DE102015212581A1 - Driver assistance and driver assistance system - Google Patents

Abstract

The invention relates to a method for driver assistance, in which a driver is assisted in carrying out repetitive driving maneuvers. The method comprises a learning mode in which, starting from a start position (50), a trajectory (52) to a target position (22) is defined by a vehicle (1) being guided by the driver and the vehicle (1) along the trajectory (52) moves from the start position (50) to the target position (22). In this case, obstacles (40, 42) in the environment of the vehicle (1) are detected via environment sensors and entered into an environment map. Subsequently, in an application mode, a current vehicle position is determined using the environment sensors and the environment map, and the vehicle (1) is automatically or semi-automatically guided from the current vehicle position to the destination position (22). It is provided that the obstacles (40, 42) detected in the learning mode are subdivided by their height into fixed obstacles (40) and temporary obstacles (42). Further aspects of the invention relate to a computer program and a driver assistance system, which are set up to carry out the method.

Description

State of the art

The invention relates to a method for driver assistance, in which a driver is assisted in carrying out repetitive driving maneuvers. Further aspects of the invention relate to a driver assistance system and a computer program which are set up to carry out the method.

Modern vehicles are equipped with a driver assistance system to assist the driver of a vehicle in performing various driving maneuvers. Automatic and semi-automatic systems are known in the prior art. In automatic systems, the driving maneuver to be performed is performed automatically by the driver assistance system, both in terms of the longitudinal guidance and in terms of the lateral guidance of the vehicle. Here, the longitudinal guidance is understood to mean the acceleration or deceleration of the vehicle, and the term transverse guidance means the steering of the vehicle. In a semi-automatic system, the driver of the vehicle either performs the longitudinal guidance and the lateral guidance is taken over by the driver assistance system or the lateral guidance is performed by the driver of the vehicle and the longitudinal guidance is taken over by the driver assistance system.

Out DE 199 47 766 A1 a device for monitoring the surroundings of a parking vehicle is known. The device comprises video cameras and object recognition sensors, which are designed, for example, as ultrasonic sensors. The cameras and sensors are used to determine the size and distance of objects. The device further comprises a maneuvering unit which, upon detection of the corresponding environment, retrieves standard maneuvering maneuvers for the automatic execution of a maneuvering maneuver.

Out DE 10 2004 016 023 A1 a method for object classification is known. In this case, data of a remote sensor aligned to the side of the vehicle is received and analyzed, which type of objects may be present. In this case, a method is used for classifying an object type in which the size of an object is compared with a size threshold value. If the size is greater than the threshold, the object is classified as a stationary object. For example, radar, laser and ultrasound are used as remote sensors.

DE 10 2008 041 679 A1 describes a method for creating a map of the environment, wherein detected objects are entered into the map. Specific features such as position, relative velocity, extent, contour, size, shape or reflectance are captured. If an object is encountered repeatedly at a location, it is classified as a static object. If, on the other hand, an object unexpectedly first appears at a known location, it is classified as a dynamic object.

Disclosure of the invention

A driver assistance method is proposed in which a driver is assisted in performing repetitive driving maneuvers. In the method, a learning mode is provided in which, starting from a start position, a trajectory to a target position is defined by guiding a vehicle through the driver. The vehicle travels along the trajectory from the start position to the target position, where obstacles in the surroundings of the vehicle are recorded via environmental sensors and entered into an environment map. In a later performed application mode of the method, the current vehicle position is determined using the environment sensors and the previously created environment map. The vehicle is then automatically or semi-automatically guided from the current vehicle position to the target position. Furthermore, it is provided that the obstacles detected in the learning mode are subdivided according to their height into stationary obstacles and temporary obstacles.

The repetitive driving maneuvers are, for example, the parking in a garage, such as a single garage or underground parking. Furthermore, the repetitive driving maneuver can be, for example, parking at a parking space, at another parking space or at parking out of a garage, from a parking space or another parking space or, for example, passing through a narrow passage such as a narrow passage or a gate act. In this case, the driver assistance system is trained in the learning mode by the driver, during which the driver of the vehicle travels along the desired trajectory from the start position to the destination position during a learning run. The term trajectory is understood to mean a trajectory along which the vehicle moves during the driving maneuver. The vehicle is usually a motor vehicle, for example an automobile, which is operated with an internal combustion engine and / or an electric motor.

