for example the right-hand area of the windscreen or the side window on the passenger side) and / or to objects detected in the vehicle environment (for example a parked vehicle or a parking space between two vehicles) , One or more parameters can then be determined for each area or object, each describing the view to this area or object. By way of example, the duration of the viewing contribution per area or object and / or the frequency of use of the view per area or object (for example over a predefined time interval Δt) can be determined as parameters.

For the areas or objects, area-related or object-related probability values for (or alternatively against) a parking and / or shunting intention of the driver can be stored. On the basis of the stored probability values, an individual probability for a parking and / or shunting intention can then be determined for each characteristic of a region or object by linking the respective probability value with the characteristic of the region or object. Based on the individual probabilities can then determine a probability of parking and / or Rangierintention.

Alternatively or additionally, the gaze behavior can be compared with one or more stored profiles of the gaze behavior (eg for a free straight-ahead ride without parkintension or for a ride-in at park intension), in particular by means of a correlation method. For example, a probability for a parking and / or maneuvering intension can also be determined from this.

In addition to the gaze behavior vehicle data can be used to detect a parking and / or maneuvering, for example, the vehicle speed or engine speed, the power window operation (for example when entering a parking garage), the mirror settings (for example, to detect the curb), the operation of the service brake , the engagement of a certain gear ratio or steering information (eg steering wheel angle, steering wheel angular velocity, steering wheel angle acceleration). Depending on information about the position and movement of the steering wheel can be concluded on a parking and maneuvering situation. As a rule, small steering wheel movements are made during a straight-ahead drive, whereas parking and maneuvering situations are significantly larger.

In addition to or alternatively to the gaze behavior, the location of the vehicle can also enter into the decision about the automatic switching on of the assistance system. For example, automatically switching on the parking assistant at certain, recurring locations without operator intervention makes sense (eg in front of one's own home, at the daily workplace, in a parking garage or in a parking lot).

The invention will be described below with reference to the accompanying drawings with reference to an embodiment.

1 shows a flowchart for an embodiment of the inventive method for switching on a parking and / or maneuvering assistance system.

In the embodiment described below, the driver's gaze behavior is measured, for example via the position, orientation and movement of the eyes and the head. By evaluating this data by means of a processing unit, a parking and / or Rangierinttension the driver can be detected. If a parking and / or maneuvering intension has been detected by the processing unit, a parking and / or maneuvering assistance system is switched on visibly for the driver.

The driver's gaze is detected by a gaze sensor system (see reference numeral 100 in 1 ). The gaze sensor system typically includes one or more driver-directed interior cameras in the vehicle interior (eg, in the instrument panel of the vehicle) for visual field recognition; this is preferably a 3-D camera system. The camera system can be used, for example, not only in the context of switching on the parking and / or maneuvering assistance system, but also for other functions, such as a fatigue warning system or to determine the seat occupancy and the head position of the passenger in the context of the control of airbag systems.

Furthermore, image processing is provided as part of the gaze sensor system, which determines, for example, the position, orientation and movement of the head and eyes of the driver in 3-dimensional space. The measurement data acquired by the gaze sensor system include, for example, the position (x _K , y _K , z _K ) and orientation (α _K , β _K , γ _K ) of the head and the position (x _A , y _A , z _A ) and orientation (α _A , β _A , γ _A ) of the eyes.

By means of an environment sensor system, the environment of the vehicle is detected (see reference numeral 110 in 1 ). Based on sensor information of one or more environment sensors (such as ultrasound sensors, radar sensors, satellite cameras, front camera, laser scanner and / or lidar sensor), an image of the environment (for example an environment map) is created relative to the vehicle (see preprocessing 120 ). In this case, a free space representation of the vehicle environment can be generated, whereby a distinction is made between free space and objects or obstacles in the environment. Also, a classification of objects in the vehicle environment can be performed. In addition to the position, extent and contour, the objects are assigned further attributes such as object type (vehicle, curb, person or similar) and object height (division into traversable and non-traversable or specification of a specific height). Downstream functions use this information to determine the distances between the ego vehicle and the objects and to detect parking spaces

Based on the measured data acquired by the gaze sensor system, the preprocessing takes place 130 a further processing of this data. In preprocessing 130 For example, the movement speeds (for example, by differentiating the position data) and the movement accelerations (for example, by differentiating the position data twice) and the rotation speeds by differentiating the orientation data, for example, and spins (for example, by differentiating the orientation data twice) are determined by the head and eyes.

Furthermore, in preprocessing 130 a determination of the current field of vision of the driver. To determine the driver's field of view, defined values are used for the vertical and horizontal field of view of a person; Here, the foveal field of view can be determined; optionally, a distinction can be made between the foveal field of view and the peripheral field of view.

