DE102022212897A1 - Method, control device and computer program for determining a probability with which an autonomous motor vehicle is located in a lane of a roadway, and computer-readable storage medium - Google Patents

Abstract

A method for determining a probability with which an autonomous motor vehicle (20) is located in a lane (34, 36) of a roadway (32) is provided. The method comprises: receiving position data that are representative of a current vehicle position of the motor vehicle (20); determining at least one reference point position of a reference point (94, 96) along the lane (34, 36) using a first lane lookup table that is assigned to the lane (34, 36) and in which reference points (94, 96) are assigned corresponding reference point positions, depending on the position data, wherein the reference points (94, 96) are representative of a course of the lane (34, 36) in the direction of travel in front of the motor vehicle (20) and wherein the determined reference point position is closest to the current vehicle position of the motor vehicle (20); determining a distance of the determined reference point position to the current vehicle position; and determining the probability with which the motor vehicle (20) is located in the lane (34) depending on the distance.

Description

The invention relates to a method, a control device and a computer program for determining a probability with which an autonomous motor vehicle is located in a lane of a roadway, as well as a computer-readable storage medium on which the computer program is stored.

An automatic emergency braking system (Advanced Emergency Braking System (AEBS)) is used to avoid or mitigate collisions in an autonomous motor vehicle. The autonomous motor vehicle can, for example, be a transport vehicle, for example a passenger transport vehicle. For example, the motor vehicle can be a means of mass transit. In this context, the motor vehicle can, for example, be referred to as a “Group Rapid Transport (GRT)” vehicle. The motor vehicle can, for example, be configured so that, with a few exceptions, it essentially drives on the roadway, whereby the roadway can be reserved for the motor vehicle and other similar motor vehicles, for example other GRTs. In this context, the roadway can be referred to as a “segregated lane”. The motor vehicle can accordingly be configured so that it does not drive on any other lanes apart from the lanes reserved for these types of motor vehicles. The exceptions can include, for example, bus stops, parking spaces or charging stations.

• In einem ersten Schritt werden mittels mindestens eines Sensors, der in dem Kraftfahrzeug angeordnet ist, ein oder mehrere Objekte in Fahrtrichtung vor dem Kraftfahrzeug erfasst. Der Sensor kann beispielsweise ein Radarsensor oder ein Lidarsensor sein. Ferner können das oder die Objekte mittels zwei oder mehr Sensoren erfasst werden, insbesondere mittels eines oder mehreren Radarsensoren und/oder mittels eines oder mehreren Lidarsensoren. Bei der Erkennung werden insbesondere die Objekte erfasst, die sich im Sichtfeld des entsprechenden Sensors befinden. Mit Hilfe des entsprechenden Sensors werden entsprechende Sensordaten erzeugt, insbesondere Radardaten bzw. Lidardaten. Die Daten können jeweils repräsentativ sein für die Positionen und/oder Geschwindigkeiten der entsprechenden Objekte, wobei die Positionen und/oder Geschwindigkeiten durch kartesische Koordinaten gekennzeichnet sein können.

An AEBS procedure underlying the AEBS system can be roughly divided into five or six steps:

• In a first step, one or more objects are detected in the direction of travel in front of the motor vehicle by means of at least one sensor arranged in the motor vehicle. The sensor can be a radar sensor or a lidar sensor, for example. Furthermore, the object or objects can be detected by means of two or more sensors, in particular by means of one or more radar sensors and/or by means of one or more lidar sensors. During detection, in particular the objects that are in the field of view of the corresponding sensor are detected. With the help of the corresponding sensor, corresponding sensor data is generated, in particular radar data or lidar data. The data can each be representative of the positions and/or speeds of the corresponding objects, whereby the positions and/or speeds can be identified by Cartesian coordinates.

In an optional second step, which is only performed if both a radar sensor and a lidar sensor are used, the radar data and the lidar data are merged into the sensor data.

In a third step, an area of interest, also known as a region of interest (ROI), is determined. The area of interest is determined so that it is located on the road in the direction of travel in front of the motor vehicle. Furthermore, the area of interest can be determined in such a way that it is highly likely that an object that is not in the area of interest does not pose a danger to the motor vehicle.

In a fourth step, the sensor data is used to check whether one or more of the detected objects are in the area of interest or not. If one or more of the detected objects are not in the area of interest, these objects can be excluded from further processing of the AEBS process.

In a fifth step, it is checked whether the objects that were not excluded in the third step and are therefore located in the area of interest pose a danger to the motor vehicle. If one or more of the objects detected in the area of interest do not pose a danger to the motor vehicle, these objects can be excluded during further processing of the AEBS procedure.

In a sixth step, braking, in other words deceleration, for example an emergency brake, of the motor vehicle is initiated in such a way that a collision of the motor vehicle with the object is avoided or at least reduced if the object represents a threat to the motor vehicle.

For the AEBS method, it is therefore particularly important, among other things, that the area of interest is determined accurately. The area of interest can normally be determined based on position, for example, whereby the normal case is characterized by the motor vehicle knowing which lane of the road it is driving in and/or that no special driving maneuver, such as turning or stopping, has been initiated by the motor vehicle. At least for the position-based determination of the area of interest, it can be helpful to be able to determine exactly how likely it is that the motor vehicle is in a certain lane of the road. The probability can can then be used to determine which lane of the road the motor vehicle is currently in.

Therefore, it is an object of the invention to provide a method, a control device and a computer program for determining a probability with which an autonomous motor vehicle is located in a lane of a roadway, which contribute to accurately determining an area of interest in which the object that may pose a danger to the motor vehicle may be located. Furthermore, it is an object of the invention to provide a computer-readable storage medium on which the computer program is stored.

These objects are achieved by the subject matter of the independent claims. Further embodiments of the invention emerge from the dependent claims and from the following description.

One aspect of the invention relates to a method for determining a probability with which an autonomous motor vehicle is located in a lane of a roadway, the method comprising: receiving position data that are representative of a current vehicle position of the motor vehicle; determining at least one reference point position of a reference point along the lane, using a first lane lookup table that is assigned to the lane and in which reference points are assigned corresponding reference point positions, depending on the position data, wherein the reference points are representative of a course of the lane in the direction of travel in front of the motor vehicle and wherein the determined reference point position is closest to the current vehicle position of the motor vehicle; determining a distance of the determined reference point position to the current vehicle position; and determining the probability with which the motor vehicle is located in the lane, depending on the distance.

One aspect of the invention relates to a control device for determining the probability with which the autonomous motor vehicle is located in the lane of the roadway, the control device comprising: a memory unit for storing the position data and the lane lookup table, which is assigned to the lane and in which reference points are assigned corresponding reference point positions, wherein the reference points are representative of the course of the lane in the direction of travel in front of the motor vehicle; and a processor that is configured to process the method explained above and below. The control device can be arranged in the motor vehicle, for example. The control device can be communicatively coupled to an external unit for retrieving and/or receiving the position data and/or the lane lookup table(s), for example via the Internet to a server on which the corresponding data is stored.

One aspect of the invention relates to a computer program for determining the probability with which the autonomous motor vehicle is located in the lane of the roadway, wherein the computer program has instructions which cause the above-mentioned method to be processed when they are executed by the control unit.

One aspect of the invention relates to a computer-readable medium on which the computer program is stored.

