DE102021207105A1 - Method for operating an assistance system of a vehicle - Google Patents

Abstract

The invention relates to a method for operating an assistance system (50) of a vehicle (10), comprising the steps: detecting a vehicle environment (100) of the vehicle (10), displaying an image (101) of the vehicle environment (100) using a display device (20 ), Defining a target position (2) of the vehicle (10) in the image (101) of the vehicle environment (100) based on a user input (4), which can be entered by means of an input unit (21), detecting objects (3) in the Vehicle environment (100), and automatic correction (5) of the target position (2) based on the objects (3) detected in the vehicle environment (100), such that an overlapping of the target position (2) with the objects (3) is avoided.

Description

State of the art

The present invention relates to a method for operating an assistance system in a vehicle and an assistance system in a vehicle.

Assistance systems of vehicles are known, which enable the vehicle to park or leave a parking space independently in parking spaces previously measured, for example by means of an environment sensor system. Furthermore, it is known to show a target position on a display in the case of driving assistance, it being possible for the target position to be shifted and/or rotated via an input unit. Such a method for driving support is in EP 1 886 094 B1 shown.

Disclosure of Invention

• - Erfassen einer Fahrzeugumgebung des Fahrzeugs,
• - Darstellen eines Abbilds der erfassten Fahrzeugumgebung mittels einer Anzeigevorrichtung,
• - Festlegen einer Zielposition des Fahrzeugs in dem Abbild der Fahrzeugumgebung basierend auf einer Benutzereingabe, wobei die Benutzereingabe mittels einer Eingabeeinheit eingebbar ist,
• - Erfassen von Objekten in der Fahrzeugumgebung, und
• - automatisches Korrigieren der Zielposition basierend auf den in der Fahrzeugumgebung erfassten Objekten. Das automatische Korrigieren erfolgt dabei derart, dass ein Überschneiden der Zielposition mit erfassten Objekten vermieden wird.

In contrast, the method according to the invention with the features of claim 1 is characterized by a particularly reliable provision of target positions, which can serve, for example, as a basis for autonomous parking maneuvers, with a high level of user comfort. This is achieved by a method for operating an assistance system of a vehicle, which includes the steps:

• - detecting a vehicle environment of the vehicle,
• - Representation of an image of the detected vehicle environment by means of a display device,
• - Defining a target position of the vehicle in the image of the vehicle environment based on a user input, wherein the user input can be entered using an input unit,
• - Detection of objects in the vehicle environment, and
• - Automatic correction of the target position based on the objects detected in the vehicle environment. The automatic correction takes place in such a way that an overlapping of the target position with detected objects is avoided.

In other words, in the method, the vehicle surroundings of the vehicle are recorded and displayed virtually using the image with the display device. The user, for example the driver of the vehicle, can then specify a target position by user input using the input unit.

For example, the target position can represent a location to which the vehicle is to be driven autonomously. The target position advantageously serves to mark a parking position at which the vehicle is to be parked, in particular by means of an autonomous parking maneuver, and to define the dimensions of the parking space. The method then corrects the target position selected manually by the user in such a way that the target position does not overlap with objects that have been recognized in the vehicle environment.

An area within the vehicle surroundings that is at least large enough for the vehicle to be parked on it is regarded as the target position. The target position preferably has at least one size that corresponds to a minimum external dimension of the vehicle. The target position is particularly preferably larger than the minimum external dimensions of the vehicle, in particular in which case a parking position at which the vehicle is to be parked can be automatically selected within the target position. It is particularly advantageous if the target position includes a minimum external dimension of the vehicle plus a maneuvering distance of, for example, 0.8 m, which is required for a parking maneuver.

The method thus offers the advantage that a corrected target position is automatically provided, which is adapted in such a way that a collision with obstacles is prevented and a high quality of the target alignment can be achieved. It is particularly advantageous that a high level of user comfort is made possible, since the user can only roughly preselect the target position using the input device, with the precise alignment of the target position then taking place through the automatic correction in such a way that a target position reliably free of obstacles is provided. This simplification saves time, which is often a decisive factor when parking. The method also allows a gap to be selected when the vehicle is in motion, e.g. by the passenger. Another advantage is that the maneuver can be planned with a minimum number of moves by adapting the parking space dimensions to the actual conditions. Without this adjustment, parking would have to take place in as space-saving a manner as possible, which would lead to a high number of moves even if there were actually a lot of space available for maneuvering for parking.