The vehicle includes environmental sensors that detect the surroundings of the vehicle can be. In particular, the environmental sensors detect obstacles in the surroundings of the vehicle. During the learning run, the environment sensors are used to create an environment map in which in particular the obstacles detected via the environmental sensors are entered. It is envisaged that a subdivision of the obstacles into stationary obstacles and temporary obstacles takes place. Stationary obstacles are immovable objects such as walls, curbs, street lamps, trees and in the case of underground garages also walls or pillars. Temporary obstacles are, for example, parked vehicles. Fixed or stationary obstacles always remain in a fixed location, whereas temporary obstacles can change their position or can be added or removed to the environment.

In the application mode of the proposed method for driver assistance, the current vehicle position is first determined using the environment sensors and the previously created environment map. Subsequently, the vehicle is guided along the trajectory, which was previously trained in the learning mode, to the target position. The leadership of the vehicle can be done semi-automatically or automatically. In an automatic guide both the longitudinal guidance, that is, the acceleration and deceleration of the vehicle, as well as the transverse guidance, so the steering of the vehicle made. The driver is responsible only for a monitoring function. In the case of a semi-automatic guiding, in the method either the longitudinal guidance is taken over, whereby the driver is responsible for the transverse guidance, or conversely the transverse guidance is taken over and the driver is responsible for the longitudinal guidance of the vehicle. The execution of the longitudinal guide and / or the transverse guide of the vehicle is preferably carried out by a driver assistance system of the vehicle.

Preferably, a first limit value is specified for the subdivision of the obstacles, obstacles having a height above the first limit value being classified as fixed obstacles. The first limit value is preferably selected in the range from 1.5 to 2.5 m, the first limit value being particularly preferably selected in the range from 1.6 to 2.0 m. For example, the first limit is 1.8 m. Typically, objects that are larger than the threshold are trees, signs, street lights or, in the context of underground car parks, walls or pillars. Typical moving objects such as passenger cars have a height that is below the first threshold.

Preferably, a second limit value is predetermined, wherein objects with a height below the second limit value are classified as stationary obstacles. If a first limit value has already been specified, then the second limit value is smaller than the first limit value. The second limit is preferably in the range of 0.1 to 0.5 m, more preferably in the range of 0.1 to 0.3 m. For example, the second limit is 0.2 m. Objects with a height below the second threshold are typically curbs and thus stationary.

Objects that have not been classified as stationary by a comparison with the first threshold and / or the second threshold are classified as temporary objects. Both the first and the second limit value are preferably specified in the classification, so that an interval can also be defined for the classification of an object as a temporary object with the second limit value as the lower limit and the first limit value as the upper limit.

In the learning mode of the method, the environment map is created based on information from the environment sensors, with detected obstacles are entered in this environment map. The environment map preferably stores information about the registered obstacles, wherein the information is selected from positions of the obstacle, height of the obstacle, contour of the obstacle, classification as a stationary or temporary obstacle, drive-over, drive-in or a combination of at least two of these information. In this case, overriding means that the obstacle can be run over safely by the vehicle, as is the case, for example, with a low curb or a threshold. With traversability is meant that the vehicle can drive safely under the obstacle, as is the case for example for transits.

In the application mode, the current vehicle position is determined with the aid of the environment sensors and the previously created environment map. The current vehicle position is preferably determined by a comparison of obstacles detected in the application mode with the environment sensors with the obstacles registered in the environment map, wherein only stationary obstacles are taken into account in the comparison or stationary obstacles are weighted more heavily than temporary obstacles in the comparison.

As a rule, in the application mode, the current vehicle position will deviate from the start position stored in the learning mode. Therefore, it is preferably provided that in the application mode, an offset between the current vehicle position and the start position is determined and a corrected trajectory is determined according to the offset. The vehicle will then travel along the corrected trajectory from the current one Vehicle position guided to the target position defined in the learning mode.

According to the invention, a computer program is also proposed according to which one of the methods described herein is performed when the computer program is executed on a programmable computer device. The computer program can be, for example, a module for implementing a driver assistance system or a subsystem thereof in a vehicle or an application for driver assistance functions, which can be executed, for example, on a smartphone or tablet. The computer program may be stored on a machine-readable storage medium, such as on a permanent or rewritable storage medium, or in association with a computing device or on a removable CD-ROM, DVD, Blu-ray Disc or a USB stick. Additionally or alternatively, the computer program may be provided for download on a computing device, such as a server, for example, via a data network such as the Internet or a communication such as a telephone line or a wireless connection.

