Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD FOR AT LEAST SEMI-AUTONOMOUSLY MANEUVRING A MOTOR VEHICLE INTO A GARAGE BASED ON ULTRASONIC SIGNALS AND CAMERA IMAGES, DRIVER ASSISTANCE SYSTEM AS WELL AS MOTOR VEHICLE
Document Type and Number:
WIPO Patent Application WO/2018/024585
Kind Code:
A1
Abstract:
The invention relates to a method for at least semi-autonomously maneuvering a motor vehicle (1) into a garage (10), in which ultrasonic signals are emitted by an ultrasonic sensor (4) of the motor vehicle (1) and ultrasonic signals reflected off objects (8) in an environment (7) of the motor vehicle (1) are received while the motor vehicle (1) is maneuvered along a drive tube (12) to the garage (10), and the objects (8) in the environment (7) are identified based on the receives ultrasonic signals and images (13) are provided by a camera (9) of the motor vehicle (1) during maneuvering, which describe the environment (7), wherein if at least one of the objects (8) is identified as an obstacle in the drive tube (12) based on the received ultrasonic signals, it is recognized based on at least one of the images (13), which describes an area in the environment (7) associated with the identified obstacle, if the received ultrasonic signals have been reflected off a component of the garage (10).

Inventors:
HEIMBERGER, Marco (Laiernstr. 12, Bietigheim-Bissingen, 74321, DE)
FRIEL, Myles (Dunmore Road, Tuam County Galway, -, -, IE)
Application Number:
EP2017/068952
Publication Date:
February 08, 2018
Filing Date:
July 27, 2017
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
CONNAUGHT ELECTRONICS LTD. (Dunmore Road, Tuam, County Galway, IE)
International Classes:
G06K9/00; B60W30/06; B62D15/02
Foreign References:
DE102012022336A12014-05-15
DE102010023162A12011-12-15
DE102011084586A12013-04-18
DE102009057837A12011-06-16
DE102012200625A12013-07-18
Attorney, Agent or Firm:
JAUREGUI URBAHN, Kristian (Laiernstr. 12, Bietigheim-Bissingen, 74321, DE)
Download PDF:
Claims:
Claims

1 . Method for at least semi-autonomously maneuvering a motor vehicle (1 ) into a garage (10), in which ultrasonic signals are emitted by an ultrasonic sensor (4) of the motor vehicle (1 ) and ultrasonic signals reflected off objects (8) in an environment (7) of the motor vehicle (1 ) are received while the motor vehicle (1 ) is maneuvered along a drive tube (12) to the garage (10), and the objects (8) in the environment (7) are identified based on the received ultrasonic signals and images (13) are provided by a camera (9) of the motor vehicle (1 ) while maneuvering, which describe the environment (7),

characterized in that

if at least one of the objects (8) is identified as an obstacle in the drive tube (12) based on the received ultrasonic signals, it is recognized based on at least one of the images (13), which describes an area in the environment (7) associated with the identified obstacle, if the received ultrasonic signals have been reflected off a component of the garage (10).

2. Method according to claim 1 ,

characterized in that

the obstacle identified based on the received ultrasonic signals remains

unconsidered in maneuvering the motor vehicle (1 ) into the garage (10) if the received ultrasonic signals are recognized as reflected off the component of the garage (10).

3. Method according to claim 1 or 2,

characterized in that

the received ultrasonic signals are recognized as reflected off the component of the garage (10) if at least a part of the garage (10) is recognized in the at least one image (13).

4. Method according to claim 3,

characterized in that

the received ultrasonic signals are recognized as reflected off the component of the garage (10) if an entrance (17) of the garage (10) is recognized in the at least one image (13).

5. Method according to claim 4,

characterized in that

the entrance (17) of the garage (10) is recognized as a rectangle (18) in the at least one image (13), wherein a brightness in the interior and the exterior of the rectangle (18) differs.

6. Method according to claim 4 or 5,

characterized in that

for recognizing the garage (10) in the at least one image (13), a feature is extracted from the at least one image (13), wherein a feature descriptor is used for extracting the feature.

