Method for detecting at least one usable parking space, motor vehicle and data processing device

Abstract

The invention relates to a method for detecting at least one usable parking space (1) by a motor vehicle (2), wherein sensor data (3) of a sensor system (4) of the motor vehicle (1), said data relating to a potential parking region (6) adjacent to a road (5) being driven on, are captured and evaluated in order to identify at least one free space (7, 8) within the potential parking region (6), the geometric dimension (9, 10) of said free area enabling the motor vehicle (2) to park on the free space (7, 8), wherein the respective free space (7, 8) is only classified as a usable parking space (1) if no exclusion condition is fulfilled for the respective free space (7, 8), wherein at least the following are evaluated as an exclusion condition (11, 12): - a statistical condition (13), the fulfilment of which depends on usage information (14) relating to a historical use of the free space (7, 8) as a parking space (1) by other motor vehicles (15, 16); and / or - an environmental condition (17), the fulfilment of which depends on whether the sensor system (4) captures, within a background region (18), which is located on the side of the free space (7, 8) facing away from the road (5), a detected motor vehicle (34) and / or a structure (20) and / or vegetation (21) classified as limiting a roadway (19).

Description

[0001] Method for detecting at least one usable parking space, motor vehicle and data processing device

[0002] DESCRIPTION:

[0003] The invention relates to a method for detecting at least one usable parking space by a motor vehicle. Sensor data from a sensor system of the motor vehicle relating to a potential parking area located adjacent to a busy road are recorded and evaluated to identify at least one open space within the potential parking area whose geometric dimensions allow the motor vehicle to park there. The respective open space is classified as a usable parking space only if no exclusion condition is met for the respective open space. The invention also relates to a motor vehicle and a data processing device.

[0004] Systems for automated or assisted parking of motor vehicles are widespread today. Such systems can be referred to as park steering assistants or intelligent parking assistants. The first step is usually to search for a parking space. To do this, the vehicle's sensors, typically while driving slowly, measure the surroundings, for example using ultrasonic sensors mounted on the side. If a gap of a certain size is detected between two vehicles or other objects, it is assumed that this is a suitable parking space. Using the determined size values, an attempt is made to distinguish between longitudinal and transverse gaps. If the sensors include a camera, parking space markings can also be detected and taken into account.If desired, the parking systems can also search for gaps without prior interaction from the driver, provided certain boundary conditions, for example regarding distances and speed, are met.

[0005] In the simplest case, it is assumed that the parking spaces found are actually usable. However, in densely populated urban areas, for example, this can lead to incorrectly identifying parking spaces that are actually unusable, for example, because they are located in a driveway or in a no-parking zone. This reduces the acceptance of parking assistance systems.

[0006] Furthermore, false-positive detection of parking spaces is highly problematic in autonomous driving in urban environments, as a journey often ends with a parking maneuver. It is difficult for users to understand why parking at the end of an automated journey requires the driver to manually select a parking space or even manually search for parking spaces.

[0007] The publication DE 10 2004 047 507 A1 therefore proposes evaluating additional information, namely map data and / or sensor-detected traffic signs and / or sound signals emitted in areas where parking is prohibited, to better identify areas where parking is prohibited. However, this approach requires the explicit marking of corresponding areas on maps or by signs or audio transmitters and is therefore not robustly applicable across the board.

[0008] The invention is therefore based on the object of further improving automated parking space detection, in particular avoiding frequent false-positive detection of parking spaces even in areas without explicit parking prohibition markings.

[0009] The object is achieved according to the invention by a method of the type mentioned at the outset, wherein as an exclusion condition at least - a statistical condition, the fulfilment of which depends on usage information regarding a historical use of the open space as a parking area by other motor vehicles, and / or

[0010] - an environmental condition, the fulfilment of which depends on whether the sensors detect a detected motor vehicle and / or buildings and / or vegetation classified as delimiting a roadway within a background area arranged on the side of the open space facing away from the road, is evaluated.

[0011] Within the scope of the invention, it was recognized that by evaluating the historical behavior of other motor vehicles and / or by detecting or classifying the aforementioned objects in the background area, the probability of a false positive detection of open spaces as usable parking areas can be significantly reduced.