On the basis of the driver's gaze behavior, fixations with a minimum fixation time Δt _F and saccades are determined. Saccades are fast eye movements that move the eyes from one object to another without taking visual information.

Subsequently, in the processing 140 The assignment of the foveal (and possibly peripheral) field of view of the driver to predefined areas of the vehicle and to objects in the vicinity of the vehicle.

Examples of areas of the vehicle are the right side, lateral area of the windscreen, the left, lateral area of the windscreen, the middle area of the windscreen (in the various areas of the windscreen can optionally be distinguished between top and bottom), the rear window, the side window on the driver's side, the side window on the passenger side, the speedometer, the central information display, the area of the passenger and the rear-view mirror. The areas may be variable and depending, for example, on the seat height of the driver.

Examples of objects in the environment are a parked vehicle or a parking space between two vehicles.

Different domain-specific probability values are stored for each of the differentiated regions in the vehicle, such a region-specific probability value representing a probability measure for the presence of a parking and / or maneuvering intension when looking at this region. Here, for example, the probability values for the area of the side window on the front passenger side and the right side area of the windscreen are higher than the probability value for the left side area of the windscreen (ie the area of the windscreen directly in front of the driver) The first two areas are typically associated with a higher probability of park and / or Rangierintention as the Blickzuwendung to the windscreen. When looking at either of the first two areas, the gaze is averted from the track ahead and instead directed at parking areas to the side of the forward driving area, which indicates a parking intention of the driver.

For objects in the environment, object-specific probabilities are calculated, such an object-specific probability value representing a probability measure for the presence of a parking and / or shunting intension when looking at this object. This is u. a. depending on the attributes of the respective object. In this case, for example, the probability value for an object which is located in front of the ego vehicle is smaller than for an object laterally of the ego vehicle. For example, the probability value for a detected gap between two vehicles whose length is greater than a defined minimum length is higher than the probability value for a number of parked vehicles without a gap having a length greater than this minimum length. The calculation of the object-specific probabilities depends on the environment sensors installed in the vehicle.

In the processing 140 various parameters P _{BF, F are} determined, which relate to the viewing of certain areas in the vehicle and objects in the vehicle environment.

• 1. die ermittelte Dauer der Blickzuwendung/Fixation je Bereich bzw. Objekt innerhalb einer Zeitspanne;
• 2. die Häufigkeit der Blickzuwendung je Bereich bzw. Objekt innerhalb einer Zeitspanne;
• 3. Ermittlung der Bereiche bzw. Objekte, welche zwischen zwei aufeinanderfolgenden Fixationen vom Blickfeld überstrichen wurden sowie die Auswertung der Abfolge der Bereiche (z. B. Frontscheibe Mitte, Frontscheibe rechte Seite, Seitenscheibe Beifahrer);
• 4. die Kopf- und/oder Blickbewegungsgeschwindigkeit/-beschleunigung bei der Blickzuwendung zu einem Bereich bzw. Objekt.

The parameters P _{BF, F} include, for example, one, several or all of the following examples:

• 1. the determined duration of gaze / fixation per area or object within a period of time;
• 2. the frequency of viewing per area or object within a time span;
• 3. Determination of the areas or objects which were swept from the field of view between two successive fixations as well as the evaluation of the sequence of the areas (eg front screen center, front screen right side, side pane front passenger);
• 4. the head and / or eye movement speed / acceleration when looking at an area or object.

Taking into account the area-specific and object-specific probabilities, a probability for a park intension W _{PI, 1} (or for a maneuvering intension W _{RI, 1} ) is determined on the basis of the areas and objects considered by the driver and the parameters evaluated for this purpose. For example, Bayesian networks, artificial neural networks or the evidence theory of Dempster and Shafer are used.

In addition, the driver's gaze behavior with stored profiles (eg for straight-ahead driving, for parking space search, during a parking process, during a maneuvering in-stallation, during a maneuvering process, in a lane change, in a turn) are compared by means of correlation methods. From this, a probability for a park intension W _{PI, 2} (or for a Rangierintension W _{RI, 2} ) can also be determined. A high degree of agreement with the profiles of the parking space search or the parking operation (or the maneuvering and shunting operations) leads to a correspondingly high probability W _{PI, 2} (or W _{RI, 2} ). A high agreement with another profile leads in turn to a low probability W _{PI, 2} (or W _{RI, 2} ).

The probabilities W _{PI, 1} , W _{PI, 2} (or W _{RI, 1} , W _{PI, 2} ) are compared with predefined, associated threshold values W _{PI, 1, S} , W _{PI, 2, S} (or W _{RI, 1 , S} , W _{PI, 2, S} ) in step 150 compared.