7. Method according to any one of claims 4 to 6,

characterized in that

the recognition of the garage (10) in the at least one image (13) is performed based on machine learning, wherein the machine learning is based on training data, which describes entrances (17) with different lighting conditions.

8. Method according to any one of the preceding claims,

characterized in that

a step (19) in the entrance (17) is recognized in the at least one image (13) as the component of the garage (10), which is disposed between an access path (20) to the garage (10) and a parking area (1 1 ) within the garage (10).

9. Method according to claim 8,

characterized in that

the step (19) in the entrance (17) is recognized as a line in the at least one image (13), which extends substantially along a horizontal.

10. Method according to claim 9 or 10,

characterized in that

the object (8) first identified as an obstacle based on the received ultrasonic signals is classified as an object (8) capable of being traversed by the motor vehicle (1 ) if the step (19) is recognized in the at least one image (13).

1 1 . Method according to any one of the preceding claims,

characterized in that

further objects in the environment (7) are recognized during maneuvering the motor vehicle (1 ) based on the received ultrasonic signals and/or the images.

12. Driver assistance system (2) for a motor vehicle (1 ), which is adapted to perform a method according to any one of the preceding claims.

13. Motor vehicle (1 ) with a driver assistance system (2) according to claim 12.

Description:
Method for at least semi-autonomously maneuvering a motor vehicle into a garage based on ultrasonic signals and camera images, driver assistance system as well as motor vehicle

The present invention relates to a method for at least semi-autonomously maneuvering a motor vehicle into a garage, in which ultrasonic signals are emitted by an ultrasonic sensor of the motor vehicle and ultrasonic signals reflected off objects in the environment of the motor vehicle are received while the motor vehicle is maneuvered to the garage along a drive tube, and the objects in the environment are identified based on the received ultrasonic signals. In addition, images are provided by a camera of the motor vehicle during maneuvering, which describe the environment. Moreover, the present invention relates to a driver assistance system for a motor vehicle. Finally, the present invention relates to a motor vehicle.

Presently, the interest is in particular directed to methods, by which a motor vehicle can be maneuvered or parked into a garage. Hereto, a parking trajectory can for example be determined, which extends from the current position of the motor vehicle outside of the garage to a park position within the garage. Therein, it can be provided that the motor vehicle is at least semi-autonomously maneuvered along this parking trajectory with the aid of a driver assistance system of the motor vehicle. Hereto, the driver assistance system can intervene in the steering and it is further left to the driver of the motor vehicle to actuate the accelerator and the brake. Further, the motor vehicle can also be fully autonomously maneuvered along the parking trajectory. Herein, the driver assistance system further takes over the intervention in a drive engine and/or a brake system of the motor vehicle. Further, methods are known from the prior art, in which the driver exits the motor vehicle and the motor vehicle is then parked into the garage in remote controlled manner.

Hereto, DE 10 2012 200 625 A1 describes a method for carrying out a parking operation of a vehicle by means of a remote control, by which longitudinal and/or lateral control of the vehicle can be effected. Therein, the implementation of the parking operation can be activated via the remote control. Herein, the longitudinal control of the motor vehicle can be controllable by a remote control. The lateral control can be taken over by a parking assistant in the vehicle. Further, the vehicle can have ultrasonic sensors, which detect a garage, in particular the garage walls. In addition, the vehicle can have a front camera and a rear camera, which capture images of the vehicle environment and analyze them for the parking assistance functions.

When the motor vehicle is moved towards the garage, usually, measurements are performed by at least one ultrasonic sensor of the driver assistance system. Based on the measurements of the ultrasonic sensor, obstacles in the environment of the motor vehicle can then be recognized and the parking maneuver into the garage can optionally be aborted. Therein, the problem arises that components of the garage are recognized as obstacles based on the measurements of the ultrasonic sensor. A possibility of avoiding this is in decreasing the sensitivity of the ultrasonic sensor. However, this entails the risk that low obstacles in the environment of the motor vehicle optionally cannot be reliably captured and the motor vehicle thus can collide with them.