In most driving situations, especially when the motor vehicle is traveling on the road at medium or high speed and neither a stop nor a turn is initiated, a position-based determination of the area of interest provides particularly accurate results. To ensure this, the probability that the autonomous motor vehicle is in the lane of the road can be determined as explained above and below and taken into account when determining the area of interest. Thus, determining this probability helps to ensure that the area of interest in which the object that may pose a danger to the motor vehicle may be located can be accurately determined.

The region of interest may also be referred to as a "region of interest." The region of interest may be the area of the roadway that the motor vehicle is expected to travel over in the near future. In this case, the region of interest is in the same lane of the roadway as the motor vehicle. However, the region of interest may also be in a different lane of the roadway, for example, in a lane adjacent to the lane in which the motor vehicle is traveling. This can help to detect objects early that could pose a threat to the motor vehicle in the near future, for example if the object is in the other lane and is approaching the lane in which the motor vehicle is traveling.

The fact that the region of interest is determined position-based means that the region of interest is determined essentially depending on the position data. This can mean in particular that the position-based determination of the region of interest also other data can be taken into account, such as a lane of the road on which the motor vehicle is currently traveling, a shape of the motor vehicle and/or a width of the region of interest to be determined. However, the fact that the region of interest is determined based on position can also mean that the region of interest is determined at least partially depending on the speed data, whereby the speed data only serves as a supplement in the position-based determination of the region of interest, for example to restrict the region of interest determined based on the position data.

The autonomous motor vehicle may, for example, be a transport vehicle, for example a passenger transport vehicle. For example, the motor vehicle may be a means of mass transportation. In this context, the motor vehicle may, for example, be referred to as a “group rapid transport vehicle”. The motor vehicle may, for example, be configured such that, with a few exceptions, it essentially drives on the roadway, whereby the roadway may be reserved for the motor vehicle and other similar motor vehicles, for example other GRTs. In this context, the roadway may be referred to as a “segregated lane”. The motor vehicle may accordingly be configured such that it does not drive on any other roadways apart from the roadways reserved for these types of motor vehicles. The exceptions may, for example, include stops, parking spaces or charging stations.

The roadway can have a first lane for a first direction of travel and a second lane for a second direction of travel that is opposite to the first direction of travel. The motor vehicle normally moves in one of the two lanes, whereby it is assumed below that the motor vehicle normally moves in the first lane and only deviates from the first lane in special cases.

The position data that is representative of the vehicle position can be generated by a position sensor, transmitted to the control unit, and received by the control unit. The position sensor can, for example, be a GPS receiver that is arranged in the motor vehicle. Optionally, the position data can be representative of an orientation of the motor vehicle. The orientation of the motor vehicle can, for example, be used to determine in which lane of the road the motor vehicle is currently traveling, as explained in more detail below. If necessary, the orientation can, for example, be generated by means of an orientation sensor, a magnetic sensor, for example a compass, GPS, and/or by means of an inertial measurement unit (IMU), transmitted to the control unit, and received by the control unit.

The reference point positions can be specified, for example, by GPS coordinates in the corresponding lane lookup table. The reference point positions and optionally the corresponding reference points can be provided with indices in the corresponding lane lookup table and ordered accordingly. For example, such a lane lookup table can have an index for each reference point and a reference point position assigned to the index, for example within a row of the corresponding lane lookup table. An order of the indices can correspond to an order of the reference points on the corresponding lane in the real world. In addition, the lane lookup table can have lane orientations that are assigned to the corresponding reference points and that are representative of how the lane is oriented at the respective reference point. If necessary, the lane orientation can be specified, for example, as an absolute lane angle, for example in relation to one or more cardinal directions. Additionally or alternatively, the lane lookup table can have lane lengths that are assigned to the corresponding reference points and that are representative of how long the lane is starting from a starting point of the corresponding lane at the respective reference point.

The motor vehicle can, for example, have a separate lane lookup table for each lane in which the motor vehicle is moving, so that the motor vehicle can check for each lane whether the motor vehicle is in it or not. The corresponding lane lookup tables can be stored in the motor vehicle, for example in a memory unit of a control unit of the motor vehicle, or made available to the motor vehicle via a communication connection, for example via the Internet.

The lane lookup tables can be created in advance depending on the real lanes by measuring the corresponding lanes and extracting the relevant measurement results and storing them in the corresponding lane lookup tables. The lane lookup tables can also contain information about one or more curves, such as corresponding curvature data, for example in the form of the lane orientations, which are assigned to the corresponding reference points.

The control signal can be representative of the braking of the motor vehicle. In particular, the control signal can be configured such that the braking device initiates braking, in particular an emergency brake, of the motor vehicle in response to receiving the control signal.

According to one embodiment, the probability with which the motor vehicle is in the lane is compared with a predetermined lane threshold; the motor vehicle is classified as not being in the lane if the probability is less than the lane threshold; and the motor vehicle is classified as being in the lane if the probability is equal to or greater than the lane threshold. The comparison of the probability with the lane threshold makes it easy to classify whether the motor vehicle is in the lane or not.

According to one embodiment, the position data are representative of a current motor vehicle orientation of the motor vehicle; in the first lane lookup table, the reference points are each assigned a lane orientation of the lane at the corresponding reference points; the lane orientation is determined at the reference point position that is assigned to the corresponding reference point and that is closest to the current motor vehicle position using the first lane lookup table; a deviation of the current motor vehicle orientation from the determined lane orientation is determined; and the probability with which the motor vehicle is in the lane is determined depending on the deviation. The comparison of the current vehicle orientation with the lane orientation at the determined first reference point position and the resulting deviation of the current motor vehicle orientation from the corresponding lane orientation can help to determine the probability with which the motor vehicle is in the first lane in a particularly suitable manner, in particular particularly precisely. In particular, the comparison of the motor vehicle orientation with the lane orientation provides an additional clue compared to the comparison of the reference point position with the vehicle position, whereby the accuracy of determining the probability can be particularly high.

According to one embodiment, the motor vehicle is configured to drive on the lane optionally in a first direction of travel or in a second direction of travel opposite to the first direction of travel; the first lane lookup table is representative of the course of the lane in the first direction of travel; a current direction of travel of the motor vehicle is determined; the reference point position that is closest to the current vehicle position of the motor vehicle is determined using the first lane lookup table if the current direction of travel corresponds to the first direction of travel; and the reference point position that is closest to the current vehicle position of the motor vehicle is determined using a second lane lookup table if the current direction of travel corresponds to the second direction of travel, wherein reference point positions in the second lane lookup table are assigned to reference points that are representative of a course of the lane in the second direction of travel.

The reference points and reference point positions stored in the first lane lookup table can also be referred to as first reference points or first reference point positions. The reference points and reference point positions stored in the second lane lookup table can also be referred to as second reference points or second reference point positions. Determining the reference point position depending on the current direction of travel of the motor vehicle using the first lane lookup table or the second lane lookup table makes it possible to determine the reference point position precisely regardless of the direction in which the motor vehicle is currently moving.

According to one embodiment, the roadway has the lane as the first lane and at least one second lane directly next to the first lane; and the reference point position that is closest to the current vehicle position of the motor vehicle is determined using a third lane lookup table, wherein reference point positions in the third lane lookup table are assigned to reference points that are representative of a course of the second lane.