If the probability W _{PI, 1} exceeds the threshold value W _{PI, 1, S} or if the probability W _{PI, 2} exceeds the threshold value W _{PI, 2, S} , it is assumed that there is a parking intention of the driver. In this case, the processing unit automatically switches on the parking assistance system (see step 160 ), for example, a semi-automatic or fully automatic parking assistance systems with parking space measurement. The driver is informed of the presence of suitable parking spaces after the automatic switch-on. Alternatively, the automatic switching on of a parking assistance system takes place in the presence of a park intension of the driver, provided that a parking space suitable for the vehicle exists, wherein the distance between the ego vehicle and the parking space is less than a threshold value.

Otherwise, the method described above is continued without switching on the parking assistance system.

If the probability W _{RI, 1} exceeds the threshold value W _{RI, 1, S} or if the probability W _{RI, 2} exceeds the threshold value W _{RI, 2, S} , the driver assumes a shunting intension. In this case, the processing unit automatically switches on a maneuvering assistance system.

• 1. Die Präsenz eines Beifahrers oder weiterer Insassen sowie deren Sitzhöhe haben Auswirkungen auf das Blickverhalten. Wird die Präsenz eines Beifahrers oder weiterer Insassen von der Verarbeitungseinheit detektiert (z. B. durch Auswertung der Türkontakte oder der Sitzbelegung), werden die hinterlegten bereichsspezifischen Wahrscheinlichkeiten auf Basis von hinterlegten Werten oder zusätzlich auf Basis der Sitzhöhe der Insassen angepasst (z. B. der Wahrscheinlichkeitswert für den Bereich der Seitenscheibe auf der Beifahrerseite).
• 2. Der Lichteinfall (z. B. über wetter-, standort-, datum- und/oder uhrzeitabhängigem Sonnenstand) wird verwendet, um die die hinterlegten bereichsspezifischen Wahrscheinlichkeiten hinsichtlich des Blickverhaltens des Fahrers anzupassen.
• 3. Das vorstehend beschriebene Verfahren arbeitet mit einem vordefinierten Satz von Wahrscheinlichkeitswerten und Schwellwerten. Das Verfahren kann aber auch lernfähig ausgelegt werden und fahrerabhängig die hinterlegten Wahrscheinlichkeitsschwellen, Übergangswahrscheinlichkeiten und Korrelationsparameter bzw. Profile anpassen. Das Anpassen geschieht beispielsweise dann, wenn ein eindeutiger Park- oder Rangiervorgang vom Fahrer durchgeführt wird (z. B. durch Drücken vom Parktaster oder Einlegen vom Rückwärtsgang), obwohl das System keine Park- bzw. Rangierintension ermittelt hat.
• 4. Es werden ein oder mehrere Geschwindigkeitsschwellen für die Auswertung des Blickverhaltens hinterlegt, beispielsweise findet eine Auswertung hinsichtlich einer Parkintension ausschließlich unterhalb der Geschwindigkeitsschwelle v_SCHWELLE,PI statt (bzw. die Auswertung hinsichtlich einer Rangierintension unterhalb der Geschwindigkeitsschwelle v_SCHWELLE,RI).
• 5. Eine optionale Auswertung des Lenkwinkelverlaufs sowie des Geschwindigkeitsverlaufs führen zu weiteren Parametern, die zusätzlich zu den Parametern P_BF,F berücksichtigt werden können. Beispielsweise führt eine hohe Änderung des Lenkwinkels über dem Zeitintervall Δt_LW,RI zu einer hohen Wahrscheinlichkeit W_PI,1 für eine Rangierintension. Eine Verzögerung der Geschwindigkeit größer als a_SCHWELLE,RI unterhalb der Geschwindigkeitsschwelle v_SCHWELLE,RI führt zu einer hohen Wahrscheinlichkeit W_RI,1 für eine Rangierintension.
• 6. Es können zu den einzelnen Blickverhaltensprofilen (wie Geradeausfahrt, Parkplatzsuche, Parkvorgang, Rangierintension, Rangiervorgang, Spurwechsel, Abbiegevorgang) zugehörige Lenkwinkel- und Geschwindigkeitsprofile hinterlegt werden. Diese werden mit Hilfe von Korrelationsverfahren mit den aktuellen Verläufen des Lenkwinkels bzw. der Geschwindigkeit verglichen und fließen in die zugehörigen Wahrscheinlichkeiten W_PI,2 (bzw. W_RI,2) ein.
• 7. Eine automatische Deaktivierung von anderen Funktionen kann bei der Aktivierung des Park- und/oder Rangierassistenzsystems einhergehen.
• 8. Es kann eine automatische Deaktivierung des Park- und/oder Rangierassistenzsystems verhindert werden (z. B. bei Überschreitung einer definierten Geschwindigkeitsschwelle zur automatischen Abschaltung), sofern das System das Vorliegen einer Park- oder Rangierintension erkannt hat.