It is the object of the present invention to demonstrate a solution how an autonomous parking operation into a garage can be more safely and reliably performed.

According to the invention, this object is solved by a method, by a driver assistance system as well as by a motor vehicle having the features according to the respective independent claims. Advantageous developments of the present invention are the subject matter of the dependent claims.

In an embodiment of a method for at least semi-autonomously maneuvering a motor vehicle into a garage, ultrasonic signals are preferably emitted by an ultrasonic sensor of the motor vehicle and ultrasonic signals reflected off objects in an environment of the motor vehicle are received while the motor vehicle is maneuvered to the garage along a drive tube. Preferably, the objects in the environment are identified based on the received ultrasonic signals. Furthermore, it is preferably provided that images are provided by a camera of the motor vehicle during maneuvering, which describe the environment. If the object is identified as an obstacle in the drive tube based on the received ultrasonic signals, it is in particular recognized based on at least one of the images describing an area in the environment associated with the identified obstacle if the received ultrasonic signals have been reflected off a component of the garage.

A method according to the invention serves for at least semi-autonomously maneuvering a motor vehicle into a garage. Herein, ultrasonic signals are emitted by an ultrasonic sensor of the motor vehicle and ultrasonic signals reflected off objects in an environment of the motor vehicle are received while the motor vehicle is maneuvered to the garage along a drive tube. Further, the objects in the environment are identified based on the received ultrasonic signals. Furthermore, images are provided by a camera of the motor vehicle during maneuvering, which describe the environment. If at least one of the objects is identified as an obstacle in the drive tube based on the received ultrasonic signals, it is recognized based on at least one of the images describing an area in the environment associated with the identified obstacle if the received ultrasonic signals have been reflected off a component of the garage.

With the aid of the method, the motor vehicle is to be at least semi-autonomously maneuvered or parked into a garage. The garage is in particular a garage bounded by walls and a ceiling and having an entrance, which can for example be closed by a garage door. The garage can for example be a single garage or a double garage. The method can for example be performed by a corresponding driver assistance system of the motor vehicle. For at least semi-autonomously maneuvering the motor vehicle, a parking trajectory can first be determined. This parking trajectory can extend from a start point or from the current position of the motor vehicle to a park position within the garage. The motor vehicle can then be maneuvered along this parking trajectory with the aid of the driver assistance system. Hereto, the driver assistance system can perform an

intervention in the steering, a drive engine and/or a brake system of the motor vehicle. Basically, it can also be provided that the motor vehicle is maneuvered along the parking trajectory in remote controlled manner.

During maneuvering the motor vehicle or during movement of the motor vehicle along the parking trajectory, measurements are continuously performed by at least one ultrasonic sensor of the motor vehicle. Hereto, the ultrasonic sensor emits the ultrasonic signal, which is then reflected off the objects in the environment of the motor vehicle. The ultrasonic signals having been reflected off the objects can then again be received by the ultrasonic sensor. The reflected ultrasonic signals can then be evaluated with the aid of an electronic control unit of the driver assistance system. For example, a digital

environmental map can be generated, which describes the environment of the motor vehicle. Further, it can be provided that the objects having been captured by the ultrasonic sensor are correspondingly classified or identified. Herein, the objects can for example be classified as high objects or as low objects. Therein, high objects present an obstacle for the motor vehicle, which can result in a collision between the motor vehicle and the object. Low objects can for example be such objects, which can be traversed by the motor vehicle without damage to the motor vehicle impending. According to an essential aspect of the present invention, it is provided that if an object is identified as an obstacle based on the received ultrasonic signals, which is located in the drive tube, the presence of this obstacle is checked based on at least one image captured by the camera. Thus, it is checked if the obstacle recognized based on the ultrasonic measurements is actually present. In particular, it is checked if this obstacle is disposed in the drive tube. Therein, the drive tube describes that area in the environment of the motor vehicle, which is occupied by the motor vehicle in the future movement of the motor vehicle along the parking trajectory. If the obstacle in the drive tube has been recognized based on the ultrasonic measurements, at least one image is used, which has been captured by the camera and which describes an area in the environment, in which the supposed obstacle is located. Based on this at least one image, it is now checked if this obstacle is actually present or if the received ultrasonic signals originate from a component of the garage, but which is not located in the drive tube. This is based on the realization that certain components or structural parts of the garage can reflect the ultrasonic signal emitted by the ultrasonic sensor such that the reflected ultrasonic signal is recognized as an obstacle in the drive tube. In that it is checked based on the at least one image if this obstacle is actually present or if the received ultrasonic signals originate from a component of the garage. Thereby, reducing the sensitivity of the ultrasonic sensor during parking the motor vehicle into the garage can for example be omitted. Thereby, it can be reliably prevented that for example low objects are not recognized based on the reduced sensitivity of the ultrasonic sensor and thus the motor vehicle collides with them. In this manner, the at least semi-autonomous parking operation into a garage can overall be more safely and reliably configured.

