DE102022203973A1 - Method for determining parking space properties - Google Patents

Abstract

The invention relates to a method for determining properties of a parking space (P4) based on parking scenario properties aggregated in the past.

Description

The invention relates to a method for determining parking space properties based on parking scenario properties aggregated in the past.

Parking assistance systems are already known which generate an environmental model of the vehicle's surroundings using the vehicle's own sensors. The environmental model contains, for example, environmental information about a free parking space, so that the vehicle can be maneuvered into the free parking space at least partially automatically based on this environmental model.

The problem with the direct use of the vehicle's environment model is that the local surroundings of a parking space are often only of limited significance for the ideal orientation of the vehicle in the parking space. This applies in particular if there are no parking space markings or if they are not or only insufficiently recognizable (e.g. due to over-parking by a neighboring third-party vehicle) and/or the third-party vehicles parked next door are not parked in the intended orientation (e.g. slanted position in a vertical parking space) .

Based on this, it is the object of the invention to provide a method for determining properties of a parking space that improves the robustness of the parking assistance system.

The task is solved by a method with the features of independent patent claim 1. Preferred embodiments are the subject of the subclaims. A driver assistance system for determining the properties of a parking space is the subject of independent patent claim 14.

• Zunächst werden Informationen zu zumindest zwei nicht belegten Parklücken ermittelt. Dies erfolgt während der Bewegung eines Fahrzeugs entlang einer Parksituation, die mehrere Parkflächen umfasst. Die Informationen können beispielsweise die direkte Umgebung der Parklücken charakterisierende Informationen sein, beispielsweise die Flanken von Fahrzeugen, Parkplatzmarkierungen etc. Diese werden vorzugsweise von einer Sensorik des Fahrzeugs bereitgestellt. Die Informationen können insbesondere Bestandteil eines Umgebungsmodells des Fahrzeugs sein, das basierend auf Informationen der Sensorik des Fahrzeugs erzeugt wird.

According to a first aspect, a method for determining properties of a parking space based on parking scenario properties aggregated in the past is disclosed. The procedure includes the following steps:

• First, information on at least two unoccupied parking spaces is determined. This occurs while a vehicle is moving along a parking situation that includes several parking spaces. The information can, for example, be information characterizing the immediate surroundings of the parking spaces, for example the flanks of vehicles, parking lot markings, etc. This is preferably provided by a sensor system of the vehicle. The information can in particular be part of an environmental model of the vehicle, which is generated based on information from the vehicle's sensors.

Parking space properties are then determined based on the information obtained about the unoccupied parking spaces, whereby the parking space properties each relate to an individual, unoccupied parking space. The parking space properties can be, for example, information about the parking space width, the parking space orientation and/or the position of the parking space edge.

Parking space metadata is then generated from the parking space properties of the individual identified, unoccupied parking spaces. The parking space metadata summarizes the parking space properties, which each contain information about the individual parking spaces, and combines them so that scenario knowledge about the currently existing parking scenario is created from several parking space properties of different parking spaces, which is contained in the parking space metadata.

As the vehicle moves further, another parking space is detected based on the vehicle's sensors.

Finally, properties of this additional free parking space are determined based on the parking space metadata.

The technical advantage of the method is that by including scenario knowledge of previously recognized free parking spaces, the parking space properties of a free parking space not only have to be determined based on information that was determined directly about this parking space, but parking space properties can be derived using the scenario knowledge . As a result, the robustness of the parking assistance system can be increased and, moreover, an automatic parking process can be initiated very early, since by using the parking space metadata, parking space properties for the further parking space can be determined at an early stage.

According to one exemplary embodiment, information in the surrounding area of the respective unoccupied parking spaces, for example edges or flanks of third-party vehicles, parking space boundaries and/or parking space markings, etc., is recorded by the vehicle's sensors. Based on the information collected, the parking space characteristics are then derived. This means that the parking space characteristics can be determined directly from the sensor information of the vehicle or from the information contained in the environment model.

According to one embodiment, the parking space properties include the width of the unoccupied parking space measured in the direction of travel of the vehicle. The width of the unoccupied parking space can be used to determine whether the parking scenario has vertical or parallel parking spaces.

According to one embodiment, the parking space metadata includes a parking space width that is determined by averaging the width of the previously recorded, unoccupied parking spaces. This means that the width of a free parking space that will be recognized in the future can be derived from the parking space metadata.