Optionally, the following extensions of the method described above can be provided:

• 1. The presence of a passenger or other inmates and their seat height have an effect on the gaze behavior. If the presence of a co-driver or other occupants is detected by the processing unit (eg by evaluating the door contacts or seat occupancy), the stored area-specific probabilities are adjusted on the basis of stored values or additionally based on the seat height of the occupants (eg. the probability value for the area of the side window on the passenger side).
• 2. The incidence of light (eg on the basis of weather, location, date and / or time-dependent sun position) is used to adjust the stored area-specific probabilities regarding the driver's gaze behavior.
• 3. The method described above uses a predefined set of probability values and thresholds. However, the method can also be designed to be adaptive and, depending on the driver, to adapt the stored probability thresholds, transition probabilities and correlation parameters or profiles. Adjustment occurs, for example, when a clear parking or maneuvering operation is performed by the driver (eg, by pressing the park button or reverse gear), even though the system has not determined a parking or maneuvering intension.
• 4. One or more speed thresholds for the evaluation of the gaze behavior are stored; for example, an evaluation takes place with respect to a parking increment exclusively below the speed _threshold v _{THRESHOLD, PI} (or the evaluation with regard to a shunting input below the speed _threshold v _{THRESHOLD, RI} ).
• 5. An optional evaluation of the steering angle course as well as the speed course leads to further parameters which can be taken into account in addition to the parameters P _{BF, F.} For example, a high change in the steering angle over the time interval .DELTA.t _{LW, RI} leads to a high probability W _{PI, 1} for a Rangierintension. A delay of the speed greater than a _{THRESHOLD, RI} below the speed _threshold v _{THRESH, RI} leads to a high probability W _{RI, 1} for a shunting intension.
• 6. It can be stored to the individual Blickverhaltensprofilen (such as driving straight ahead, parking space search, parking, Rangierintension, maneuvering, lane change, turn) associated steering angle and speed profiles. These are compared by means of correlation methods with the current progressions of the steering angle or the speed and flow into the associated probabilities W _{PI, 2} (or W _{RI, 2} ).
• 7. Automatic deactivation of other functions may be accompanied by the activation of the parking and / or maneuvering assistance system.
• 8. An automatic deactivation of the parking and / or maneuvering assistance system can be prevented (eg if a defined speed threshold for automatic shutdown is exceeded), provided that the system has detected the presence of a parking or maneuvering intension.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102006058213 A1 [0012]
• DE 102006052575 A1 [0013]
• DE 102006032171 A1 [0014]
• DE 102008040149 A1 [0015]

Cited non-patent literature

• "System Ergonomic Analysis of Display and Operating Concepts to Support the Parking Process", Christian Doisl, Department of Ergonomics, Technische Universität München, 22.11.2007 [0021]

Claims (10)

Method for automatically switching on a parking and / or maneuvering assistance system in a motor vehicle, comprising the steps of: - determining the gaze behavior of the driver; - evaluating the gaze behavior; and - Automatic switching on the parking and / or maneuvering assistance system depending on the evaluation of the gaze behavior. The method of claim 1, wherein the step of evaluating gaze behavior comprises: Detecting a parking and / or Rangierentention the driver based on the gaze behavior of the driver; and the parking and / or maneuvering assistance system is automatically switched on depending on the detected parking and / or maneuvering intention. The method of claim 2, wherein based on the gaze behavior of the driver, a probability for parking and / or Rangierentention the driver is determined and a threshold value comparison of the probability is performed with a threshold value. Method according to one of the preceding claims, wherein the step of evaluating the gaze behavior comprises: - Determining one or more characteristics that The gaze away from the road ahead of the vehicle and / or The view reference to one or more vehicle surroundings areas lying outside this travel path, in particular a parking area, affect. 4. The method of claim 4, wherein the parking and / or maneuvering assistance system is turned on when the one or more characteristics indicate that gazing away from the forward pathway and / or gazing beyond one or more vehicle surrounding areas exceeds a certain extent , Method according to one of the preceding claims, wherein the step of evaluating the gaze behavior comprises: - Determining one or more characteristics that relate to the viewing of certain areas in the vehicle or objects in the vehicle environment. The method of claim 6, wherein the one or more characteristics include: - the duration of the viewing contribution per area or object or - the frequency of viewing per area or object. Method according to one of the preceding claims, wherein the step of evaluating the gaze behavior comprises: - Compare the determined gaze behavior with one or more stored profiles of the gaze behavior, in particular by means of a correlation method. Method according to one of the preceding claims, wherein in each case an object-specific probability value is determined for detected objects in the vehicle environment, wherein such an object-specific probability value represents a probability measure for the presence of a parking and / or Rangierintension when Blickzuwendung to this object. Method according to one of the preceding claims, wherein area-specific probability values are stored for areas in the vehicle, such a area-specific probability value representing a probability measure for the presence of a parking and / or maneuvering intension when looking at this area.

Images