Preferably, the obstacle identified based on the received ultrasonic signals remains unconsidered in maneuvering the motor vehicle into the garage if the received ultrasonic signals are recognized as reflected off the component of the garage. For example, the at least semi-autonomous parking maneuver can be performed such that the parking maneuver is aborted and/or the motor vehicle is decelerated if an obstacle is recognized in the drive tube. If the check based on the image of the camera now has yielded that the obstacle identified based on the ultrasonic measurements actually is not present, this supposed obstacle can remain unconsidered in the parking maneuver. This in particular means that the method is not aborted or the motor vehicle is not decelerated. This allows reliable performance of the parking maneuver.

According to a further embodiment, the received ultrasonic signals are recognized as reflected off the component of the garage if at least a part of the garage has been recognized in the at least one image. The camera of the motor vehicle can for example be oriented and/or formed such that it is able to capture images from an area of the environment, which is located in front of the motor vehicle in direction of travel. If now the garage or at least a part of the garage has been recognized within the image having been captured by the camera, it can be assumed that the ultrasonic signals received by the ultrasonic sensor have been reflected off a part or a component of the garage. Such a component of the garage can for example be a wall of the garage, which bounds the garage or a parking space within the garage. The component can also be a ceiling or a floor of the garage. A garage door, by which an entrance of the garage can be closed, can also be recognized as the component of the garage. Thus, if such a component or part of the garage is arranged in at least one image describing the area of the supposed obstacle, it can be assumed with high likelihood that the reflected ultrasonic signals originate from this component of the garage.

Therein, it is in particular provided that the received ultrasonic signals are recognized as reflected off the component of the garage if an entrance of the garage is recognized in the at least one image. In particular, the entrance of the garage describes that opening of the garage, through which the motor vehicle is moved into the garage in parking. The entrance of the garage can in particular be closed by a garage door. Walls of the garage, the ceiling of the garage and/or a floor of the garage can adjoin to the entrance of the garage. Herein, it has turned out that these components of the garage adjoining to the entrance of the garage can result in reflections of the emitted ultrasonic signal with high likelihood, which can then be identified as an obstacle in the drive tube. Now, if the entrance of the garage and/or components of the garage adjoining to the entrance are recognized in the at least one image, it can be assumed with high likelihood that the reflected ultrasonic signals originate from one of these components.

According to a further configuration, the entrance of the garage is recognized as a rectangle in the at least one image, wherein a brightness differs in the interior and exterior of the rectangle. Thus, it can be checked if a rectangle is located in the at least one image, which is associated with the allegedly recognized obstacle. Herein, it can in particular be checked if an area within the rectangle is darker than an area outside of the rectangle. Therein, it is taken into account that for example the walls bounding the entrance of the garage are brighter than the area within the non-illuminated garage. For example, the presence of a rectangle in the image can be checked with the aid of a corresponding image processing program. If a rectangle has been recognized, it can be checked if areas or pixels within the rectangle have a lower brightness than pixels outside of the rectangle. Thus, the entrance of the garage can be recognized in the at least one image in simple and reliable manner.