[0012] If it is detected, for example on the basis of position data of the other motor vehicles, that the other motor vehicles only stop very rarely and / or only for a very short time in this open space, the open space is at least historically not used as a parking space and it can therefore be assumed that the open space will continue not to be available as a parking space, for example because parking there is prohibited or because it blocks an entrance or an intersection.

[0013] It was also recognized that open spaces where parking would block a driveway, such as a property entrance, a garage entrance, or an intersection, can be easily detected by detecting or classifying the aforementioned objects in the background area behind the open space. If a motor vehicle is detected in the background area, this indicates that the background area should be passable, which in turn can indicate that access should be via the open space. Since it can be difficult to robustly classify the driveway itself in the sensor data for property entrances, building entrances, and the like, it is proposed to detect relevant boundary objects instead, or at least in addition to them, i.e. specifically the boundary of the driveway caused by buildings or vegetation.

[0014] For example, posts that limit the roadway or an archway can be recognized as buildings, whereby in addition to the pure geometry of the structure, additional features, such as a potentially existing height or width limit sign, can also be evaluated.

[0015] For example, hedges or bushes bordering the road can be identified as limiting vegetation.

[0016] A distinction, for example, between a roadway and a footpath can be made by evaluating the distance between the lateral boundaries of the path caused by buildings or vegetation.

[0017] The sensor technology can comprise multiple sensors and / or multiple sensor types. It is possible that partial data from different sensors or sensor types can be used to detect the open space and to detect the aforementioned objects in the background. For example, the presence of open spaces can be detected using ultrasonic and / or radar sensors, and objects in the background can be classified based on image data from a camera. However, it is also possible to use the same sensor data for both purposes—for example, image data from a camera or LIDAR data—or to use a fusion of data from multiple sensors for both purposes.

[0018] The evaluation of the sensor data to identify the open space and / or the evaluation of the exclusion condition or at least parts of the exclusion conditions and / or the classification of the open space as a parking space, which depends on this, can be carried out by a processing device of the motor vehicle. Alternatively, however, parts or all of these steps can also be implemented in the method according to the invention by a vehicle-external processing device, for example, a server or a cloud solution, whereby, due to the use of the vehicle-mounted sensors, the usable parking space is still detected by the motor vehicle.

[0019] Shifting at least parts of the processing steps to an external device may be expedient, for example, if the external device serves to provide detected parking spaces to motor vehicles or other end devices, for example, via a web interface. However, it can also be used in addition or as an alternative to supplementing the computing power available in the vehicle.

[0020] The usage information can be provided, in particular, by a device external to the vehicle, such as a server or a cloud, which aggregates information regarding the historical use of the open space. For example, corresponding data can be provided by the other vehicles themselves to the device external to the vehicle, where it can be collected and evaluated.

[0021] As will be explained in more detail later for an advantageous embodiment, the usage information can be taken from map data stored locally in the motor vehicle or on a server or in the cloud, which map data assigns different usage information to different positions. The position of the open space relative to the motor vehicle is known. With the help of conventional positioning devices, for example with a GPS sensor and a compass, the position and orientation of the motor vehicle relative to a global coordinate system can also be determined, whereby the usage information assigned to the respective position of the open space can be taken from the map. In addition to at least one of the two exclusion conditions mentioned above, at least one further exclusion condition can be taken into account.The additional exclusion condition can, for example, detect and evaluate traffic signs and / or road markings and / or lowered curbs in the sensor data to identify, for example, no-parking zones and / or fire department access points, and / or evaluate annotated map data that describe, for example, no-parking zones and / or fire department access points. Such map data can also specify time-dependent or other parking restrictions, for example, resident parking zones.

[0022] In some countries and regions, it may also be appropriate that an additional exclusion condition is always met, as long as, for example, no blue or yellow curb, as used in some countries to mark permitted parking areas, is detected adjacent to the open space.

[0023] If at least the statistical condition is evaluated as an exclusion condition, the additional motor vehicles can each implement a driver assistance system by means of which usable parking spaces are detected, wherein, after the detection of a respective usable parking space, the respective additional motor vehicle can determine parking information which indicates whether the additional motor vehicle is actually parked in this parking space and / or whether parking in this parking space is refused by the driver of the additional vehicle, and can transmit this to a device external to the vehicle, wherein the device external to the vehicle can determine or update the usage information or preliminary information on the basis of which the usage information can be determined on the basis of the parking information received from the additional motor vehicles.