In other words, the reference point position closest to the current vehicle position of the motor vehicle can be searched not only in the first lane lookup table, but also in the third lane lookup table. For example, the reference point position closest to the current vehicle position of the motor vehicle can be searched in the third lane lookup table if the distance of the current vehicle position to the closest reference point position from the first lane lookup table is greater than the predetermined lane threshold. If then the distance of the current vehicle position to the closest reference point position from the third lane lookup table is smaller than the specified lane threshold, the motor vehicle is classified as being in the second lane. However, if the distance of the current vehicle position to the closest reference point position from the first lane lookup table is already smaller than the specified lane threshold, consultation of the third lane lookup table can be omitted and the motor vehicle can be classified as being in the first lane.

Alternatively, the closest reference point to the current vehicle position from the first lane lookup table and the closest reference point to the current vehicle position from the third lane lookup table may be searched for, and the corresponding distances to the current vehicle position may be determined. If the distance between the current vehicle position and the determined reference point from the first lane lookup table is less than the distance between the current vehicle position and the determined reference point from the third lane lookup table, the motor vehicle may be classified as being in the first lane. If the distance between the current vehicle position and the determined reference point from the first lane lookup table is greater than the distance between the current vehicle position and the determined reference point from the third lane lookup table, the motor vehicle may be classified as being in the second lane. However, if both distances are greater than the predetermined lane threshold, the motor vehicle may be classified as being in neither the first nor the second lane.

The reference points and reference point positions stored in the third lane lookup table can also be referred to as third reference points or third reference point positions. If the motor vehicle can move in the second lane in a first direction of travel and a second direction of travel opposite to the first direction of travel, a fourth lane lookup table can be specified that is representative of a course of the second lane in the second direction of travel. When determining the distance and/or the deviation and accordingly when determining the probability of which lane the motor vehicle is in and/or in which direction the motor vehicle is moving, the fourth lane lookup table can then also be taken into account accordingly.

It is to be understood that features of the method as described above and below may also be features of the control device, the computer program and/or the computer-readable medium and vice versa. The computer-readable medium may be a hard disk, a USB storage device, a RAM, a ROM, an EPROM or a FLASH memory. The computer-readable medium may also be a data communications network, such as the Internet, which enables the download of a program code.

• 1 zeigt ein Ausführungsbeispiel eines autonomen Kraftfahrzeugs.
• 2 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines allgemeinen Verfahrens zum Vermeiden oder Vermindern einer Kollision des autonomen Kraftfahrzeugs mit einem Objekt.
• 3 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines speziellen Verfahrens zum Vermeiden oder Vermindern einer Kollision des autonomen Kraftfahrzeugs mit dem Objekt.
• 4 zeigt ein Display auf dem eine beispielhafte Fahrsituation dargestellt ist.
• 5 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines speziellen Verfahrens zum Vermeiden oder Vermindern einer Kollision des autonomen Kraftfahrzeugs mit dem Objekt.
• 6 zeigt eine Draufsicht auf eine beispielhafte Fahrsituation, auf das Kraftfahrzeug und auf einen interessierenden Bereich vor dem Kraftfahrzeug.
• 7 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines Verfahrens zum positionsbasierten Ermitteln eines interessierenden Bereichs.
• 8 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines Verfahrens zum Ermitteln, ob sich das Objekt in dem interessierenden Bereich in Fahrrichtung vor dem Kraftfahrzeug befindet.
• 9 zeigt ein Fahrzeugmodell des Kraftfahrzeugs zum geschwindigkeitsbasierten Ermitteln des interessierenden Bereichs.
• 10 zeigt eine Prinzipskizze zum Erläutern des geschwindigkeitsbasierten Ermittelns des interessierenden Bereichs.
• 11 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines Verfahrens zum geschwindigkeitsbasierten Ermitteln des interessierenden Bereichs.
• 12 zeigt ein Beispiel von zwei Fahrlinien für das Kraftfahrzeug.
• 13 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines Verfahrens zum Ermitteln einer Wahrscheinlichkeit, mit der sich das Kraftfahrzeug auf einer Fahrspur einer Fahrbahn befindet.
• 14 zeigt ein Diagramm, in dem mehrere beispielhafte Verläufe von Zeitdauern die vergehen, bis das Kraftfahrzeug mit dem Objekt kollidiert, in Abhängigkeit einer Geschwindigkeit des Kraftfahrzeugs dargestellt sind.
• 15 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines speziellen Verfahrens zum Vermeiden oder Vermindern der Kollision des autonomen Kraftfahrzeugs mit dem Objekt.
• 16 zeigt ein Diagramm, in dem mehrere beispielhafte Längen des interessierenden Bereichs in Abhängigkeit von der Geschwindigkeit des Kraftfahrzeugs dargestellt sind.
• 17 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines speziellen Verfahrens zum Vermeiden oder Vermindern der Kollision des autonomen Kraftfahrzeugs mit dem Objekt.

In the following, embodiments of the invention are described in detail with reference to the accompanying figures.

• 1 shows an embodiment of an autonomous motor vehicle.
• 2 shows a flowchart of an embodiment of a general method for avoiding or reducing a collision of the autonomous motor vehicle with an object.
• 3 shows a flow chart of an embodiment of a special method for avoiding or reducing a collision of the autonomous motor vehicle with the object.
• 4 shows a display showing an example driving situation.
• 5 shows a flow chart of an embodiment of a special method for avoiding or reducing a collision of the autonomous motor vehicle with the object.
• 6 shows a top view of an exemplary driving situation, of the motor vehicle and of an area of interest in front of the motor vehicle.
• 7 shows a flow chart of an embodiment of a method for position-based determination of a region of interest.
• 8th shows a flow chart of an embodiment of a method for determining whether the object in the area of interest is located in front of the motor vehicle in the direction of travel.
• 9 shows a vehicle model of the motor vehicle for speed-based determination of the area of interest.
• 10 shows a schematic diagram to explain the speed-based determination of the region of interest.
• 11 shows a flow chart of an embodiment of a method for speed-based determination of the region of interest.
• 12 shows an example of two driving lines for the motor vehicle.
• 13 shows a flow chart of an embodiment of a method for determining a probability with which the motor vehicle is located in a lane of a roadway.
• 14 shows a diagram in which several exemplary courses of time periods that pass until the motor vehicle collides with the object are shown as a function of a speed of the motor vehicle.
• 15 shows a flow chart of an embodiment of a special method for avoiding or reducing the collision of the autonomous motor vehicle with the object.
• 16 shows a diagram in which several exemplary lengths of the region of interest are represented as a function of the speed of the motor vehicle.
• 17 shows a flow chart of an embodiment of a special method for avoiding or reducing the collision of the autonomous motor vehicle with the object.

The reference symbols used in the figures and their meaning are listed in summary form in the list of reference symbols. In principle, identical or similar parts are provided with the same reference symbols.

Generally

1 shows an embodiment of an autonomous motor vehicle 20. The motor vehicle 20 has a passenger cabin 21 for accommodating one or more passengers (not shown), a door 22 for entering or leaving the passenger cabin 21, wheels 23 for moving the motor vehicle 20, a braking device 24 for braking, in other words decelerating, the motor vehicle 21, and a control unit 25 for operating the motor vehicle 22. The motor vehicle 20 can be configured such that both the wheels 23 on a front axle of the motor vehicle 20 and the wheels 23 on a rear axle of the motor vehicle 20 are steerable, in particular pivotable relative to the corresponding axle. As a result, in addition to or as an alternative to driving straight ahead and cornering the motor vehicle 20, a sideways movement, in other words "crabwise motion", of the motor vehicle 20 is possible, in particular by turning both the wheels 23 on the front axle and the wheels 34 on the rear axle in the same direction.