The dependent claims relate to preferred developments of the invention.

The automatic correction preferably takes place by means of the smallest possible change in position of the target position, which was determined based on the user input via the input unit. In other words, during the automatic correction, the target position specified by the user is moved to the next possible free space in the image of the vehicle environment, which has a size that corresponds at least to the size of the target position. The automatic correction can thus be viewed as a kind of "magnetic snapping" of the manually defined target position to the next possible free space in the image of the vehicle environment. A particularly high level of user comfort of the method can thereby be made possible, since the user's original wish with regard to the placement of the target position is taken into account in the best possible way.

The detection of objects particularly preferably includes an identification of obstacles and/or roadway boundaries and/or roadway markings. The identification can preferably be carried out by means of an object classification, which in particular can be part of an environment sensor system. As a result, objects that are particularly relevant for parking maneuvers can be recognized in order, for example, to provide improved options for aligning the target position using the identified objects. In particular, irrelevant objects, such as specific roadway boundaries and/or roadway markings, can also be identified. For example, objects can be identified as irrelevant objects if an overlap of the target position with such an object is permissible, which can be the case, for example, with parking areas that overlap with lane markings or curbs or the like. For example, such an irrelevant boundary can also be identified as a virtual boundary, with the method explicitly allowing the target position to overlap with such virtual boundaries. This can prevent recognized markings, such as parking area markings, from being unnecessarily driven around during an autonomous parking maneuver.

Preferably, the target position in the image of the vehicle environment, which is displayed on the display device, can be rotated and/or moved by means of the user input, which can be entered via the input unit, in particular in one plane of the image. That is, the user can manually rotate and/or move the target position shown in the image to easily and conveniently set a desired placement of the target position.

The automatic correction also preferably includes a rotation of the target position, in particular in a plane of the image, in order to correct a misalignment of the target position. The rotation preferably takes place based on identified lane boundaries and/or lane markings, particularly preferably in such a way that at least one boundary of the corrected target position is aligned parallel to at least one lane boundary and/or lane marking. Alternatively or additionally preferably, the automatic correction includes a shift in the target position in such a way as to correct a distance between the target position and at least one object detected in the vehicle environment. The distance is preferably corrected based on a predefined minimum distance. As a result, an optimal automatic alignment of the corrected target position can be made possible, as a result of which, for example, optimal utilization of available parking spaces and/or sufficient space for boarding and alighting is ensured.

The step of detecting objects in the area surrounding the vehicle is preferably carried out after the step of determining the target position. This allows particularly convenient operation for the user since, for example, there is no need to wait for the environment detection to be completed.

Objects in the surroundings of the vehicle are particularly preferably recognized exclusively within a predefined surrounding area, which surrounds the target position manually defined via the input unit. The surrounding area preferably has a width of at least 0.2 m, in particular a maximum of 2 m, particularly preferably a maximum of 1 m. In particular, the width is measured in the radial direction starting from a center of the target position and from an outer boundary of the target position. As a result, the method can be operated particularly efficiently, since only a small area of the vehicle's surroundings has to be scanned for objects, which allows particularly resource-saving operation.

• - Ermitteln von zumindest einer freien Parkfläche in der Fahrzeugumgebung, und
• - Anzeigen einer Initial-Zielposition mittels der Anzeigevorrichtung, wobei die Initial-Zielposition innerhalb der freien Parkfläche liegt.

More preferably, the step of recognizing objects in the area surrounding the vehicle is carried out before the step of defining the target position. The method also includes the following steps, which are carried out before the step of determining the target position:

• - determining at least one free parking space in the vehicle environment, and
• - Displaying an initial target position by means of the display device, the initial target position being within the free parking area.

An area of the vehicle environment, or the image of the vehicle environment, is regarded as a free parking area in which there is no object and which has a size that corresponds at least to the size of the target position, in particular plus a minimally required maneuvering space. In other words, before the target position is set manually using the user input, a free initial target position is determined and displayed in the image using the display device. The initial target position can thus be viewed as a suggestion to the user, which the user can select, for example, without the user having to worry about placing the target position.