[0024] The driver assistance system of the respective additional motor vehicle can, for example, implement the method according to the invention to detect the respective usable parking space. In particular, the usage information or the advance information can be adapted iteratively by detecting, each time the method according to the invention is carried out after the detection of the usable parking space, whether this parking space is actually being used for parking or whether use is being rejected. For example, the usage information or advance information can be updated after each detection and transmission of this parking information, or after a certain period of time or a certain number of newly received parking information items for a respective position or area.

[0025] A driver can, for example, explicitly refuse parking by means of an operator input or implicitly by continuing to drive or parking outside the respective detected usable parking space.

[0026] In particular, the respective additional motor vehicle can report the detection of the usable parking space and the parking there, or the refusal of parking there, to a server, for example, a server of the vehicle manufacturer, or a cloud system, after which the server or cloud system maintains corresponding statistics on the use of the usable parking areas or parking transactions at their locations. In general, swarm data can thus be used to improve the detection of usable parking spaces. The analysis of parking transactions can be carried out for very large areas, for example, entire countries or continents, or locally, for example, for specific cities or neighborhoods.

[0027] The usage information or advance information may also indicate a conditional use of certain areas as parking areas by the other motor vehicles, for example if the other motor vehicles actually parked in this area only for a short period of time and / or only on certain days or at certain times.

[0028] The preliminary information can specify position information as a map for various positions on the map, whereby the respective position information describes how often, after a usable parking space at the respective position was determined, a respective one of the other motor vehicles actually parked at this position and / or how often parking there was refused by the driver of the respective other motor vehicle, whereby the position information for the position of the determined open space is used as usage information.

[0029] In particular, the advance information can be stored by the vehicle or another device external to the vehicle, and the vehicle can, for example, only transmit the position of the determined open space to this device in order to query the usage information. Alternatively, it would also be possible to provide the advance information to the vehicle, for example, at the start of the journey or when entering a relevant area, and to select the usage information from the map locally in the vehicle itself.

[0030] If at least the environmental condition is evaluated as an exclusion condition, pillars and / or a gate and / or a façade opening can always be classified as structures delimiting the driveway, or if an additional condition is met, and / or hedge sections and / or planted beds can be classified as vegetation delimiting the driveway. In particular, open gates and / or archways can also be classified as gates and thus as structures delimiting the driveway. Façade openings are often used in urban environments, for example, to provide access to backyards or similar.

[0031] The additional condition can evaluate the distance between pillars, hedge sections, or flowerbeds, or the width of the gate or façade opening, and / or the height of the gate or façade opening, and / or the height of a lower edge of the façade opening. In particular, the additional condition can be met if the distance lies within a predefined interval, for example, if the distance is greater than 2 m, greater than 2.5 m, and / or less than 4 m, or less than 8 m. Very small distances do not allow a motor vehicle to pass and therefore are unlikely to constitute structures in front of which parking should be prohibited. If the distances are too large, it can generally also be assumed that the aforementioned objects do not delimit a driveway or similar structure.Regarding the height extension, the additional condition can be met, for example, if the height extension lies within a specified height interval, for example, if the height extension is greater than 2 m or greater than 3 m and / or less than 6 m or less than 10 m. This size range is typical for facade openings. Considering the height of the lower edge can be useful to more robustly distinguish driveways from shop windows, which could potentially be mistakenly identified as facade openings, and similar structures.

[0032] If at least the ambient condition is evaluated as an exclusion condition, the motion state of the detected vehicle can be determined by evaluating the sensor data upon detection of a detected vehicle in the background area. The fulfillment of the ambient condition depends on the motion state. In particular, the exclusion condition can only be fulfilled upon detection of a stationary detected vehicle in the background area, since a stationary vehicle in the background area allows for particularly robust detection of situations in which a vehicle would be parked in an open space or a driveway would be blocked when the vehicle is parked.