The autonomous motor vehicle 20 can be, for example, a transport vehicle, for example a passenger transport vehicle. For example, the motor vehicle 20 can be a means of mass transportation. In this context, the motor vehicle can be referred to, for example, as a “Group Rapid Transport vehicle” (“GRT” for short).

The motor vehicle 20 can, for example, be configured so that, with a few exceptions in which special driving maneuvers are necessary or at least carried out, it essentially travels on a predetermined roadway 32 (see 4 ). The lane 32 may be reserved for the motor vehicle 20 and other similar motor vehicles 20, for example other GRTs. In this context, the lane 32 may be referred to as a "segregated lane". The motor vehicle 20 may accordingly be configured such that it does not travel on any lanes other than the lanes 32 reserved for this type of motor vehicle. The exceptions may include, for example, bus stops, parking spaces or charging stations. The motor vehicle 20 may perform the sideways movement at or near these exceptions.

The roadway 32 may, for example, be a first lane 34 (see 4 ) and a second lane 36 adjacent to the first lane 34. For example, the first lane 34 can be provided for a first direction of travel of the motor vehicle 20, and the second lane 36 can be provided for a second direction of travel of the motor vehicle 20. Alternatively, each of the lanes 34, 36 can be used by the motor vehicle 20 in both directions. For example, the motor vehicle 20 can be a first motor vehicle 20 that can move in both directions on the first lane 34, and a second motor vehicle 20 can move in both directions on the second lane 36. Normally, the motor vehicle 20 moves on one of the two lanes 34, 36, wherein it is assumed below that the motor vehicle 20 normally moves forward in the first lane 34 in the direction of travel and only deviates from the first lane 34 in special cases.

The control unit 25 can, for example, comprise or form a vehicle control system of the motor vehicle 20. The control unit 25 has a memory unit 26 for storing data and a processor 27 for processing the data. The control unit 25 can be used to retrieve and/or receive sensor data, position data, speed data and/or other data be communicatively coupled to an external unit (not shown), for example via the Internet to a server on which the corresponding data is stored.

The control unit 22 may be configured to avoid or reduce a collision of the motor vehicle 20 traveling on the roadway 32 with an object 50 (see 4 ). In this context, the storage unit 26 serves to store sensor data, in particular radar data and/or lidar data, position data, speed data and/or one or more time threshold values, and the processor 27 is configured in this context to execute a method for avoiding or reducing the collision of the motor vehicle 20 traveling on the roadway 32 with the object 50.

The time threshold values may be stored in the memory unit 26 in the form of a time threshold lookup table in which predetermined speeds of the motor vehicle 20 are assigned to corresponding time threshold values.

The control unit 22 can be configured to generate a control signal for braking the motor vehicle 20, wherein the control signal is representative of the braking of the motor vehicle 20. In particular, the control signal can be configured such that the braking device 24 initiates the braking of the motor vehicle 20 in response to receiving the control signal. The braking device 24 can be hardware that communicates with the control unit 22 and with mechanical and/or hydraulic brakes (not shown) of the motor vehicle 20. Alternatively, the braking device 24 can be implemented as software and processed by the control unit 22.

The control unit 22 may be configured to detect a region of interest 60 (see 6 ) that is located on a lane, for example the first lane 34, in the direction of travel in front of the motor vehicle 20, is determined based on position. In this context, the storage unit 26 serves to store sensor data, position data and/or speed data, and the processor 27 is configured in this context to execute a method for position-based determination of the region of interest 60.

The control unit 22 can be configured to determine the region of interest 60 based on speed. In this context, the storage unit 26 serves to store sensor data, position data and/or speed data, and the processor 27 is configured in this context to execute a method for determining the region of interest 60 based on speed.

The control unit 22 can be configured to determine a probability with which the motor vehicle 20 is located on one of the lanes 34, 36 of the roadway 32. In this context, the storage unit 26 serves to store position data and at least one lane lookup table which is assigned to the lane 34, 36 corresponding to it and in which reference points 94, 96 (see 14 ) corresponding reference point positions are assigned, wherein the reference points 94, 96 are representative of a course of the corresponding lane 34, 36 in the direction of travel in front of the motor vehicle 20, and the processor 27 is configured in this context to execute a method for determining the probability.

2 shows a flow chart of an embodiment of a general method for avoiding or reducing a collision of the autonomous motor vehicle 20 with the object 50 (see 4 ). Further embodiments, for example special embodiments, of the method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50 are explained with reference to the following figures.

In a step S2, sensor data can be received. The sensor data is representative of the object 50 that is located in front of the motor vehicle 20 in the direction of travel. The sensor data can be generated by at least one sensor (not shown). The sensor can be a radar or a lidar sensor, for example. The sensor data can alternatively be generated by two or more sensors, wherein the sensors can have at least one radar and at least one lidar sensor, for example. In this case, the sensor data can have radar data or lidar data, or the radar data and the lidar data can be fused to form the sensor data. The sensor data can be transmitted from the sensor(s) to the control unit 25 of the motor vehicle 20 that is carrying out the method, and received by the control unit 25.

The sensor data can be in the form of a list that contains all objects 50 in front of the motor vehicle 20 that are detected by the sensor(s). For example, the radar data can be in the form of a radar list in which the objects 50 detected by radar are stored as radar objects. Alternatively or additionally, the lidar data can be in the form of a lidar list in which the objects 50 detected by lidars are stored as lidar objects.

In an optional step S4, the sensor data can be fused, in particular if the radar data and the sensor data are received in step S2. If appropriate, the fused sensor data can be in the form of a fusion list, which can include fusion objects, radar-only objects and/or lidar-only objects, as explained in more detail below.

In a step S6, the region of interest 60, which is located on the roadway 32 in the direction of travel in front of the motor vehicle 20, can be determined. The region of interest 60 can be determined, for example, depending on position data that are representative of a current vehicle position of the motor vehicle 20 and/or depending on speed data that are representative of a current speed of the motor vehicle 20.

The region of interest 60 may also be referred to as a “region of interest” (ROI). The region of interest 60 may be the region of the roadway 32 that the motor vehicle 20 is expected to travel over in the near future. In this case, the region of interest 60 is located in the same lane 34, 36 of the roadway 32 as the motor vehicle 20. However, the region of interest 60 may also be located in a different lane 34, 36 of the roadway 32, for example in a lane 34, 36 that is adjacent to the lane 34, 36 on which the motor vehicle 20 is traveling.

The position data can be generated by a position sensor (not shown), transmitted to the control unit 25 and received by the control unit 25. The position sensor can be arranged in the motor vehicle 20. The position sensor can be a GPS receiver, for example. The position data can also be representative of an orientation of the motor vehicle 20, in other words a vehicle orientation. In particular, the position data can include orientation data that is representative of the vehicle orientation of the motor vehicle 20. The orientation data can be generated, for example, by means of an orientation sensor, a magnetic sensor, for example a compass, GPS, and/or by means of an inertial measurement unit (IMU), transmitted to the control unit and received by the control unit. The vehicle orientation can be specified, for example, as a yaw angle, for example as an absolute yaw angle. The absolute yaw angle can be specified, for example, with reference to one or more of the cardinal directions or in a proprietary reference system of the control unit.