• - Ermitteln einer weiteren Benutzereingabe, und
• - Ändern der korrigierten Zielposition basierend auf der weiteren Benutzereingabe.

The method preferably also includes the following steps, which are carried out after the automatic correction step:

• - detecting further user input, and
• - Changing corrected target position based on further user input.

The change can thus be viewed as overwriting the automatically corrected target position. This means that in this case the user has the option of changing the automatically corrected target position again. For example, inadvertently incorrect inputs can be corrected in the previous setting, and/or an incorrect automatic correction of the target position can be changed.

• - Ermitteln einer an eine Fahrbahn angrenzende freie Parkfläche in der Fahrzeugumgebung. Der Schritt des automatischen Korrigierens umfasst dabei zusätzlich den Schritt:
• - Ausrichten der Zielposition längs oder quer zur Fahrbahn in Abhängigkeit einer Ausrichtung der ermittelten freien Parkfläche.

The method particularly preferably also includes the step:

• - Determining a free parking area adjacent to a lane in the vehicle environment. The step of automatic correction also includes the step:
• - Alignment of the target position along or across the road depending on an alignment of the determined free parking space.

In other words, the target position set manually by the user is automatically adapted to a parking space arranged along or perpendicular to the roadway. A particularly convenient selection of a parking position can thus be made possible, it being possible to avoid the user having to turn the target position manually.

The target position is preferably displayed as a rectangular frame or as a vehicle contour in the image of the vehicle surroundings by means of the display device. The rectangular frame or the vehicle contour preferably has a dimension which corresponds to at least one dimension of the vehicle. The dimensions of the rectangular frame or the vehicle contour are particularly preferably chosen in such a way that there is enough space available for a parking maneuver. As a result, a particularly simple and clear display can be provided. Alternatively, if the specified target position is greater than the vehicle dimension, it is shown as a rectangle with a vehicle shown inside the rectangle, which makes it clear that the driver also defines the minimum maneuvering space required for the maneuver.

The method is preferably carried out while the vehicle is driving. Because the user only has to roughly position the target position and the system corrects the target position independently, simple operation by the driver is also possible while driving. As a result, a particularly high level of user comfort can be provided for the user.

• - autonomes Einparken des Fahrzeugs in eine innerhalb der korrigierten Zielposition liegende Position in der Fahrzeugumgebung. Dadurch kann ein besonders komfortables und zuverlässig genaues autonomes Einparkmanöver des Fahrzeugs ermöglicht werden. Je nach Größe der korrigierten Zielposition kann das Fahrzeug dabei automatisch innerhalb der Zielposition positioniert werden, beispielsweise um optimale Abstände zu den umliegenden Objekten einzuhalten. Alternativ, wenn die Zielposition und Fahrzeugabmessung gesagt oder knapp aufeinander abgestimmt sind, kann das Fahrzeug auch exakt mittig in eine der Zielposition entsprechende Positionen der Fahrzeugumgebung eingeparkt werden.

More preferably, the method also includes the step:

• - Autonomous parking of the vehicle in a position lying within the corrected target position in the vehicle environment. A particularly comfortable and reliably accurate autonomous parking maneuver of the vehicle can thereby be made possible. Depending on the size of the corrected target position, the vehicle can be automatically positioned within the target position, for example in order to maintain an optimal distance from the surrounding objects. Alternatively, if the target position and vehicle dimensions are stated or closely matched, the vehicle can also be parked exactly in the center of a position in the vehicle environment that corresponds to the target position.

Advantageously, the vehicle is parked autonomously up to a maximum speed of 20 km/h, preferably a maximum of 10 km/h. That is, the autonomous parking maneuver is performed at a low speed of the vehicle.

• - Speichern der korrigierten Zielposition in einer Umgebungskarte. Vorzugsweise werden dabei zusätzlich Umfelddaten, wie beispielsweise Positionsdaten, insbesondere GPS-Daten und Daten der Umfeldsensorik, die eine genaue Lokalisierung des Fahrzeugs im Umfeld ermöglichen, mit gespeichert. Dadurch kann die korrigierte Zielposition beispielsweise bei wiederkehrenden Fahrten im selben Bereich der Fahrzeugumgebung wiederverwendet werden, um ein besonders einfaches Assistenzsystem bereitzustellen.