[0033] After detecting one or more usable parking spaces, an assistance system of the motor vehicle can control at least one actuator of the motor vehicle to automatically perform longitudinal and / or lateral driving interventions, allowing the motor vehicle to be guided at least partially automatically to the usable parking space or a selected one of the usable parking spaces. Excluding unsuitable open spaces is particularly relevant for automated parking searches or at least partially automated parking.Alternatively or additionally, the detection of the usable parking space can also serve to provide a driver of the motor vehicle with an indication of the usable parking space during manual control of the vehicle, for example, via a display, and / or to provide information regarding the usable parking space via an external device, for example, via a website and / or an app and / or to the navigation systems of other motor vehicles. After an open space has been identified as a usable parking space, occupancy statistics can also be determined and provided, for example, or similar.

[0034] The usage information may additionally indicate which parking direction was predominantly used by the other motor vehicles when parking in the parking area during historical use, whereby, on the one hand, the fulfillment of the statistical condition and / or a further exclusion condition may depend on the parking direction and / or whereby, on the other hand, the parking direction may be output to a driver of the motor vehicle via an output device of the motor vehicle and / or may be taken into account in the context of the at least partially automated guidance of the motor vehicle onto the parking area.

[0035] In particular, a distinction can be made between parallel parking, perpendicular parking, and diagonal parking. These parking directions can be differentiated, for example, depending on the angle of the longitudinal direction of the respective parked additional motor vehicle relative to the road. For example, it can be checked whether this angle lies within an angle interval assigned to the respective parking direction.

[0036] In addition, particularly in the case of perpendicular and / or diagonal parking, a distinction can be made between forward parking, where the front of the vehicle is facing away from the road, and reverse parking, where the front of the vehicle is facing the road. Taking the parking direction into account as part of the statistical condition or the further exclusion condition can be useful, as different parking directions require different lengths of open space parallel to the road. By outputting the parking direction to a driver, for example via a vehicle display or acoustically via a loudspeaker, the driver can be informed of a preferred parking direction in unclear parking situations. At least partially automated guidance can be implemented more easily and robustly if a suitable parking direction is already known in advance.

[0037] In addition to the method according to the invention, the invention relates to a motor vehicle comprising a sensor system and a processing device, wherein the motor vehicle is configured to participate in the method according to the invention as the motor vehicle and / or the further motor vehicle, wherein the processing device is configured to detect parking spaces usable by the motor vehicle depending on the sensor data from the sensor system. Thus, the method according to the invention can be implemented using a motor vehicle according to the invention. Irrespective of this, the features explained for the method according to the invention, with the advantages mentioned therein, can be transferred to the motor vehicle according to the invention, and vice versa.

[0038] Furthermore, the invention relates to a data processing device that is configured, on the one hand, to participate in the method according to the invention as the vehicle-external device, and / or, on the other hand, to provide the usage information and / or the advance information to the motor vehicle as the or an external vehicle-external device in the method according to the invention. The features explained for the method according to the invention, in particular with regard to the determination and / or provision of the advance information and / or the usage information, can be transferred to the data processing device according to the invention with the advantages mentioned therein, and vice versa. Further advantages and details of the invention emerge from the following exemplary embodiments and the associated drawings. These schematically show:

[0039] Fig. 1 shows a traffic situation in which an embodiment of the motor vehicle according to the invention and an embodiment of the data processing devices according to the invention are used to carry out an embodiment of the method according to the invention for detecting at least one usable parking space,

[0040] Fig. 2 is a flow chart of the method performed in Fig. 1, and

[0041] Fig. 3 the determination of prior information on the basis of which the

[0042] The usage information evaluated using the methods shown in Fig. 1 and 2 is determined.

[0043] Fig. 1 shows a traffic situation in which a motor vehicle 2 is guided along a road 5, wherein at least one parking area 1 usable by the motor vehicle 2 is to be detected.

[0044] In the following, it is assumed by way of example that a highly automated or fully automated parking is to take place, in which the motor vehicle 2 is guided along the road 5 without intervention by the driver by the assistance system 39 implemented by the processing device 41 of the motor vehicle 2 by suitable control of at least one actuator 40 of the motor vehicle 2 until a suitable parking space 1 is found, and is then automatically parked there.