The speed data may be generated by a speed sensor (not shown) located in the motor vehicle 20, transmitted to the controller 25, and received by the controller 25. Alternatively, the speed data may be retrieved from the memory unit 26 of the motor vehicle 20 and received by the controller 25.

In a step S8, it can be checked whether the object 50 or one, two or more other objects 50 are located in the region of interest 60, depending on the sensor data and the determined region of interest 60. In particular, step S8 can be carried out for all objects 50 that are coded in the sensor data, the radar data and/or the lidar data. To check whether the object or objects 50 are located in the region of interest 60, object positions of the objects 50 can be determined depending on the sensor data, for example in Cartesian coordinates, for example in a first coordinate system. In addition, the region of interest 60 can be specified in Cartesian coordinates, for example also in the first coordinate system. Then, in step S8, it can be easily checked based on the corresponding Cartesian coordinates whether one, two or more of the objects 50 are in the region of interest 60 or not. If the condition of step S8 is met, processing can be continued in an optional step S10. If the condition of step S8 is not met, processing can be continued again in step S2.

In the optional step S10, it can be determined whether one or more of the objects 50 located in the area of interest 60 pose a danger to the motor vehicle 20. For example, one or more of the objects 50 can be so small and/or light that they do not pose a danger to the motor vehicle 20. In this case, one of the objects 50 can be, for example, a bird sitting on the roadway 32. Alternatively or additionally, one or more of the objects 50 can cross the roadway 32 so quickly that there is no danger of a collision. In this case, one of the objects 50 can be, for example, a bird flying over the roadway 32. If the condition of step S10 is met, processing can continue in a step S12. If the condition of step S10 is not met, the processing can be continued again in step S8, for example with another of the objects 50. As an alternative to carrying out step S10, it can be assumed that all detected objects 50 in the area of interest 60 pose a danger to the motor vehicle 20.

In step S12, the control signal for the braking device 24 of the motor vehicle 20 can be generated. If the optional step S10 is not processed, the control signal is generated when one or more of the objects 50 are located in the area of interest 60. If the optional step S10 is processed, the control signal is only generated when one or more of the objects 50 in the area of interest 60 pose a danger to the motor vehicle 20. Braking, in other words deceleration, of the motor vehicle can be a "normal" braking process, an emergency brake, or a rapid braking of the motor vehicle. During the normal braking process, passenger comfort can have the highest priority. The emergency brake can be configured in such a way that little or no consideration is given to driving comfort and the sole focus is on braking the motor vehicle 20 as quickly as possible. The emergency brake can be configured in such a way that driving comfort is taken into account and the focus is on braking the motor vehicle 20 as quickly as possible. The braking distance of the motor vehicle 20 is normally shorter with the emergency brake than with the emergency brake, and shorter with the emergency brake than with normal braking. The amount of negative acceleration during braking can, for example, be in a range from 1 m/s ² to 10 m/s ² , for example from 2 m/s ² to 5 m/s ² , for example around 3 m/s ² . The negative acceleration can be significantly higher with the emergency brake than with the normal braking process. The amount of negative acceleration can be greater with the emergency brake than with the normal braking process and smaller than with the emergency brake. With each of these braking processes, the motor vehicle can be braked until it comes to a standstill.

Switching logic

3 shows a flow chart of an embodiment of a special method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50. The method described with reference to 3 The special procedure explained can largely be applied to the procedure referred to 2 general procedures explained above. In particular, the procedure with reference to 3 specific methods explained include steps corresponding to steps of the method with reference to 2 or even identical with the general procedure explained above. Therefore, only the steps of the special procedure according to 3 in which the special procedure according to 3 from the general procedure according to 2 deviates.

A step S20 of the special method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50 can correspond to the step S2 of the general method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50. Furthermore, in the step S20, a processing of the sensor data corresponding to the step S4 can be carried out, in particular if the sensor data comprises the radar data and the lidar data.

In a step S22, the position data can be received, for example as explained in more detail above with reference to step S6.

In a step S24, the speed data can be received, for example as explained in more detail above with reference to step S6.

In a step S26, it can be checked whether a predetermined exclusion criterion is met that speaks against determining the region of interest 60 in a position-based manner, depending on the position data and/or the speed data. In most driving situations, in particular when the motor vehicle 20 is moving at medium or high speed on the roadway 32 and neither a stop nor a turn is initiated, the position-based determination of the region of interest 60 can provide more accurate results than the speed-based determination of the region of interest 60. In particular, in these driving situations, the region of interest 60 can be adapted more precisely to the roadway 32, in particular the corresponding lane 34, 36, by the position-based determination than with the speed-based determination. However, in the case of special driving maneuvers, for example at stops, charging stations or in tight curves, and/or in the case of a sideways movement (“crabwise motion”) of the motor vehicle 20, the speed-based determination of the region of interest 60 can provide more accurate results. Therefore, the position-based determination of the region of interest 60 can be specified as standard and switched to the speed-based determination only when the one or one, two or more further exclusion criteria characterizing the special driving maneuvers are met.

A first of the exclusion criteria can be met, for example, if the motor vehicle 20 is currently in an area in which the area of interest 60 is to be determined exclusively on the basis of speed. There can be areas on and/or at the roadway 32 for which it is known that the motor vehicle 20 is in the these areas must perform special driving maneuvers, for example sideways movement or a tight curve, for which the position-based determination of the region of interest 60 is unsuitable. For example, in the area in which the region of interest 60 is to be determined exclusively based on speed, a probability that the speed-based determination of the region of interest 60 will provide a more accurate result than the position-based determination of the region of interest 60 may be relatively high, in particular compared to a situation in which the motor vehicle is not currently in the area in which the region of interest 60 is to be determined exclusively based on speed. For example, the area in which the region of interest 60 is to be determined exclusively based on speed may have a parking lot, a bus stop or a charging station for the motor vehicle 20, or an intersection at which the roadway 32 crosses another roadway 32. The other roadway 32 may also be reserved for the motor vehicle 20 or the other similar motor vehicles 20, for example other GRTs. Thus, the other lane 32 can also be a separate lane 32, in particular a “segregated lane”. The method can be configured such that the region of interest 60 is only determined based on position if the first exclusion criterion is not met and if no other exclusion criterion speaks against the position-based determination.

The first exclusion criterion may be met, for example, if the motor vehicle 20 is currently at a stop or is approaching the stop. If the motor vehicle 20 is at the stop or is approaching it, the probability that the speed-based determination of the region of interest 60 will provide a more accurate result than the position-based determination of the region of interest 60 may be relatively high, particularly compared to a situation in which the motor vehicle 20 is not currently at a stop and/or is not approaching the stop. The stop may be located next to the roadway 32. The stop may serve to allow one or more passengers to board and/or disembark from the motor vehicle 20. The position data may be representative of the motor vehicle being at the stop or approaching it. In this context, the position data may be generated independently of the GPS receiver of the motor vehicle 20. For example, in this case, the position data may be generated by a sensor that is arranged at the stop or in the vicinity of the stop and that is configured to detect the motor vehicle 20 approaching the stop or standing at it. Alternatively, initiating and/or performing the sideways movement may be representative of the motor vehicle being in the area of the stop or approaching it.