According to one embodiment, the parking space properties include an alignment angle of the parking space, the alignment angle being derived based on environmental objects that delimit the respective parking space and/or parking space boundary markings. The alignment angle can in particular indicate what angle the parking space has relative to the longitudinal extent of the street from which the parking lot is entered. This makes it possible to determine the orientation of sloping parking spaces in order to position the vehicle appropriately within the parking space when parking.

According to one embodiment, the parking space metadata includes a parking space alignment angle, the parking space alignment angle being determined by averaging the alignment angles of the previously detected, unoccupied parking spaces. The averaging ensures that, despite deviations of individual parked vehicles from a target alignment, the alignment angle with which the vehicle is placed in a free parking space is adapted as closely as possible to the target alignment of the parking situation.

According to one exemplary embodiment, the parking space metadata includes information about the parking space type, the parking space type indicating whether the previously recorded, unoccupied parking spaces in the parking situation are a vertical parking space, a parallel parking space or an inclined parking space. The parking space types “vertical parking space” and “parallel parking space” can be classified across width ranges of the parking space width, for example in such a way that the parking spaces in a parking situation, which, for example, have an average width of less than or equal to 3m, are classified as vertical parking spaces and parking spaces with a width of 5m or larger than parallel parking spaces. It is understood that the threshold values mentioned only serve as examples and the threshold values can also have other values.

According to one embodiment, the parking space type is determined based on an averaging of the alignment angle of the previously detected, unoccupied parking spaces. The alignment angle can be used as a criterion in particular when an inclined parking situation is to be detected.

According to one embodiment, the parking space properties include the location of the edge of the parking space that faces the lane of the vehicle. The edge of the parking space indicates the boundary at which the parking space ends or the road from which the parking space is accessed begins. By determining the edge of the parking space, it is possible to determine how far a vehicle should be placed in the depth direction into a future identified parking space in order not to protrude onto the road.

According to one exemplary embodiment, the parking space metadata includes information about the position of the parking space edge, wherein the information about the position of the parking space edge is determined by averaging the position of the parking space edges of the previously recorded, unoccupied parking spaces. This makes it possible to derive information from the parking scenario as to how far in the depth direction the vehicle should be placed in a future identified parking space, without having to orientate oneself to the immediate surroundings, as this can often only provide inadequate information about the depth placement of the vehicle (for example due to different lengths or widths of third-party vehicles).

According to one embodiment, determining properties of the additional free parking space based on the parking space metadata includes deriving the orientation of the additional free parking space and/or the edge of the parking space that faces the vehicle's lane. “Deriving” here means either the reading of the respective information from the parking space metadata and its direct application to the placement of the vehicle in the parking space or the modification of this information based on environmental information recognized locally in the other parking space (e.g. adjustment based on recognizable parking space markings or due to other boundary conditions of the additional parking space).

According to one exemplary embodiment, the properties of the further free parking space are determined based on the parking space metadata in such a way that the vehicle can drive past the further free parking space without first driving past gap is controlled directly into this. This means that the parking process can be significantly shortened and the driver's acceptance of the parking assistance system can be increased.

According to one exemplary embodiment, when the vehicle enters the further free parking space, the properties of the further free parking space are at least partially modified based on the environmental information recorded by the sensor system of the vehicle. The vehicle is then controlled to take a final parking position based on the modified properties of the remaining free parking space. This means that the parking space metadata is used as a starting point in order to be able to offer the free parking space for use at an early stage. The positioning of the vehicle in the parking space can then be adjusted if, based on the environmental information recorded by the vehicle's sensors, it can be seen that the positioning of the vehicle needs to be modified differently from the parking space metadata.

• - Ermitteln von Informationen zu zumindest zwei nicht belegten Parklücken während der Bewegung des Fahrzeugs entlang einer Parksituation, die mehrere Parkflächen umfasst;
• - Ermitteln von Parklückeneigenschaften zu den einzelnen nicht belegten Parklücken;
• - Erstellen von Parklückenmetadaten aus den Parklückeneigenschaften der einzelnen ermittelten, nicht belegten Parklücken;
• - Erkennen einer weiteren Parklücke nach dem Weiterbewegen des Fahrzeugs; und
• - Bestimmen von Eigenschaften der weiteren freien Parklücke basierend auf den Parklückenmetadaten.