[0045] However, the steps explained for detecting the usable parking space 1 can also be used, for example, to inform a manually operating driver of the motor vehicle 2 about a suitable parking space 1, for example by outputting a corresponding notification via the output device 33 of the motor vehicle 2, which can be a display, for example. The operation of the motor vehicle 2 or the method implemented for detecting at least one usable parking space 1 is explained below with additional reference to Fig. 2, which shows a flowchart of an exemplary implementation of such a method.

[0046] In step S1, sensor data 3 are first acquired by a sensor system 4 of the motor vehicle 2. These data relate to both a potential parking area 6 adjacent to the road 5, for example, a parkable shoulder, and a background area 18 located on a side of the potential parking area facing away from the road. In the example, the sensor system 4 comprises several cameras 44, which provide the partial data 46 of the sensor data 3, and several ultrasonic sensors 45, which provide the partial data 47 of the sensor data 3.

[0047] In step S2, the sensor data 3 are evaluated by the processing device 41 in order to identify existing open spaces 7, 8 within the potential parking area 6, the geometric dimensions 7, 8 of which, in particular their length, allow the motor vehicle 2 to park in the respective open space 7, 8. For this purpose, the occupancy of the potential parking area 6 can be detected while driving past, in particular via the ultrasonic sensors 45 provided on the side of the potential parking area 6 on the motor vehicle 2, so that it may be sufficient, for example, to evaluate the partial data 46 of the sensor data 3. Additionally or alternatively, image data from the cameras 44 can also be used to determine the open spaces 7, 8.

[0048] In the example shown in Fig. 1, only the open space 8 is actually a usable parking space 1, since the open space 7 is part of a driveway 19, namely a driveway. However, corresponding driveways are often not explicitly marked or signposted, which is why in the example shown in Fig. 2, two separate exclusion conditions 11, 12 are used to exclude open spaces that are not actually suitable for parking the motor vehicle 2. Actual parking, or information to the driver that a usable parking space 1 has been detected, only occurs in step S10 if both exclusion conditions 11, 12 are not met.

[0049] The first exclusion condition 11 is a statistical condition 13, the fulfillment of which depends on usage information regarding a historical use of the respective open space 7, 8 as a parking area by other motor vehicles 15, 16. In order to obtain such usage information 27, the motor vehicle 1 first transmits the position 27 of the respective open space 7, 8 to a vehicle-external device 24 by means of a communication device 42 in step S3. The position 27 of the respective open space 7, 8 can be determined based on a position of the motor vehicle 2, which can be determined, for example, via a GPS sensor (not shown), and the relative position of the respective open space 7, 8 to the motor vehicle 2 determined from the sensor data 3.

[0050] The vehicle-external device 24 can be implemented, for example, by a suitably programmed data processing device 42 and aggregates, as will be explained later with reference to Fig. 3 using an example, the parking behavior of a plurality of further motor vehicles 15, 16 at a plurality of positions 27, 28, for example in the form of a map 26, as is schematically shown in Fig. 3.

[0051] The map 26 specifies respective position information 29, 30 for different positions 27, 28. Based on the received position 27 of the respective open space 7, 8, the vehicle-external device 24 transmits the position information 29 associated with the received position 27 to the motor vehicle 2 as usage information 14 in step S3.

[0052] In the example, the respective position information 29, 30 and thus also the usage information 27 includes, on the one hand, a frequency 31, either for how often the other motor vehicles 15, 16 have parked on the respective open space 7, 8 after the respective open space 7, 8 has been classified as a usable parking space 1 by a driver assistance system 17 of the respective other motor vehicle 15, 16, or for how often such parking has been rejected.

[0053] On the other hand, the position information 29, 30 or the usage information in the example includes a parking direction 32, which indicates the direction in which the other motor vehicles 15, 16 were predominantly parked on the respective open space 7, 8 or at the respective position 27. The parking direction 32 can, for example, indicate whether parallel parking, diagonal parking, or perpendicular parking, or whether forward or reverse parking, was predominantly carried out.