According to the invention, a driver assistance system is furthermore provided, which comprises environment sensors for detecting obstacles in the surroundings of a vehicle, a control device and a memory unit. The driver assistance system is preferably designed and / or set up to carry out the methods described herein. Accordingly, the features described in the context of the method apply correspondingly to the driver assistance system and, conversely, the features described in the context of the driver assistance system apply correspondingly to the methods.

The function of the control unit or parts or units thereof can be implemented in the form of a computer program. In addition, it is conceivable that the control unit or parts or units thereof is designed as an application-specific integrated circuit (ASIC). Preferably, the control unit is further configured to communicate with other systems of the vehicle or is designed as part of a system connected to a vehicle, in particular a driver assistance system.

The memory unit of the driver assistance system is in particular designed to store information about obstacles detected in the learning mode and make them available again in connection with the application mode. The storage unit may be embodied, for example, as a separate unit or as part of another unit. In particular, the storage unit may be part of the control unit.

The environment sensors are designed, for example, as ultrasonic sensors, optical cameras, stereo cameras, radar, lidar or a combination of at least two of these sensor types. The environment sensors can also be associated with other systems of a vehicle, in particular further driver assistance systems of the vehicle.

Advantages of the invention

In the context of driver assistance systems that assist a driver in the execution of repetitive maneuvers, it is necessary to determine the current position of the vehicle in relation to a previously created environment map. The position determination is improved with the aid of the proposed method or the proposed driver assistance system, since a distinction is made between stationary and temporary obstacles during position determination with the aid of an environment map created in the learning mode of the method. In this case, stationary obstacles for determining position are particularly well suited, since it is ensured that their position and other properties such as height, contour and the like have not changed since the creation of the environment map in the learning mode. Advantageously, the classification can already take place in the learning mode of the method, so that the reliability of the position determination at any time is independent of the position and number of temporary obstacles.

In situations in which the vehicle environment is dominated by temporary objects during the learning mode, a position determination in the application mode is only possible by the method according to the invention, so that the availability of the driver assistance system is increased.

Brief description of the drawings

Embodiments of the invention are illustrated in the figures and explained in more detail in the following description.

Show it:

1 a schematic representation of a vehicle with a driver assistance system,

2 Parking in an underground car park with activated learning mode and

3 Parking in an underground car park with activated application mode.

Embodiments of the invention

In the following description of the embodiments of the invention, the same or similar elements are designated by the same reference numerals, wherein a repeated description of these elements is dispensed with in individual cases. The figures illustrate the subject matter of the invention only schematically.

In 1 is a vehicle 1 with a driver assistance system 10 shown. The driver assistance system 10 includes a controller 12 , which with environment sensors 18 and odometry sensors 15 communicates. The environment sensors 18 For example, they are configured as ultrasonic sensors, optical cameras, stereo cameras, radar, lidar or a combination of at least two of these sensor types. Here are the environmental sensors 18 designed and arranged so that these obstacles are around the vehicle 1 can detect and determine a height of the obstacles.

The control unit 12 of the driver assistance system 10 is also available with other systems of the vehicle 1 in conjunction, leaving the controller 12 For example, can also determine a steering wheel angle, to guide the vehicle 1 can generate a steering wheel angle and / or the vehicle 1 for example, can accelerate and decelerate. In the 1 These are connections with a connection to a steering wheel 17 and to a pedal 16 indicated.

In the method according to the invention, the driver assistance system is provided 10 into a learning mode. After activating the learning mode, the vehicle becomes 1 by his driver from a starting position 50 to a target position 22 along a trajectory 52 led, compare 2 , During the driving maneuver becomes the trajectory 52 recorded for this purpose, for example, odometry sensors 15 , the operation of pedals 16 and an impact of the steering wheel 17 be evaluated. Furthermore, it is continuously updated on the environment sensors 18 the environment of the vehicle 1 recorded, with an environment map of the environment of the vehicle 1 is created. Here are in the environment map in particular on the environment sensors 18 detected obstacles registered. Depending on the embodiment of the invention can be provided to enter each obstacle in the environment map. Alternatively, a filtering may be provided, with only obstacles are entered in the environment map that meet certain criteria.