According to a further configuration, a feature is extracted from the at least one image for recognizing the garage in the at least one image, wherein a feature descriptor is used for extracting the feature. For example, a corresponding pre-processing of the captured image can first be performed. Furthermore, it can be provided that a corresponding segmentation is for example performed. Subsequent thereto, feature extraction can be performed, in which the entrance of the garage and in particular the rectangle

representing the entrance of the garage is recognized. Hereto, methods of machine vision or computer vision can be used. For feature extraction, a corresponding descriptor or feature descriptor can for example be used. For example, the histogram of oriented gradients (HOG) can be used as the feature descriptor. Moreover, local binary patterns (LBP) can for example be used as the feature descriptor. Thus, the rectangle can be reliably recognized in the image.

Furthermore, it is advantageous if the recognition of the garage in the at least one image is performed based on machine learning, wherein machine learning is based on training data, which describes entrances with different lighting conditions. Furthermore, the machine learning can be used to recognize the entrance of the garage or the rectangle representing the entrance of the garage in the at least one image. Therein, different methods of machine learning and in particular the so-called deep learning can be used. This machine learning can be based on training data or training images, which describe entrances of garages with different lighting conditions. These different lighting conditions can for example arise by different light conditions at day and at night in the environment of the garage. Further, the different lighting conditions can arise in that the garage is illuminated in the interior or that the garage is illuminated in the interior by the headlights of the motor vehicle. These different lighting conditions can be trained by the training data and thus the entrance of the garage can be reliably recognized in the images.

In a further configuration, a step in the entrance is recognized as the component of the garage in the at least one image, which is disposed between an access path to the garage and a parking area within the garage. The parking area, on which the motor vehicle can be parked for parking, can for example be disposed on a floor of the garage. This floor of the garage can for example be situated higher compared to an access path leading to the garage. Thereby, a step arises between the access path and the parking area of the garage. It has turned out that the ultrasonic signals emitted by the ultrasonic sensor are in particular reflected off this step such that they are recognized as obstacles in the drive tube. This in particular exists if a metal frame or sheet metal is disposed on this step or around the entrance. Thus, if such a step or such a sheet metal is recognized in the images, it can be assumed that the reflected ultrasonic signals originate from this step and/or from this sheet metal.

Herein, it is in particular provided that the step in the entrance is recognized as a line in the at least one image, which substantially extends along a horizontal. For example, a corresponding object recognition algorithm can be used, which is able to recognize lines in the at least one image. If the sheet metal is to be recognized, this sheet metal can be recognized with the aid of a corresponding object recognition algorithm based on its shaping, its color or the like. This allows reliable recognition of the step and/or the sheet metal.

Furthermore, it is advantageous if further objects in the environment are recognized during maneuvering the motor vehicle based on the received ultrasonic signals and/or the images. During the movement of the motor vehicle to the garage, they can be captured or recognized based on the ultrasonic measurements and/or the images provided by the camera. Here too, it can be checked if these objects present an obstacle for the motor vehicle. If a recognized object is for example classified as an obstacle and it is recognized that this obstacle has not been caused by an incorrect measurement originating from a component of the garage, a corresponding warning can be output to the driver of the motor vehicle or the parking trajectory can be newly calculated. It can also be provided that the parking maneuver is aborted. This overall allows secure and reliable parking maneuver.

A driver assistance system according to the invention for a motor vehicle is adapted for performing a method according to the invention and advantageous configurations thereof. For example, the driver assistance system can have at least one ultrasonic sensor. In particular, it is provided that the driver assistance system has a plurality of ultrasonic sensors, which are for example disposed distributed at the motor vehicle. Further, the driver assistance system can have at least one camera. Here too, it can be provided that the driver assistance system has a plurality of cameras. In addition, it is in particular provided that the driver assistance system has an electronic control unit, which can receive measured values from the ultrasonic sensor on the one hand, which describe the received ultrasonic signals. In addition, the images can be received by the electronic control unit, which are captured by the at least one camera. A motor vehicle according to the invention includes a driver assistance system according to the invention. The motor vehicle is in particular formed as a passenger car.