Preferably, the method further includes the step:

• - Saving the corrected target position in an environment map. In this case, environmental data, such as position data, in particular GPS data and data from the environmental sensor system, are preferably used in addition Enable localization of the vehicle in the area, with saved. As a result, the corrected target position can be reused, for example, in the case of repeated journeys in the same area of the vehicle environment, in order to provide a particularly simple assistance system.

Furthermore, the invention leads to an assistance system of a vehicle, which is designed in particular for autonomous control of the vehicle. The assistance system includes a control device, an environment sensor system that is set up to detect a vehicle environment of the vehicle, a display device, and an input unit. The control device is set up to carry out the method described above. The environment sensor system can include at least one camera, for example, which is set up to capture images of the vehicle environment. The surroundings sensor system can preferably include further sensors which are set up to detect the vehicle surroundings and in particular objects in the vehicle surroundings.

The input unit preferably includes a touch-sensitive screen. In this case, the touch-sensitive screen particularly preferably also forms the display device. This makes a particularly simple and user-friendly assistance system possible. Alternatively or additionally, the input unit can include a joystick and/or an operating element of the vehicle.

character list

• 1 eine vereinfachte schematische Ansicht einer Parkplatzsituation, bei welcher ein Verfahren zum Betreiben eines Assistenzsystems eines Fahrzeugs gemäß einem bevorzugten Ausführungsbeispiel der Erfindung eingesetzt wird,
• 2 eine Durchführung des Verfahrens der 1,
• 3 Schritte einer Durchführung des Verfahrens der 1 bei einer weiteren Parkplatzsituation,
• 4 eine weitere Parkplatzsituation, bei welcher das Verfahren eingesetzt wird,
• 5 eine weitere Parkplatzsituation, bei welcher das Verfahren eingesetzt wird, und
• 6 einen weiteren Schritt des Verfahrens der 1 bei einer weiteren Parkplatzsituation.

The invention is described below using an exemplary embodiment in conjunction with the figures. In the figures, components that are functionally the same are each identified by the same reference symbols. It shows:

• 1 a simplified schematic view of a parking lot situation in which a method for operating an assistance system of a vehicle according to a preferred embodiment of the invention is used,
• 2 an implementation of the method of 1 ,
• 3 Steps of carrying out the method of 1 in the event of another parking situation,
• 4 another parking lot situation in which the method is used,
• 5 another parking lot situation in which the method is used, and
• 6 another step of the process of 1 in the event of another parking situation.

Preferred Embodiments of the Invention

1 shows a simplified schematic view of a parking lot situation in which a method for operating an assistance system 50 of a vehicle 10 according to a preferred exemplary embodiment of the invention is used. In 2 an implementation of the method is shown in detail.

Assistance system 50 includes (not shown in detail) a control device, a display device 20, an input unit 21, and an environment sensor system. A vehicle environment 100 of vehicle 10 can be detected by means of the environment sensor system. The vehicle 10 can preferably also include a location system, by means of which location data, for example GPS data, of the vehicle 10 in the surroundings 100 can be determined.

Autonomous parking maneuvers can be carried out using the assistance system 50 . In this case, vehicle 10 is moved to a target position 2 (cf. 2 ) parked. The method includes a combination of manual selection and automatic correction of the target position 2, as described in detail below.

In particular, the method is therefore suitable for being able to park the vehicle 10 autonomously in complex parking lot situations with a particularly high level of user comfort. 1 shows such a parking lot situation as an example. The vehicle 10 is still on a lane 30 and is to be parked autonomously in a parking lot 15, which is partially surrounded by objects 3, such as obstacles 31, for example in the form of walls.

First, the vehicle environment 100 of the vehicle 10 is detected, with objects 3, such as obstacles 31, roadway boundaries 32 and roadway markings 33 (cf. also 3 until 6 ) are recorded and identified. The detection can take place, for example, when the vehicle 10 is stationary or alternatively when it is driving past. An image 101 of vehicle surroundings 100 is then displayed using display device 20 (cf. 2 ).

Like in the 2 To recognize, the detected objects 3 can be imaged incompletely, for example with gaps 34, which can result from the fact that the objects 3 can partially cover each other during the detection.