A second exclusion criterion can be met, for example, if the motor vehicle 20 has initiated or is already performing the sideways movement. In particular, during the sideways movement, the vehicle orientation, which can be used for the position-based determination of the region of interest, for example to check which lane the motor vehicle is in, can deviate from a direction of movement of the motor vehicle, so that the position-based determination of the region of interest 60 can be unsuitable in the case of sideways movement of the motor vehicle 20. The method can be configured such that the region of interest 60 is only determined position-based if the second exclusion criterion is not met and if no other exclusion criterion speaks against the position-based determination.

Depending on the position data, a probability can be determined with which the motor vehicle 20 is moving in a lane 34 of the roadway 32. This probability can be determined, for example, by means of the method explained below for determining the probability with which the autonomous motor vehicle 20 is in the lane 34, 36 of the roadway 32. A third of the exclusion criteria can be met if this probability is less than a predetermined probability threshold. The predetermined probability threshold can, for example, be in a range from 80% to 100%, for example from 80% to 90%, for example approximately 80%. The probability threshold can, for example, be empirically determined and specified by a manufacturer of the motor vehicle 20 by storing the probability threshold in the memory unit 26 of the control unit 25.

If the probability with which the motor vehicle 20 moves on the lanes 34, 36 of the roadway 32 is smaller than the predetermined probability threshold, the probability that the speed-based determination of the region of interest 60 provides a more accurate result than the position-based determination of the region of interest 60 may be relatively high, in particular compared to a situation in which the probability is equal to or greater than the predetermined probability threshold. The method may be configured such that the region of interest 60 is only is determined on a position-based basis if the third exclusion criterion is not met and if no other exclusion criterion speaks against the position-based determination.

A current speed of the motor vehicle 20 can be determined depending on the speed data. A fourth exclusion criterion can be met if the determined speed is less than a predetermined speed threshold. The predetermined speed threshold can, for example, be in a range from 0.5 m/s to 10 m/s, for example from 1 m/s to 5 m/s, for example 2 m/s. The speed threshold can, for example, be empirically determined by the manufacturer of the motor vehicle 20 and specified by storing the speed threshold on the memory unit 26 of the control unit 25.

If the speed is less than the speed threshold, there may be an increased probability that the motor vehicle 20 will leave its current lane 34, 36 or the roadway 32 in the near future. In this case, the probability that the speed-based determination of the region of interest 60 will provide a more accurate result than the position-based determination of the region of interest 60 may be relatively high, particularly compared to a situation in which the speed is equal to or greater than the predetermined speed threshold. The method may be configured such that the region of interest 60 is only determined position-based if the fourth exclusion criterion is not met and if no other exclusion criterion speaks against the position-based determination.

In the above, the exclusion criteria are referred to as "first", "second", "third" and "fourth" exclusion criteria. However, this numbering does not represent any limitation and serves solely to distinguish one selection criterion from another. In particular, this numbering does not represent an order or a priority of the exclusion criteria. If several of the exclusion criteria are checked, they can be checked in any combination, priority and/or order. Furthermore, one, two or three of the selection criteria with a higher number can be checked without checking a selection criterion with a lower number. For example, only the second and fourth selection criteria can be checked, whereas the first and third selection criteria are not checked. This example can be applied to any other combination of checking the selection criteria.

If the condition of step S26 is not met, processing can be continued in step S28. If the condition of step S26 is met, processing can be continued in step S30.

In step S28, the region of interest 60 is determined in a position-based manner depending on the position data, for example by means of a method for position-based determination of the region of interest 60, which is located on the lane 34, 36 in the direction of travel in front of the motor vehicle 20, as explained in more detail below. The fact that the region of interest 60 is determined in a position-based manner means that the region of interest 60 is determined essentially in dependence on the position data. This can mean in particular that other data can also be taken into account in the position-based determination of the region of interest 60, such as the lane 34, 36 of the roadway 32 on which the motor vehicle 20 is currently driving, a current vehicle orientation of the motor vehicle 20, a lane orientation of the lane 34, 36 at or near the position of the motor vehicle 20, a shape of the motor vehicle 20 and/or a width of the region of interest 60 to be determined. However, the fact that the region of interest 60 is determined in a position-based manner can also mean that the region of interest 60 is determined at least partially depending on the speed data, wherein the speed data in the position-based determination of the region of interest 60 only serve as a supplement, for example to restrict the region of interest 60 determined depending on the position data, for example its length I _ROI (see 10 ). The method may be configured such that position-based determination of the region of interest 60 is the standard for determining the region of interest 60.

In step S28, the region of interest 60 is determined in a speed-based manner depending on the speed data, for example by means of a method for speed-based determination of the region of interest 60, which is located on the lane 34, 36 in the direction of travel in front of the motor vehicle 20, as explained in more detail below. The fact that the region of interest is determined in a speed-based manner means that the region of interest is essentially determined in a speed-based manner depending on the speed data. This can mean in particular that other data can also be taken into account in the speed-based determination of the region of interest 60, such as a yaw rate ω _ego of the motor vehicle 20, a shape of the motor vehicle 20 and/or a width of the region of interest 60 to be determined.

Steps S32, S34 and S36 can be processed correspondingly to steps S8, S10 and S12 respectively.

Fusion of radar and lidar data

4 shows a display 30 on which an exemplary driving situation is shown. In particular, 4 and in particular the display 30 the roadway 32 with the first lane 34 and the second lane 36, which are separated by a center line 35. A 4 right side edge of the roadway 32 can be referred to as the first side edge 37 and a 4 left side edge of the roadway 32 can be referred to as second side edge 39.

Several objects 50 are located next to the roadway. The objects 50 include radar objects 38 that are encoded in the radar data, contour points 40 that are encoded in the lidar data, and fusion objects 42. One, two, or more of the contour points 40 that are close to each other are representative of a lidar object. Accordingly, the display 30 shows several lidar objects that may also be encoded in the lidar data. Radar objects 38 on or in which none of the contour points 40 is located may also be referred to as radar-only objects. Lidar objects for which no radar object 38 exists may also be referred to as lidar-only objects.

For each of the contour points 40 that lies on or in one of the radar objects 38, one of the fusion objects 42 is generated. In other words, the radar objects 38 and the lidar objects, in particular the corresponding radar data and the corresponding lidar data, are fused, as described with reference to 5 explained in more detail. If several of the contour points 40 that correspond to one of the lidar objects are located on or in one of the radar objects 38, it may be that based on one of the corresponding contour points 40 together with the corresponding radar object, only one of the fusion objects 42 is generated, wherein any of these contour points 40 can be used for the fusion with the corresponding radar object 38.

Furthermore, 4 and the display 30 several velocity vectors 44 which are representative of the directions of movement and velocities of the radar-only objects and which are in 4 drawn with a dash-dot line, several velocity vectors 46, which are representative of the directions of movement and velocities of the song-only objects and which are in 4 drawn with a dotted line, and several velocity vectors 48, which are representative of the directions of movement and velocities of the fusion objects 42 and which are in 4 marked with a solid line.

5 shows a flow chart of an embodiment of a special method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50.

In a step S40, radar data can be received that is representative of one or more of the radar objects 38 that are located in front of the motor vehicle 20 in the direction of travel. The radar data can be generated by the radar sensor, transmitted to the control unit 25 and received by the control unit 25. The radar sensor can be arranged in the motor vehicle 20, for example. Optionally, the radar data can be generated by two or more radar sensors, which can each be arranged in the motor vehicle 20, for example. The radar sensor(s) can be part of a radar system.