The environment map can save to the obstacles more data, in particular one on the environment sensors 18 certain height of the respective obstacle is stored. During the learning run in learning mode, the environment map is continuously updated and the environment map is stored in a memory unit for later use 14 stored.

Has the driver of the vehicle 1 with his vehicle 1 the desired target position 22 reached, the learning mode is terminated. The trajectory 52 , along the vehicle 1 from the starting position 50 to the target position 22 was conducted, cf. 2 , is through the control unit 12 in the storage unit 14 stored.

The trajectory defined in the learning mode 52 can then be used in an application mode of the method, to this driving maneuver automatically by the vehicle 1 to let repeat. To do this, the driver of the vehicle drives 1 near the starting position 50 and moves the driver assistance system 10 in the application mode. The control unit 12 now determines a current vehicle position 56 with the help of environment sensors 18 and the previously created environment map, compare 3 , In particular, the positions of the obstacles entered in the environment map with those of the environmental sensors are used for this purpose 18 currently detected obstacles compared. It is advantageously provided that either exclusively classified as stationary obstacles are used for the comparison, or that are classified as stationary classified obstacles in the position determination higher. The classification of the obstacles into stationary obstacles and temporary obstacles takes place over the determined height of the obstacles, taking advantage of the fact that curbs and walls and pillars are typically fixed and easily distinguished by their size from other objects, such as other vehicles.

After determining the current position 56 of the vehicle 1 will be an offset from the current vehicle position 56 and the start position previously defined in learning mode 50 determined. Then, with the help of the determined offset and the trajectory defined in the learning mode 52 a corrected trajectory 54 determined, compare 3 , The vehicle 1 will now be along the corrected trajectory 54 from the current vehicle position 56 to the target position 22 guided. The vehicle is preferred 1 fully automatically guided, whereby both the longitudinal and the transverse guidance by the driver assistance system 10 is taken over. It is preferred during the guidance of the vehicle 1 the environment continues with the help of environment sensors 18 monitors, taking the lead of the vehicle 1 can be canceled or interrupted if a predetermined minimum distance between the vehicle 1 and falls below an obstacle.

In 2 is the parking of a vehicle 1 in an underground car park 30 with activated learning mode of the driver assistance system 10 shown. In the underground garage 30 There are a variety of parking spaces, some of which are parked vehicles 36 are occupied. Furthermore, the underground garage 30 walls 32 and pillars 34 on.

After activation of the learning mode of the driver assistance system 10 becomes the vehicle 1 by his driver along a trajectory 52 from the starting position 50 to the target position 22 guided. Here are the starting position 50 and the target position 22 to a reference position 20 between the wheels of the rear axle of the vehicle 1 based.

While the vehicle 1 along the trajectory 52 to the target position 22 is conducted via environment sensors 18 of the vehicle 1 created the environment map. It will be through the environment sensors 18 of the vehicle 1 obstacles 40 . 42 detected. The environment sensors detect this 18 the walls 32 the underground car park 30 , the buttresses 34 the underground car park 30 as well as the parked vehicles 36 , For a subdivision of the detected obstacles 40 . 42 in stationary obstacles 40 and temporary obstacles 42 will be the height of the obstacles 40 . 42 determined and compared with a first limit. For example, the first limit is set to 1.8 m. The height of a typical vehicle is less than 1.8 m, so the parked vehicles 36 as temporary obstacles 42 get ranked. The pillars 34 as well as the walls 32 the underground car park 30 are high and thus higher than the first limit. Thus, the walls become 32 and the pillars 34 each as stationary obstacles 40 classified. After reaching the target position 22 Both the trajectory 52 as well as the environment map saved.

In 3 is the underground car park again 30 out 2 but are located in the in 3 situation illustrated no parking vehicles 36 in the underground garage 30 , Furthermore, it can be seen that the current vehicle position 56 of the vehicle 1 from the start position defined during the learn trip 50 differs.

The maneuver learned in the learning mode is now to be repeated in the application mode, so that the vehicle 1 automatically to the destination position 22 to be led.