The display device 20 is a touch-sensitive screen, so that the display device 20 can simultaneously serve as an input unit 21 for a haptic user input 4 .

The user can manually specify the target position 2 in the image 101 by means of his user input 4, which is in the form of a rectangular frame whose relative dimensions in relation to the vehicle surroundings 100 correspond to the dimensions of the vehicle 10 including a maneuvering distance. The user can use the user input 4 to move and rotate the target position 2 in the image 101, as in FIG 2 (a) implied.

The method also includes the step of an automatic correction 5, which in the 2 B) is shown. The target position defined by the user using his user input 4 is shifted and/or rotated in such a way as to avoid overlapping with the detected objects 3 in the image 101 .

The automatic correction 5 is carried out by means of the smallest possible change in position of the target position 2 specified by the user. In the parking lot situation shown, this means, for example, that the target position 2 automatically snaps to the next possible free space that is large enough after the user input 4 and is anchored there as a corrected target position 2′ and remains visible.

The autonomous parking maneuver into the corrected target position 2' preferably takes place after a subsequent optional confirmation by the user.

In the method, the possibility of overwriting the corrected target position 2′ can preferably also be provided. Here, after the automatic correction 5, the user can change the corrected target position 2' again by means of a new user input 4. For example, errors can be corrected in this way or deviating individual wishes of the user can be implemented.

3 shows an alternative parking lot situation, in which additional steps of the method are carried out. The parking lot situation is a free parking lot 15, which is arranged along the edge of a roadway 30 and between two other vehicles, which can be viewed as obstacles 31.

is shown in 3 the image 101 of the vehicle environment 100. The user has already set the target position 2 in the image 101 manually by means of his user input 4. Like in the 3 (a) As can be seen, the target position 2 is arranged in this case with a certain skewing and with a certain overlap with one of the obstacles 31 .

The automatic correction 5 additionally includes a rotation of the target position 2 in order to correct the misalignment. For this purpose, the target position 2 is aligned based on identified roadway boundaries 32 and roadway markings 33, in particular in such a way that the corrected target position 2' (cf. 3(c) ) is parallel to the lane boundary 32 and lane markings 33 . In addition, the corrected target position 2 ′ is arranged in the middle between the two obstacles 31 .

In order to enable the method to be carried out particularly efficiently, objects 3 in vehicle surroundings 100 are only detected within a predefined surrounding area 25, which surrounds manually specified target position 2 (cf. 3 (b) ). The surrounding area 25 is defined by a predefined maximum distance 26 of approximately 0.5 m inside and outside the boundary of the target position 2, respectively. In other words, the surrounding area 25 is defined by two rectangles lying one inside the other, with the rectangle of the target position 2 lying between these two rectangles.

In the 4 until 6 further exemplary parking lot situations are shown, on the basis of which further functionalities of the method are explained.

4 (a) shows intentional parking between two obstacles 31 in the form of vehicles that are parked parallel and next to one another in front of a lane boundary 32 . After the manual determination of the target position 2, which in the example shown overlaps with the left obstacle, the automatic correction 5 takes place in such a way that the corrected target position 2' is arranged in the middle between the two obstacles 31.

If only one obstacle 31, such as in the 4 (b) , is present, the automatic correction 5 takes place in such a way that the corrected target position 2 ′ is arranged at a predefined minimum distance 27 from the obstacle 31 in order, for example, to enable comfortable entry and exit from the vehicle 10 .

In 5 is the same parking situation as in the 4 (a) shown. Based on 5 a corrected longitudinal alignment of target position 2 is illustrated. 5 (a) shows a manually set target position 2, which is too far over the hind Nisse 31 protrudes and protrudes into an adjacent lane 30. The corrected target position 2 ′ is shifted in the direction of the lane boundary 32 by the automatic correction 5 in such a way that it does not intersect with the lane 30 .

5(b) shows a situation similar to that 2 B) and 3 (a) , wherein the manually specified target position 2 has an overlap with the lane boundary 32. The automatic correction 5 then arranges the corrected target position 2′ in such a way that it does not overlap with the lane boundary 32 .