[0054] The parking direction 32 can be used in particular in step S10 for automated parking, but can also be used, for example, to take into account the different required dimensions for the different parking directions 32 and, for example, to reject open spaces 7, 8 that are too short for parallel parking if the respective open space 7, 8 is predominantly used for such parking according to the usage information 14.

[0055] In step S5, the statistical condition 13 is then evaluated, which in the example is to be fulfilled if the frequency 31 for the parking of the other motor vehicles 15, 16 at position 27 of the respective open space 7, 8 falls below a predetermined threshold, or additionally or alternatively, if the frequency of refusal of such parking exceeds a threshold. In other words, the statistical condition 13 evaluates whether the other motor vehicles 15, 16 or their drivers tend to use the respective open space 7, 8 as a parking space or whether they rarely or never use the respective open space 7, 8 for parking, even after a suggestion from a driver assistance system.In the second case, this indicates that the open space 7, 8 is actually not suitable or at least not well suited for parking, whereby the detection of this open space 7, 8 by the driver assistance system 17 of the other motor vehicles 15, 16 appears to represent a false detection, whereby the corresponding open space 7, 8 is discarded in step S 11 and is not detected as a usable parking space if the statistical condition 13 is fulfilled.

[0056] In a further development of the method explained, the frequency 31 provided to the motor vehicle 2 could, for example, be time-dependent, for example if the open spaces 7, 8 are used for parking by the other motor vehicles 15, 16 with different frequencies at different times of day, and / or it could be taken into account how long each of the other motor vehicles 15, 16 has been parked at the respective position 27. This makes it possible, for example, to take into account restrictions on the standing time or temporary parking bans applicable to the respective open space 7, 8.

[0057] If the statistical condition 13 is not met in step S5, the respective open space 7, 8 is potentially usable as a parking space and thus not yet excluded. In the example shown in Fig. 1 for parking space 7, this may be the case, for example, if the entrance there is already sufficiently well detected by the driver assistance systems 17 of the other motor vehicles 15, 16 that there are no sufficiently good statistics available to exclude the open space 7 based solely on the usage information 14.

[0058] Therefore, the following steps serve to additionally evaluate the environmental condition 17, which should be met when certain indicators in the vicinity of the respective open space 7, 8 indicate that it is not suitable for parking. The environmental condition 17 used depends specifically on whether the sensor system 4 of the motor vehicle 2 detects a motor vehicle 34 within a background area 18 arranged on the side of the respective open space 7, 8 facing away from the road 5 and / or whether a building 20 or vegetation 21 is detected that is classified as delimiting a roadway 19. As already explained in the general section, an entrance, an intersection, or the like can be easily recognized based on the features mentioned, even if the roadway 19 itself cannot be easily recognized, for example, in the image data from the cameras 44.

[0059] In principle, it would also be possible, within the framework of environmental condition 17 or at least one further exclusion condition, to consider other indicators of the unsuitability of the respective open space 7, 8 for parking, such as traffic signs or markings indicating a parking ban. However, the consideration of traffic signs and road markings, or even the use of map data that explicitly describe parking bans or similar, will not be discussed further in this example.

[0060] In preparation for the evaluation of the environmental condition 17 in step S9, the sensor data 3, in particular the partial data 47 of the cameras 44, are first evaluated in step S6 in order to detect a potentially present building 20, vegetation 21, and / or a potentially present motor vehicle 34 in the respective background area 18. The detection of various objects in image data is well known from other application areas of the prior art and will therefore not be explained in detail.

[0061] In particular, pillars, a gate, hedge sections, planted beds, or the like can be detected as development 20 or vegetation 21 that could be relevant for delimiting a driveway 19. In the example, a planted bed and pillars, for example of a wall or a fence, are located as vegetation 21 in the background area 18 as development 20. And to determine whether these objects are likely to delimit a driveway 19, an additional condition 35 is evaluated in step S7, which is fulfilled if the distances 36, 37 of the pillars or planted beds shown in Fig. 1 lie within a predetermined range. The predetermined range is selected such that the minimum width is somewhat larger than the typical width 43 of motor vehicles, which is shown as an example for the detected motor vehicle 34, and does not exceed this by too much, for example by less than a factor of 4 or 6.This makes it possible, for example, to distinguish between a roadway 19 bordered by buildings 20 or vegetation 21 and other buildings or vegetation.