According to a further aspect, a driver assistance system for determining properties of a parking space based on parking scenario properties aggregated in the past is disclosed. The driver assistance system is coupled to a sensor system of the vehicle and includes a computer unit with a processor and a memory unit that is designed to process the environmental information provided by the sensor system. The computing unit is configured to perform the following steps:

• - Determining information about at least two unoccupied parking spaces while the vehicle is moving along a parking situation that includes several parking spaces;
• - Determining parking space characteristics for the individual unoccupied parking spaces;
• - Creation of parking space metadata from the parking space properties of the individual identified, unoccupied parking spaces;
• - Detecting another parking space after the vehicle has moved on; and
• - Determine properties of the additional free parking space based on the parking space metadata.

The term “parking area” within the meaning of the present disclosure is understood to mean a local area that can be used for parking. A parking area can be occupied by a vehicle (parked parking area) or free (unparked parking area).

The term “parking space” in the sense of the present disclosure is to be understood as meaning a parking area that is free, i.e. a parking area that is not parked.

Further developments, advantages and possible applications of the invention also emerge from the following description of exemplary embodiments and from the figures. All described and/or illustrated features, individually or in any combination, are fundamentally the subject of the invention, regardless of their summary in the claims or their relationship. The content of the claims is also made part of the description.

• 1 beispielhaft eine schematische Darstellung eines Fahrzeug, das ein Fahrerassistenzsystem mit einer Rechnereinheit und einer Sensorik zur Erfassung von Umgebungsinformationen aufweist;
• 2 beispielhaft und schematisch die Darstellung einer Schrägparksituation, bei der ein Fahrzeug an einer Parkreihe mit Schrägparkplätzen vorbeifährt und bei der mehrere freie Parklücken vorhanden sind;
• 3 beispielhaft eine schematische Darstellung, bei der ein Fahrzeug direkt in eine freie Parklücke einfährt, ohne vorher an dieser vorbeizufahren, um Parklückeneigenschaften zu erfassen;
• 4 beispielhaft und schematisch die Darstellung einer Senkrechtparksituation, bei der ein Fahrzeug an einer Parkreihe mit Senkrechtparkplätzen vorbeifährt und bei der direkt nebeneinander mehrere freie Parklücken vorhanden sind;
• 5 beispielhaft und schematisch die Darstellung einer Senkrechtparksituation, bei der ein Fahrzeug an einer Parkreihe mit Senkrechtparkplätzen vorbeifährt und bei der eine freie Parklücke beidseitig von schräg parkenden Fahrzeugen umgeben ist;
• 6 beispielhaft und schematisch die Darstellung einer Senkrechtparksituation, bei der ein Fahrzeug an einer Parkreihe mit Senkrechtparkplätzen vorbeifährt und bei der eine freie Parklücke beidseitig von sehr tief parkenden, kurzen Fahrzeugen umgeben ist; und
• 7 beispielhaft ein Blockdiagramm, das die Verfahrensschritte des Verfahrens zur Bestimmung von Eigenschaften einer Parklücke basierend auf in der Vergangenheit aggregierten Parkszenario-Eigenschaften verdeutlicht.

The invention is explained in more detail below using the figures and exemplary embodiments. Show it:

• 1 By way of example, a schematic representation of a vehicle that has a driver assistance system with a computer unit and a sensor system for recording environmental information;
• 2 By way of example and schematically, the representation of an inclined parking situation in which a vehicle drives past a row of parking spaces with inclined parking spaces and in which there are several free parking spaces;
• 3 For example, a schematic representation in which a vehicle drives directly into a free parking space without first driving past it in order to record parking space properties;
• 4 By way of example and schematically, the representation of a vertical parking situation in which a vehicle drives past a row of parking spaces with vertical parking spaces and in which there are several free parking spaces right next to each other;
• 5 By way of example and schematically, the representation of a vertical parking situation in which a vehicle drives past a row of parking spaces with vertical parking spaces and in which a free parking space is surrounded on both sides by vehicles parked at an angle;
• 6 By way of example and schematically, the representation of a vertical parking situation in which a vehicle drives past a row of parking spaces with vertical parking spaces and in which a free parking space is surrounded on both sides by very low, short vehicles; and
• 7 For example, a block diagram that shows the procedural steps of the method for determining the properties of a parking space based on parking scenario characteristics aggregated in the past.