After activating the application mode of the method are via the environment sensors 18 of the vehicle 1 obstacles 40 . 42 in the vicinity of the vehicle 1 detected. In the in 3 the situation shown will be the walls 32 and the pillars 34 the underground car park 30 detected. These are determined by their height as fixed obstacles 40 classified. The stationary obstacles 40 be with the stationary obstacles stored in the environment map 40 compared, so that the current vehicle position 56 in relation to the previously stored stationary obstacles 40 determine. This has an advantageous effect that when creating the environment map a classification of obstacles in stationary obstacles 40 and temporary obstacles is possible. In this case, in an advantageous variant of the invention, a filtering can take place, in which only the stationary obstacles 40 be entered in the environment map, so that the in 3 illustrated situation with respect to the stationary obstacles 40 is identical to the in 2 and environment map created during the learning mode.

After determining the current vehicle position 56 will be an offset between the current vehicle position 56 and the start position defined in learning mode 50 certainly. This offset is then used to define the trajectory defined in the learning mode 52 correct, so that a corrected trajectory 54 is determined. The vehicle 1 is then along the corrected trajectory 54 to the target position 22 guided.

It is preferred while driving along the corrected trajectory 54 the drive over the environment sensors 18 supervised. It is provided, the leadership of the vehicle 1 to interrupt as soon as an approach to an obstacle falls below a predefinable safety distance 40 . 42 is recognized.

The invention is not limited to the embodiments described herein and the aspects highlighted therein. Rather, within the scope given by the claims a variety of modifications are possible, which are within the scope of expert action.

QUOTES INCLUDE IN THE DESCRIPTION

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

Cited patent literature

• DE 19947766 A1 [0003]
• DE 102004016023 A1 [0004]
• DE 102008041679 A1 [0005]

Claims (10)

Method for driver assistance, in which a driver is assisted in carrying out repetitive driving maneuvers, with a learning mode in which, starting from a starting position (FIG. 50 ) a trajectory ( 52 ) to a target position ( 22 ) is defined by a vehicle ( 1 ) is guided by the driver and the vehicle ( 1 ) along the trajectory ( 52 ) from the starting position ( 50 ) to the target position ( 22 ), whereby via environment sensors ( 18 ) Obstacles ( 40 . 42 ) in the vicinity of the vehicle ( 1 ) and entered in an environment map, and with an application mode in which a current vehicle position ( 56 ) using the environment sensors ( 18 ) and the environment map, and the vehicle ( 1 ) automatically or semi-automatically from the current vehicle position ( 56 ) to the target position ( 22 ), characterized in that the obstacles detected in the learning mode ( 40 . 42 ) by their height into stationary obstacles ( 40 ) and temporary obstacles ( 42 ). Method according to claim 1, characterized in that obstacles ( 40 . 42 ) with a height above a first limit value and / or a height below a second limit value as fixed obstacles ( 40 ), where the second limit is less than the first limit and not fixed as temporary obstacles ( 42 ) get ranked. A method according to claim 2, characterized in that the first limit value in the range of 1.5 to 2.5 m is selected and / or that the second limit value in the range of 0.1 to 0.5 m is selected. Method according to one of claims 1 to 3, characterized in that the environment map information about the registered obstacles ( 40 . 42 ), the information being selected from position, altitude, stationary or temporary obstacle rating, drivability, wheelchair accessibility, or a combination of at least two of these information. Method according to one of claims 1 to 4, characterized in that the determination of the current vehicle position ( 56 ) by comparing in the application mode with the environment sensors ( 18 ) ( 40 . 42 ) with the obstacles listed in the environment map ( 40 . 42 ), whereby only stationary obstacles ( 40 ) or fixed obstacles ( 40 ) are weighted more heavily than temporary obstacles ( 42 ). Method according to one of claims 1 to 5, characterized in that in the application mode, an offset between the current vehicle position ( 56 ) and the starting position ( 50 ) and a corrected trajectory ( 54 ) is determined according to the offset. Method according to one of claims 1 to 6, characterized in that in the application mode, the longitudinal guide and / or the transverse guide of the vehicle ( 1 ) by a driver assistance system ( 10 ) he follows. A computer program executing the method of any one of claims 1 to 7 when running on a computer. Driver assistance system ( 10 ) comprising environmental sensors ( 18 ) for detecting obstacles ( 40 . 42 ) in the vicinity of a vehicle ( 1 ), a control device ( 12 ) and a storage unit ( 14 ), characterized in that the driver assistance system ( 10 ) is arranged to carry out the method according to one of claims 1 to 7. Driver assistance system ( 10 ) according to claim 9, characterized in that the environment sensors ( 18 ) Comprise ultrasonic sensors, optical cameras, stereo cameras, radar, lidar or a combination of at least two of these sensor types.

Images