The preferred embodiments presented with respect to the method according to the invention and the advantages thereof correspondingly apply to the driver assistance system according to the invention as well as to the motor vehicle according to the invention.

Further features of the invention are apparent from the claims, the figures and the description of figures. The features and feature combinations mentioned above in the description as well as the features and feature combinations mentioned below in the description of figures and/or shown in the figures alone are usable not only in the respectively specified combination, but also in other combinations or alone without departing from the scope of the invention. Thus, implementations are also to be considered as encompassed and disclosed by the invention, which are not explicitly shown in the figures and explained, but arise from and can be generated by separated feature combinations from the explained implementations. Implementations and feature combinations are also to be considered as disclosed, which thus do not have all of the features of an originally formulated independent claim. Moreover, implementations and feature combinations are to be considered as disclosed, in particular by the implementations set out above, which extend beyond or deviate from the feature combinations set out in the relations of the claims.

Now, the invention is explained in more detail based on preferred embodiments as well as with reference to the attached drawings.

There show:

Fig. 1 a motor vehicle according to an embodiment of the present invention, which has a driver assistance system;

Fig. 2 the motor vehicle, which is at least semi-autonomously maneuvered into a garage; and

Fig. 3 an image of the garage having an entrance. In the figures, identical and functionally identical elements are provided with the same reference characters.

Fig. 1 shows a motor vehicle 1 according to an embodiment of the present invention in a plan view. Presently, the motor vehicle 1 is formed as a passenger car. The motor vehicle 1 includes a driver assistance system 2, which can assist a driver of the motor vehicle 1 in at least semi-autonomously maneuvering or in parking. In particular, the driver assistance system 2 is adapted to at least semi-autonomously park the motor vehicle 1 into a garage 10.

Further, the driver assistance system 2 includes an electronic control unit 3. Furthermore, the driver assistance system 2 includes at least one ultrasonic sensor 4. In the present embodiment, the driver assistance system 2 includes eight ultrasonic sensors 4, wherein four ultrasonic sensors 4 are disposed in a front area 5 and four ultrasonic sensors 4 are disposed in a rear area 6 of the motor vehicle 1 . Objects 8 in an environment 7 of the motor vehicle 1 can be captured with the aid of the ultrasonic sensors 4. Hereto, ultrasonic signals can be emitted by the respective ultrasonic sensors 4. The ultrasonic sensors reflected off objects 8 in the environment 7 can again be received by the ultrasonic sensors 4. Measured values can then be provided with the aid of the respective ultrasonic sensors, which describe the reflected ultrasonic signals. These measured values can then be transmitted to the electronic control unit 3. The electronic control unit 3 is connected to the respective ultrasonic sensors 4 for data transmission. Presently, corresponding data lines are not illustrated for the sake of clarity. The objects 8 in the environment 7 can then be classified or identified based on the measured signals with the aid of the electronic control unit 3. It can also be provided that a digital environmental map is provided by the electronic control unit 3, which describes the environment 7 and the objects 8.

Moreover, the driver assistance system 2 includes a camera 9, which can for example be formed as a CCD camera or as a CMOS camera. Image sequences or video data can be provided with the aid of the camera 9, which describe the environment 7 and/or the objects 8. In the present embodiment, the camera 9 is disposed such that an area in front of the motor vehicle 1 in direction of travel can be captured by it. Basically, the

arrangement of the camera 9 is arbitrary. It can also be provided that the driver assistance system 2 has a plurality of cameras 9. The image data provided by the camera 9 can also be evaluated by the electronic control unit 3. Fig. 2 shows the motor vehicle 1 , which is maneuvered into a garage 10 with the aid of the driver assistance system 2. Here, a parking trajectory can first be determined by the driver assistance system 2, along which the motor vehicle 1 is maneuvered. Hereto, the driver assistance system 2 can perform corresponding interventions in the steering, the drive engine and/or the brake system of the motor vehicle 1 . Thereby, the motor vehicle 1 can be maneuvered from a current start position outside of the garage 10 to a final position within the garage 10, which is associated with a parking area 1 1 of the garage 10.