1 shows an example and schematic of a vehicle 1 that has a driver assistance system. The driver assistance system includes in particular a parking assistance system, by means of which the vehicle can be maneuvered into a free parking space at least partially automatically. The vehicle 1 has a sensor system 2 for detecting the vehicle surroundings. The sensor system 2 can have several sensors distributed around the vehicle 1. The sensors can be, for example, ultrasonic sensors, one or more cameras, one or more radar sensors and/or one or more LIDAR sensors. It goes without saying that the vehicle can have sensors of different sensor types.

The driver assistance system has a computer unit 3. The computer unit 3 comprises a processor and a memory unit in order to be able to execute a program for carrying out the method steps described below.

The driver assistance system is designed to recognize a parking situation. Such a parking situation can be present, for example, in a parking lot, in a parking garage, etc. and can be characterized by the fact that there are a large number of parking areas that have the same size and orientation, at least in some areas. In such a parking situation, the driver assistance system is designed to determine properties of these free parking spaces based on free parking spaces in order to be able to draw conclusions about which properties another free parking space that the vehicle 1 will recognize in the future will most likely have to be able to offer this parking space to the vehicle 1 as early as possible for automated parking, in particular in such a way that the vehicle 1 does not first have to be moved past the parking space in order to recognize the parking space properties for an automated parking process, but rather into the parking space directly, i.e. without a prior Drive past, can be driven in.

2 shows a parking situation and a vehicle 1 that drives along this parking situation. It should be noted that the in the 2 The different positions of the vehicle 1 shown are taken at different times due to its further movement in the direction of travel FR.

The parking situation is characterized by a large number of parking areas, some of which are parked. The parked parking areas are each marked by an oval. Between the parked parking areas there are two unparked parking areas, namely parking spaces P1, P2. In the exemplary embodiment shown, the parking spaces P1, P2 are each laterally delimited by surrounding objects in the form of third-party vehicles, so that the parking space properties can be derived. For example, the flanks of vehicles parked adjacent to each other can be used to determine the characteristics of the parking space.

Alternatively or additionally, other features of the parking space can also be used to derive parking space properties, for example parking space markings applied to the ground.

The parking space properties can include, for example, the width of the parking space, the orientation of the parking space and/or the location of the edge of the parking space that is adjacent to the roadway on which the vehicle is moving.

The width b of the parking space can be used to determine whether it is a vertical parking space or a parallel parking space. Based on the orientation of the parking space, it can be deduced what angular position the parking space has relative to the direction of travel of the vehicle 1 or the longitudinal direction of the road along the row of parking spaces. The angular position of the parking space can be specified, for example, by the angle α, which is included between the longitudinal axis of the parking space and a straight line parallel to the direction of travel of the vehicle 1 or the longitudinal extent of the road over which the parking space is driven. Based on the edge of the parking space, it can be determined where the parking space begins in order to be able to estimate how far the vehicle 1 has to be maneuvered into the parking space in order to prevent the vehicle 1 from protruding into the road from which the parking space is being driven.

The driver assistance system of the vehicle 1 determines the parking space characteristics while the vehicle 1 drives along the parking spaces. This is done, for example, by creating an environment model based on the sensor system 2 of the vehicle 1. The environment model can contain information about the free parking spaces P1, P2. These can be, for example, parking space boundary lines, contours of the surrounding objects that laterally delimit the parking space (e.g. third-party vehicles parked next to the parking space), etc. The parking space properties can then be determined from this information, for example the width b of the parking space, the alignment angle α of the parking space and/or the position of the edge of the parking space PR.

When the vehicle 1 moves along several free parking spaces P1, P2, parking space properties for the respective free parking spaces P1, P2 are recorded and stored. Based on these parking space characteristics, parking space metadata is then calculated.

When calculating the parking space metadata, the parking space properties of the individual free parking spaces P1, P2 are processed and combined so that the parking space metadata is determined taking into account information about several parking spaces that characterize the parking scenario.

The parking space metadata can, for example, contain information about the width of the parking spaces. This information is determined, for example, by averaging the widths b of the previously recorded free parking spaces P1, P2.

In addition, the parking space metadata can contain, for example, information about the alignment angle α of the parking spaces. This information is determined, for example, by averaging the alignment angle α of the previously detected free parking spaces P1, P2.