The radar data may be in the form of a digital radar object list in which the radar objects 38 are listed. For example, each radar object in the radar object list may be assigned a unique identification number (ID), a size and/or a position of the corresponding radar object.

Furthermore, depending on the radar data, directions of movement of the objects 50, in particular the radar objects, can be determined. Depending on the determined directions of movement, orientations of the corresponding objects 50 can be determined and the radar data can be supplemented so that the radar data are representative of the orientations of the corresponding objects 50. For example, the orientations can be assigned to the corresponding radar objects in the radar object list. If necessary, the supplemented radar data can be merged with the lidar data when the radar data and the lidar data are later merged.

Normally, the radar data are exclusively representative of a position and size of the object 50. The directions of movement of the objects 50, in particular the radar objects, can be determined using radar data that are recorded at different points in time, in particular successive points in time. If the objects 50 are other road users, for example other vehicles, the orientations of the objects 50 can correspond to their directions of movement. The orientations can therefore be determined depending on the directions of movement, for example by aligning the orientations parallel to the corresponding directions of movement. gen or identical to the corresponding directions of movement. The fact that one of the objects 50 is another road user can be recognized, for example, by the fact that the corresponding object 50 is located on the roadway 32. Alternatively, radar systems are known which each have one or more radar sensors and a corresponding evaluation logic, wherein the evaluation logic has an object recognition which is able to recognize whether one or more of the radar objects 38 are other road users or not and/or which road user(s) the object(s) 50 are.

In particular, the radar data can be representative of classes to which the radar objects 38 belong. Sizes of the radar objects 38 can be assigned to the classes of objects 50. The classes can include, for example, "truck", "van", "car", "motorcycle", "bicycle" and/or "pedestrian", wherein the (standard) sizes assigned to these classes become smaller in the list from the class "truck" to the class "pedestrian". The radar data can be adapted so that the radar data is representative of the radar objects 38 with the corresponding sizes of the radar objects 38. For example, the sizes can be assigned to the corresponding radar objects in the radar object list. If necessary, when merging the radar data with the lidar data, the adapted radar data can be merged with the lidar data.

A radar system that has a suitable radar sensor and that provides the aforementioned functions is currently sold, for example, by the company “Conti”, for example the model ARS 404-21 or the model ARS 408-12, which can be used as standardized ARS interfaces and are each described in the corresponding technical documentation from Conti, in particular in “Technical Documentation, ARS 404-21 (Entry), ARS 408-21 (Premium)”, version 1.91, May 18, 2018. This and documentation of other suitable radar systems can be obtained, for example, from https://conti-engineering.com/components/ars-408/. These radar systems provide as outputs, among other things, the position of the detected object 50, the size of the detected object 50, the speed of the detected object 50, the speed of the motor vehicle 20, and the class of the detected object 50.

In a step S42, the lidar data can be received that are representative of one or more of the lidar objects that lie in front of the motor vehicle 20 in the direction of travel. The lidar data can be in the form of a digital lidar object list that includes the lidar objects. The lidar data can be generated by a lidar sensor, transmitted to the control unit 25 and received by the control unit 25. The lidar sensor can be arranged in the motor vehicle 20, for example. Optionally, the lidar data can be generated by two or more lidar sensors, each of which can be arranged in the motor vehicle 20, for example. The lidar data can include contour data that are representative of contour points 40, wherein each of the lidar objects is represented by one or more of the contour points 40. In the lidar object list, the contour points can each be assigned a unique identification number (ID), a lidar object, a speed of the corresponding lidar object and a position of the corresponding lidar object.

A lidar system having a suitable lidar sensor is currently sold, for example, by the company “ibeo automotive”, for example from the “LUX” series, in particular the “Ibeo LUX 4L”, and is described in the corresponding technical documentation from ibeo, in particular in “Interface Specification for ibeo LUX, ibeo LUX systems and ibeo Evaluation Suite”, version 1.48.2, July 28, 2017, available at www.ibeo-as.com. This and documentation of other suitable lidar systems can be obtained, for example, from https://www.ibeo-as.com/en/products/sensors/IbeoLUX. This lidar system provides as outputs a position of a detected object, a speed of the detected object, a speed of the motor vehicle and a class of the motor vehicle.

In a step S44, the radar data, possibly the supplemented radar data, and the lidar data can be fused to form the sensor data, the sensor data being representative of one or more of the sensor objects that correspond to the radar objects 38 and/or the lidar objects. The fact that the radar data and the lidar data are fused to form the sensor data means that the radar data and the lidar data are combined, for example in order to be mutually verified and/or mutually supplemented. The sensor data can be in the form of the fusion list that contains the sensor objects. When merging the radar data and the lidar data to form the sensor data, it can be checked for each of the contour points 40 whether the corresponding contour point 40 overlaps with one of the radar objects 38. If one of the contour points 40 overlaps with one of the radar objects 38, a corresponding fusion object 42 can be generated for the contour point 40 and all other contour points 40 of the same lidar object and encoded in the sensor data. If none of the contour points 40 of one of the lidar objects overlaps with one of the radar objects 38, a corresponding lidar object, a lidar-only object can be generated and encoded in the sensor data. For each of the radar objects 38 for which no overlapping contour point 40 is found, a radar-only object can be generated and encoded in the sensor data. The sensor objects can thus be classified into the radar-only objects whose corresponding radar objects 38 do not correspond to any of the lidar objects, into the lidar-only objects whose corresponding lidar objects do not correspond to any of the radar objects 38 and/or into the fusion objects 42 whose radar objects 38 each correspond to at least one of the lidar objects.

In other words, the sensor objects may be divided into radar-only objects 38, lidar-only objects and fusion objects 42, wherein no corresponding lidar objects were found for radar-only objects 38, no corresponding radar objects were found for lidar-only objects, and a corresponding radar object 38 and at least one corresponding lidar object were found for fusion objects 42. The sensor data may be in the form of the digital fusion list, which includes the sensor objects, in particular fusion objects 42, radar-only objects 38 and/or lidar-only objects.

The radar data may include radar timing information associated with the radar objects 38 and representative of a first time at which the radar objects 38 were detected. The lidar data may include lidar timing information associated with the lidar objects and representative of a second time at which the lidar data was detected. In merging the radar data and the lidar data, the most recent radar data and the most recent lidar data may then be fused into the sensor data. Furthermore, each time new radar data or lidar data is received, this new radar or lidar data may be fused with the most recent lidar or radar data to generate a new fusion list or to update the last fusion list.

A detection of the objects 50 based exclusively on radar data has the advantages over a detection of the objects 50 based exclusively on lidar data that clearly identified objects 50 are consistently recognized as such over time, that the probability that the detected objects 50 are real is relatively high and that a speed of the radar sensor and thus of the motor vehicle 20 can be determined based on the radar data. However, detection based on radar data also has a disadvantage compared to detection based on lidar data, namely that determining the position and/or size of the detected objects 50 is relatively poor.

The detection of the objects 50 based on the lidar data has the advantage over the detection of the objects 50 based on the radar data that contour point positions of the contour points 40 along the contours of the objects 50 can be determined very precisely. However, the detection based on the lidar data also has a disadvantage compared to the detection based on radar data, namely that objects 50 that are actually clearly identified can be recognized again and again as new objects 50 over time, that no conclusion can be drawn about the speed of the lidar sensor and thus of the motor vehicle 20 and that object positions, object dimensions and object orientations of recognized objects 50 are recognized relatively poorly.