Another aspect of the process is 6 shown. 6 shows doing similar to the 4 and 5 , vehicles or obstacles 31 parked transversely to a roadway 30. The target position 2 manually defined by the user is initially along the roadway 30, as in FIG 6 (a) shown, arranged. If it is determined that the free parking area 35 adjacent to the lane 30 is not large enough for parallel parking, or alternatively, for example, that parking is required perpendicular to the lane 30, automatic correction 5 automatically aligns the corrected target position 2' perpendicular to the lane Lane 30, as in 6 (b) shown. This means that the automatic correction 5 can include rotating the target position 2 depending on the orientation of the determined free parking area 35 .

QUOTES INCLUDED IN DESCRIPTION

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

Patent Literature Cited

• EP 1886094 B1 [0002]

Claims (16)

Method for operating an assistance system (50) of a vehicle (10), comprising the steps: - detecting a vehicle environment (100) of the vehicle (10), - Displaying an image (101) of the vehicle environment (100) by means of a display device (20), - Defining a target position (2) of the vehicle (10) in the image (101) of the vehicle environment (100) based on a user input (4), which can be entered using an input unit (21), - Detection of objects (3) in the vehicle environment (100), and - Automatic correction (5) of the target position (2) based on the objects (3) detected in the vehicle environment (100), such that an overlapping of the target position (2) with objects (3) is avoided. procedure after claim 1 , wherein the automatic correction (5) takes place by means of the smallest possible change in position of the target position (2) defined on the basis of the user input (4). Method according to one of the preceding claims, wherein the detection of objects (3) comprises an identification of obstacles (31) and/or roadway boundaries (32) and/or roadway markings (33). Method according to one of the preceding claims, wherein the target position (2) in the image (101) of the vehicle environment (100) can be rotated and/or moved by means of the user input (4). Method according to one of the preceding claims, - wherein the automatic correction (5) comprises a rotation of the target position (2) to correct a misalignment of the target position (2), and/or - wherein the automatic correction (5) includes a shift in the target position (2) such that a distance from the target position (2) to at least one object (3) detected in the vehicle environment (100), in particular based on a predefined minimum distance (27) to correct. Method according to one of the preceding claims, wherein the detection of objects (3) in the vehicle environment (100) is carried out after the target position (2) has been defined. procedure after claim 6 , wherein the detection of objects (3) in the vehicle environment (100) only for a predefined maximum area (25) surrounding the manually specified target position (2) is performed. Procedure according to one of Claims 1 until 5 , wherein the detection of objects (3) in the vehicle environment (100) before the determination of the target position (2) is carried out, and wherein the method further comprises the steps before determining the target position (2): - determining at least one free Parking area in the vehicle environment (100), and - displaying an initial target position (2) by means of the display device (20), the initial target position (2) being within the free parking area. Method according to one of the preceding claims, further comprising the steps, after the automatic correction (5): - determining a further user input (4), and - Changing corrected target position (2`) based on further user input (4). A method according to any one of the preceding claims, further comprising the step of: - Determining a free parking area (35) adjacent to a lane (30) in the vehicle environment (100), the automatic correction (5) additionally comprising the step: - Aligning the target position (2) along or across the roadway (30) depending on an alignment of the determined free parking area (35). Method according to one of the preceding claims, wherein the target position (2) is displayed as a rectangular frame in the image (101) of the vehicle surroundings (100), in particular wherein the rectangular frame has a dimension which is at least one external dimension of the vehicle (10), in particular plus a predetermined maneuvering space. Method according to one of the preceding claims, wherein the method is carried out while the vehicle (10) is traveling. A method according to any one of the preceding claims, further comprising the step of: - Autonomous parking of the vehicle (10) in a position within the corrected target position (2') in the vehicle environment (100). A method according to any one of the preceding claims, further comprising the step of: - Saving the corrected target position (2) in an environment map. Assistance system of a vehicle (10), in particular for autonomously controlling the vehicle (10), comprising: - a control device, - an environment sensor system for detecting a vehicle environment (100) of the vehicle (10), - a display device (20), and - an input unit (21), - wherein the control device is set up to carry out the method according to one of the preceding claims. assistance system claim 15 , wherein the input unit (21) comprises a touch-sensitive screen and/or a joystick and/or a control element of the vehicle (10).

Images