By using the information about the parking space width and/or the alignment angle α of the parking space contained in the parking space metadata, the parking space type can be classified as a vertical parking space, parallel parking space or slanted parking space. For this purpose, for example, a threshold value for the width b of the parking space can be used. If the width b of the parking space is smaller than the threshold value of the parking space width, the parking space can be classified as a vertical or inclined parking space. However, if the width b of the parking space is greater than the parking space width threshold, the parking space can be classified as a parallel parking space. The alignment angle contained in the parking space metadata can also be used to classify the parking space, in particular to distinguish a vertical parking space from an inclined parking space.

In addition, the parking space metadata can contain, for example, information about the location of the parking space edge PR. This information is determined, for example, by averaging the position of the parking space edges PR of the previously recorded free parking spaces P1, P2.

Based on the parking space metadata, it is possible, immediately after detecting another free parking space, to provide the information as to how the vehicle 1 should be positioned in the free parking space, i.e. the final position and final orientation of the vehicle 1 in the further free parking space.

3 shows an example of an inclined parking scenario in which a vehicle 1 recognizes this parking space before driving past a free parking space.

Due to the parking space metadata that the driving assistance system of the vehicle 1 has obtained based on previously detected free parking spaces, it is possible to drive into the parking space directly, i.e. without driving past. This can significantly increase parking comfort.

When starting the parking process, parking space properties for the free parking space are preferably first derived from the parking space metadata. These parking space properties can be suitably adapted, for example, after the vehicle 1 has partially entered the parking space, if the sensor system 2 of the vehicle 1 determines parking space properties that deviate from the parking space properties of the parking space metadata.

4 shows a parking scenario in which, after driving past several free parking spaces P1, P2, P3, another free parking space P4 is detected by the sensor system 2 of the vehicle 1. The width of the additional free parking space is significantly larger than the width of the previously recognized free parking spaces P1, P2, P3. Since the parking space metadata has classified the parking scenario as a vertical parking scenario and the average width of the parking spaces P1, P2, P3 is much smaller than the width of the other free parking space P4, it can be concluded that the other free parking space is not a parallel parking space, but rather several free vertical parking spaces. Accordingly, the parking assistance system can directly provide the driver with a suggestion for a vertical parking maneuver in one of the free vertical parking spaces.

5 illustrates another vertical parking scenario in which, after driving past several free parking spaces P1, P2, P3, another free parking space P4 is detected by the sensor system 2 of the vehicle 1. The other free parking space P4 is bordered on the side by third-party vehicles F1, F2. The third-party vehicles F1, F2 park differently from the specified parking orientation, namely the third-party vehicle F1 is turned to the right and the third-party vehicle F2 is turned in the opposite direction, ie to the left. In such a parking scenario, it would be unfavorable to determine an orientation for parking space P4 based solely on the immediate surrounding objects. Based on the scene knowledge, i.e. the parking space metadata based on the free parking spaces P1, P2, P3 have been determined, it is possible to derive an orientation for the additional free parking space P4, thereby improving the parkability of the parking space P4.

6 shows a third vertical parking scenario, in which, after driving past several free parking spaces P1, P2, P3, another free parking space P4 is detected by the sensor system 2 of the vehicle 1. The other free parking space P4 is in turn bordered laterally by two third-party vehicles F1, F2. In contrast to the parking scenario in 5 The third vehicles F1, F2 are significantly shorter than the free parking space P4. As a result, the third-party vehicles F1, F2 are arranged far back from the edge of the parking space PR. In such a parking scenario, it would be unfavorable to determine the parking space edge PR of the parking space P4 based solely on the immediate surrounding objects. Based on the scene knowledge, ie the parking space metadata, which was determined based on the free parking spaces P1, P2, P3, it is possible to derive the parking space edge PR for the further free parking space P4, thereby improving the placement of the vehicle 1 in the parking space P4 becomes.

7 shows a schematic block diagram illustrating the operations of the method for determining properties of a parking space based on parking scenario properties aggregated in the past.

First, information about at least two unoccupied parking spaces is determined while a vehicle is moving along a parking situation that includes several parking spaces (S10). Parking space characteristics are then determined for the individual unoccupied parking spaces (S11).

Parking space metadata is then created from the parking space properties of the individual identified, unoccupied parking spaces (S12).

After the vehicle has moved on, another parking space is then detected based on a sensor system of the vehicle (S13).

Finally, properties of the additional free parking space are determined based on the parking space metadata (S14).

The invention has been described above using exemplary embodiments. It is understood that numerous changes and modifications are possible without departing from the scope of protection defined by the patent claims.