[0062] In addition, further indications that vegetation or buildings limit a roadway can be taken into account, such as signs indicating maximum permissible heights and / or widths for a vehicle passing through a narrow section.

[0063] Since a stationary detected motor vehicle 34 is a particularly robust indication that the respective open space 7, 8 is located in the area of ​​a driveway, in the example in step S8 a movement state 38 of the detected motor vehicle 34 is additionally determined by evaluating the sensor data 3.

[0064] In step S9, the environmental condition 17 can then be evaluated as exclusion condition 12. In particular, this can be fulfilled if, in step S7, a detected structure 20 or vegetation 21 was classified as delimiting a roadway 19 and / or if, in step S8, it was determined that a detected motor vehicle 34 is located in the background area 18.

[0065] If the environmental condition 17 is met, the respective open space 7, 8 is discarded in step S11 and thus not detected as a usable parking space 1.

[0066] If the environmental condition 17 is not met, however, the method continues in step S10, in which the assistance system 39 of the motor vehicle 2 controls at least one actuator 40 of the motor vehicle 2 in order to carry out automated driving interventions 48, which serve to guide the motor vehicle 2 into the remaining open space 8, i.e., into the detected parking space 1. Additionally or alternatively, it is possible in step S10, for example, to output an indication of the presence of the usable parking space 1 to a driver of the motor vehicle 2 via the output device 33. During automated parking or the indication, the parking direction 32 previously received as part of the usage information 14 can be taken into account.

[0067] In the exemplary embodiment shown in Fig. 1, the described procedure would, for example, result in the motor vehicle 2 initially being guided along the road 5 past the open space 7, which is recognized as such but not detected as a usable parking space 1 because it is rejected due to the exclusion condition 11 or 12. Thus, the journey of the motor vehicle 2 is continued to the open space 8, which in the example is detected as a usable parking space 1, whereby automated or assisted parking or a notification to the driver can take place there.

[0068] As explained above, the usage information 27 is intended to describe information regarding a historical use of the respective open space 7, 8 as a parking area by other motor vehicles 15, 16. A possible approach for determining such usage information is explained below with additional reference to Fig. 3.

[0069] A plurality of additional motor vehicles 15, 16, for example, the entire vehicle fleet of at least one manufacturer, are used, each implementing a driver assistance system 17 for detecting usable parking spaces. The driver assistance system can, in particular, implement the method for detecting usable parking spaces explained above.

[0070] After a usable parking space 1 is detected by the respective additional motor vehicle 15, 16, it determines whether the parking space is actually being used for parking or whether such parking is being rejected by a driver, and relevant parking information 22, 23 is transmitted to the vehicle-external device 24. This information can also additionally describe the parking direction used.

[0071] The vehicle-external device 24 determines for various positions 27, 28 how frequently other motor vehicles 15, 16 actually park there when a usable parking space is detected at the respective position 27, 28. This frequency and, optionally, the predominantly used parking direction can then be stored as position information 29, 30 assigned to the respective position 27, 28, in particular as a map 26, which can form preliminary information 25 of the usage information 27. As explained above, the usage information 27 can then be extracted from such a map 26 depending on the position 27 of an open space 7, 8 to be assessed.

Claims

PATENT CLAIMS:

1. A method for detecting at least one usable parking area (1) by a motor vehicle (2), wherein sensor data (3) of a sensor system (4) of the motor vehicle (1), which relate to a potential parking area (6) arranged adjacent to a road (5) with traffic, are recorded and evaluated in order to identify at least one open space (7, 8) within the potential parking area (6), the geometric dimensions (9, 10) of which allow parking of the motor vehicle (2) on the open space (7, 8), wherein the respective open space (7, 8) is classified as a usable parking area (1) only if no exclusion condition is met for the respective open space (7, 8), characterized in that as an exclusion condition (11, 12) at least - a statistical condition (13), the fulfilment of which depends on usage information (14) relating to a historical use of the open space (7, 8) as a parking area (1) by other motor vehicles (15, 16), and / or - an environmental condition (17), the fulfilment of which depends on whether a detected motor vehicle (34) and / or a building (20) and / or vegetation (21) classified as delimiting a roadway (19) is detected by the sensor system (4) within a background area (18) arranged on the side of the open space (7, 8) facing away from the road (5), is evaluated.