Therein, a drive tube 12 is represented, which describes the area in the environment 7 of the motor vehicle 1 , which the motor vehicle 1 will occupy in the future upon parking into the garage 10. While the motor vehicle 1 is maneuvered, measurements are continuously performed by the ultrasonic sensors 4. Thus, it can for example be determined if an obstacle is located in the drive tube 12, which can result in collision.

When the motor vehicle 1 approaches the garage 10, the ultrasonic signals emitted by the respective ultrasonic sensors 4 can be reflected off a component of the garage 10 and again return to the ultrasonic sensor 4. Therein, it can be the case that an ultrasonic signal is reflected such that the reflected ultrasonic signal or the measuring signal generated therefrom is classified as an obstacle or as a high object 8 located within the drive tube 12 by the electronic control unit 3. This can result in the driver assistance system 2 aborting the parking maneuver or performing deceleration.

If an obstacle is identified in the drive tube 12 based on the measuring signals provided by the ultrasonic sensors 4 during the parking maneuver into the garage 10, an image 13 is captured by the camera 9, which is associated with an area in the environment 7, in which the supposed obstacle is located. It can also be provided that images 13 are continuously captured during travel of the motor vehicle 1 towards the garage 10 and at least the image 13 is used, which describes the area, in which the supposed obstacle is located.

Such an image 13 is shown in Fig. 3. In the image 13, it is apparent that the garage 10 has walls 14, a ceiling 15, a floor 16 as well as an entrance 17 as components. Now, if the garage 10 or a part thereof is recognized in the image 13, it can be assumed that the reflected ultrasonic signal, based on which the supposed obstacle has been recognized, originates from a reflection of the ultrasonic signal off a component of the garage 10. In particular, this can be assumed if the entrance 17 of the garage 10 is recognized in the image 13. It has turned out that in particular components of the garage 10, which are disposed in the entrance 17 or surround it, can cause reflections of the ultrasonic signals, which are recognized as obstacles. Moreover, it can be provided that it is checked if the obstacle recognized based on the measurements by the ultrasonic sensors is actually present in the image.

For recognizing the entrance 17 of the garage 10, it can in particular be checked if a rectangle 18 can be recognized in the image 13. In particular, it can be checked if the pixels within the rectangle 18 are darker than the pixels outside of the rectangle 18. Thus, an internally non-illuminated garage 10 can for example be reliably recognized. For recognizing the rectangle 18 in the image 13, methods of machine vision or computer vision can be used. For example, a feature extraction can be performed. The rectangle 18 in the image 13 can also be recognized with the aid of a corresponding classifier.

Furthermore, it can be provided that the rectangle 18 is recognized in the image 13 based on machine learning. In order to train a corresponding image recognition algorithm or object recognition algorithm, for example, training data or training images can be used. These training images can show different garages 10 and/or different entrances 17.

Similarly, this training data can show garages 10 or entrances 17 with different lighting conditions.

Further, it can be checked if a step 19 is recognized in the image 13, which is recognized between the floor 15 or the parking area 10 and an access path 20, which leads to the garage 10. In the present example, a sheet metal 21 is disposed at this step 19 and around the entrance 17. With the aid of a corresponding object recognition algorithm, the step 19 can for example be recognized as a line in the image 13. The sheet metal 21 can for example be recognized due to its shaping, its color and/or its reflection characteristics in the image 13. In particular, if the step 19 and/or the sheet metal 21 are recognized in the image 13, it can be assumed that the reflected ultrasonic signals can originate from the step 19 and/or the sheet metal 21 .

If it has resulted based on the evaluation of the image 13 that the supposed obstacle, which has been recognized based on the measurements by the ultrasonic sensors 4, is actually not present, this supposed obstacle can remain unconsidered in the parking maneuver of the motor vehicle 1 into the garage 10. For example, the obstacle can be classified as a low object, which can be traversed by the motor vehicle 1 . If a digital environmental map has been provided, the allegedly recognized obstacle can be deleted from the environmental map. This overall allows safe and reliable performance of the autonomous parking maneuver into the garage 10.

PAGE

INTENTIONALLY LEFT BLANK