Reference symbol list

1

vehicle

2

Sensor technology

3

Computer unit

α

Alignment angle

b

Width of the parking space

F1, F2

Third vehicle

FR

Direction of travel

P1

parking lot

P2

parking lot

P3

parking lot

P4

another parking space

PR

Parking space edge

Claims (14)

Method for determining properties of a parking space (P4) based on parking scenario properties aggregated in the past, the method comprising the following steps: - Determining information about at least two unoccupied parking spaces (P1, P2, P3) while a vehicle (1) is moving along a parking situation that includes several parking spaces (S10); - Determining parking space characteristics for the individual unoccupied parking spaces (P1, P2, P3) (S11); - Creation of parking space metadata from the parking space properties of the individual identified, unoccupied parking spaces (P1, P2, P3) (S12); - Based on a sensor system (2) of the vehicle (1), detection of another parking space (P4) after the vehicle (1) has moved further (S13); and - Determine properties of the additional free parking space (P4) based on the parking space metadata (S14). Procedure according to Claim 1 , characterized in that information in the surrounding area of the respective unoccupied parking spaces (P1, P2, P3) is recorded by the sensor system (2) of the vehicle (1) and that the parking space properties are derived based on the recorded information. Procedure according to Claim 1 or 2 , characterized in that the parking space properties include the width (b) of the unoccupied parking space (P1, P2, P3) measured in the direction of travel (FR) of the vehicle (1). Procedure according to Claim 3 , characterized in that the parking space metadata include a parking space width, which is determined by averaging the width (b) of the previously recorded, unoccupied parking spaces (P1, P2, P3). Method according to one of the preceding claims, characterized in that the parking space properties include an alignment angle (α) of the parking space (P1, P2, P3), the alignment angle (a) being based on surrounding objects that occupy the respective parking space (P1, P2, P3). limit, and/or parking space boundary markings are derived. Procedure according to Claim 5 , characterized in that the parking space metadata includes a parking space alignment angle, the parking space alignment angle being determined by averaging the alignment angles (a) of the previously recorded, unoccupied parking spaces (P1, P2, P3). Method according to one of the preceding claims, characterized in that the parking space metadata includes information about the parking space type, the parking space type indicating whether the previously recorded, unoccupied parking spaces (P1, P2, P3) of the parking situation are a vertical parking space, a parallel parking space or is a sloping parking lot. Procedure according to Claim 7 , characterized in that the parking space type is determined based on an averaging of the alignment angle (α) of the previously recorded, unoccupied parking spaces (P1, P2, P3). Method according to one of the preceding claims, characterized in that the parking space properties include the position of the parking space edge (PR) which faces the lane of the vehicle (1). Procedure according to Claim 9 , characterized in that the parking space metadata includes information about the position of the parking space edge, the information about the position of the parking space edge being determined by averaging the position of the parking space edges (PR) of the previously recorded, unoccupied parking spaces (P1, P2, P3). Method according to one of the preceding claims, characterized in that determining properties of the further free parking space (PR) based on the parking space metadata derives the orientation of the further free parking space and/or the edge of the parking space which faces the lane of the vehicle (1). , includes. Method according to one of the preceding claims, characterized in that the determination of properties of the further free parking space (P4) based on the parking space metadata takes place in such a way that the vehicle (1) directly drives past the further free parking space (P4) without previously driving past this can be controlled. Method according to one of the preceding claims, characterized in that when the vehicle (1) enters the further free parking space (P4), an at least partial modification of the properties of the further free parking space (P4) based on the sensor system of the vehicle (1) captured environmental information and that the vehicle (1) is controlled to take a final parking position based on the modified properties of the other free parking space (P4). Driver assistance system for determining properties of a parking space (P4) based on parking scenario properties aggregated in the past, the driver assistance system being coupled to a sensor system (2) of the vehicle (1) and comprising a computer unit (3) with a processor and a memory unit, which is designed to process the environmental information provided by the sensor system (2), the computer unit (3) being configured to carry out the following steps: - Determining information about at least two unoccupied parking spaces (P1, P2, P3) while the vehicle (1) is moving along a parking situation that includes several parking spaces; - Determining parking space characteristics for the individual unoccupied parking spaces (P1, P2, P3); - Creation of parking space metadata from the parking space properties of the individual identified, unoccupied parking spaces (P1, P2, P3); - Detecting another parking space (P4) after the vehicle (1) has moved on; and - Determine properties of the additional free parking space (P4) based on the parking space metadata.

Images