2. Method according to claim 1, wherein at least the statistical condition (13) is evaluated as exclusion condition (11, 12), characterized in that the further motor vehicles (15, 16) each implement a driver assistance system (17) by which usable parking areas (1) are detected, wherein the respective further motor vehicle (15, 16) after each detection of a respective usable parking area (1) generates parking information (22, 23) which indicates whether the further motor vehicle (15, 16) is actually parked on this parking area (1 ) and / or parking on this parking area (1 ) is refused by the driver of the further vehicle (15, 16), and is transmitted to a vehicle-external device (24), wherein the vehicle-external device (24) determines or updates the usage information (14) or preliminary information (25) on the basis of which the usage information (14) is determined on the basis of the parking information (22, 23) received from the further motor vehicles (15, 16).

3. Method according to claim 2, characterized in that the preliminary information (25) as a map (26) specifies position information (29, 30) for different positions (27, 28) on the map (26), wherein the respective position information (29, 30) describes how often, after a usable parking area (1) at the respective position (27, 28) has been determined, a respective one of the further motor vehicles (15, 16) has actually parked at this position (27, 28) and / or how often parking there has been refused by the driver of the respective further motor vehicle (15, 16), wherein the position information (29, 30) for the position (27) of the determined open space (7, 8) is used as usage information (14).

4. Method according to one of the preceding claims, wherein at least the environmental condition (17) is evaluated as exclusion condition (11, 12), characterized in that always or when an additional condition (35) is fulfilled, pillars and / or a gate and / or a façade opening are classified as buildings (20) delimiting the driveway (19) and / or hedge sections and / or planted beds are classified as vegetation (21) delimiting the driveway (19).

5. Method according to claim 4, characterized in that that the additional condition (35) evaluates the distance (36, 37) of the pillars or hedge sections or flowerbeds from each other or the width of the gate or the façade opening and / or a height extension of the gate or the façade opening and / or the height of a lower edge of the façade opening.

6. Method according to one of the preceding claims, wherein at least the ambient condition (17) is evaluated as an exclusion condition (11, 12), characterized in that upon detection of a detected motor vehicle (34) in the background area, a movement state (38) of the detected motor vehicle (34) is determined by evaluating the sensor data (14), wherein the fulfillment of the ambient condition (17) depends on the movement state (38).

7. Method according to one of the preceding claims, characterized in that after the detection of a usable parking space (1) or several of the usable parking spaces (1) by an assistance system (39) of the motor vehicle (2), at least one actuator (40) of the motor vehicle (2) is controlled in order to carry out automated longitudinal and / or transverse driving interventions (48), whereby the motor vehicle is guided at least partially automatically to the usable parking space (1) or a selected one of the usable parking spaces (1).

8. Method according to one of the preceding claims, characterized in that the usage information (27) additionally indicates which parking direction (32) was predominantly used by the other motor vehicles (15, 16) during the historical usage when parking on the parking area (1), wherein on the one hand the fulfillment of the statistical condition (14) and / or a further exclusion condition depends on the parking direction (32) and / or wherein on the other hand the parking direction (32) is output to a driver of the motor vehicle (2) via an output device (33) of the motor vehicle (2) and / or is taken into account within the framework of the at least partially automated guidance of the motor vehicle (2) onto the parking area (1).

9. Motor vehicle, comprising a sensor system and a processing device (41), characterized in that the motor vehicle (2, 15, 16) is set up to participate in the method according to one of the preceding claims as the motor vehicle (2) and / or the further motor vehicle (15, 16), wherein the processing device (41) is set up to detect parking spaces (1) that can be used by the motor vehicle (2) as a function of the sensor data (3) of the sensor system (4).

10. Data processing device, characterized in that it is set up, on the one hand, to participate in the method according to claim 2 or 3 as the vehicle-external device (24), and / or, on the other hand, is set up to provide the usage information (14) and / or the advance information (25) to the motor vehicle (2) in a method according to one of claims 1 to 8 as the or a vehicle-external device (24).

Images