Optionally, the fusion objects 42 can be assigned a predefined first probability with which the fusion objects 42 actually exist, for example in the fusion list. Optionally, the radar-only objects can be assigned a predefined second probability with which the radar-only objects actually exist, for example in the fusion list. Optionally, the lidar-only objects can be assigned a predefined third probability with which the lidar-only objects actually exist, wherein the first probability is greater than the second probability and greater than the third probability. If necessary, the determination explained further below of whether one or more of the sensor objects are located in the region of interest 60 and/or the determination of whether one or more of the sensor objects located in the region of interest 60 poses a danger to the motor vehicle 20 can then be carried out depending on the corresponding probabilities.

For example, if one of the objects 50 that is located in front of the motor vehicle 20 in the direction of travel is detected by both radar and lidar, the probability that the corresponding object 50 actually exists and is actually located in front of the motor vehicle 20 in the direction of travel is greater than if the object 50 is detected exclusively by radar or exclusively by lidar. This circumstance can be taken into account by assigning the probabilities and in particular by assigning the greatest probability to the fusion objects 42.

Optionally, for each of the fusion objects 42, it can be checked whether the corresponding fusion object 42 satisfies at least one fusion crossing criterion that is representative of the fact that the corresponding fusion object 42 crosses the roadway 32 before the Motor vehicle 20 crosses. In addition to the corresponding fusion object 42, a corresponding crossing fusion object 42 can then be encoded in the sensor data if the corresponding fusion object 42 satisfies the fusion crossing criterion.

As explained above, the fusion objects 42 can be stored in the fusion list. The fusion objects 42 can initially be marked in the fusion list as not crossing the roadway 32. If the corresponding crossing fusion object 42 is found for one of the fusion objects 42, the crossing fusion object 42 can be stored in the fusion list in addition to the corresponding fusion object 42. Optionally, it can be checked whether the corresponding fusion object 42 meets two or more of the fusion crossing criteria, each of which is representative of the fact that the corresponding fusion object 42 crosses the roadway 32 in front of the motor vehicle 20. In this case, the crossing fusion object 42 can be encoded in the sensor data in addition to the corresponding fusion object 42 if the corresponding fusion object 42 meets one, two or more of the fusion crossing criteria.

The fusion crossing criterion(s) to be checked may, for example, originate from a group comprising the group: the corresponding lidar object is relevant; the corresponding lidar object has an age that is greater than a predetermined age threshold, for example greater than 100 ms, for example greater than 500 ms, for example greater than 1000 ms; a lateral velocity of the corresponding lidar object is greater than a predetermined lidar lateral velocity threshold, for example greater than 1 m/s, for example greater than 1.5 m/s, for example greater than 1.75 m/s; all contour points 40 belonging to the corresponding lidar object do not overlap with any moving radar object 38; a probability of existence of the corresponding radar object 38 is greater than a predetermined radar object existence threshold, for example greater than 80%, for example greater than 85%, for example greater than 90%; a lateral velocity of the corresponding radar object 38 is greater than a predetermined radar lateral velocity threshold, for example greater than 1 m/s, for example greater than 1.5 m/s, for example greater than 1.75 m/s; a longitudinal velocity of the corresponding radar object 38 is less than a predetermined radar longitudinal velocity threshold, for example less than 10 m/s, for example less than 5 m/s, for example less than 2 m/s; the corresponding radar object 38 is not stationary.

The age of the lidar object refers to how many lidar data that were recorded consecutively immediately before the current lidar data already contain the corresponding lidar object. In other words, the age of the lidar object refers to how many lidar object lists that were created consecutively immediately before the current lidar object list already contain the corresponding lidar object. The more lidar data or lidar object lists the corresponding lidar object is contained in, the older the corresponding lidar object is.

The probability of existence of the radar object can be determined, for example, depending on the age of the radar object. The age of the radar object can, for example, be determined corresponding to the age of the lidar object.

Optionally, for each of the lidar-only objects, it can be checked whether the corresponding lidar-only object satisfies at least one lidar crossing criterion that is representative of the fact that the corresponding lidar-only object crosses the roadway 32 in front of the motor vehicle 20. If necessary, in addition to the lidar-only object, a corresponding crossing lidar-only object can be encoded in the sensor data if the corresponding lidar-only object satisfies the lidar crossing criterion.

As explained above, the lidar-only objects can be stored in the fusion list. The lidar-only objects can be marked in the fusion list as not crossing the roadway 32. If the corresponding lidar crossing fusion object 42 is found for one of the lidar-only objects, the lidar crossing fusion object 42 can be stored in the fusion list in addition to the corresponding lidar-only object. Optionally, it can be checked whether the corresponding lidar-only object meets two or more lidar crossing criteria, each of which is representative of the corresponding lidar-only object crossing the roadway 32 in front of the motor vehicle 20. In this case, the crossing lidar-only object can be encoded in the sensor data in addition to the corresponding lidar-only object if the corresponding lidar-only object meets one, two or more of the lidar crossing criteria.

The lidar crossing criteria(s) to be checked may, for example, come from a group comprising: an age of the lidar-only object is greater than a predefined lidar age threshold, for example greater than 100 ms, for example greater than 500 ms, for example greater than 1000 ms; a distance of the lidar-only object to the motor vehicle 20 is greater than a predefined lidar distance threshold, from a center Z (see 9 ) of the motor vehicle 20 measured from, for example, greater than 1 m, for example greater than 5 m, for example approximately 6 m; a length of the lidar-only object is less than a predetermined length threshold (it is unlikely that there is no corresponding radar object 38 for a very long lidar object), for example, less than 3 m, for example, less than 2 m, for example, less than 1.5 m; a width of the lidar-only object is less than a predetermined width threshold (it is unlikely that there is no corresponding radar object 38 for a very wide lidar object), for example, less than 3 m, for example, less than 2 m, for example, less than 1.5 m; a lateral velocity of the lidar-only object is greater than a predetermined lidar lateral velocity threshold, for example, greater than 1 m/s, for example, greater than 1.5 m/s, for example, greater than 1.75 m/s; a relative lateral velocity of the lidar-only object is less than a predetermined lidar relative lateral velocity threshold, for example less than 5 m/s, for example less than 2 m/s, for example approximately 1.75 m/s; a number of contour points 40 belonging to the corresponding lidar-only object is less than a predetermined lidar-only number threshold (it is unlikely that no corresponding radar object 38 exists for a very large lidar object), for example less than 20, for example less than 15, for example less than approximately 11.

The age of one of the lidar-only objects can be determined from historical lidar data by checking whether the corresponding lidar object is already encoded in previously acquired lidar data, in particular in previous lidar data that was acquired consecutively and immediately before the current lidar data. The age can be assumed to be longer the more previous lidar data the corresponding lidar object is encoded in. In other words, the age of one of the lidar-only objects can be assumed to be longer the more previous lidar object lists the corresponding lidar object is included in.

Optionally, for each fusion object 42, radar-only object and/or lidar-only object in the fusion list, it can be checked whether the corresponding object 50 meets one or more relevance criteria. The corresponding fusion object 42, radar-only object or lidar-only object can be marked as relevant in the fusion list if the corresponding fusion object 42, radar-only object or lidar-only object meets one or more of the relevance criteria. The corresponding fusion object 42, radar-only object or lidar-only object can be marked as not relevant if the corresponding fusion object 42, radar-only object or lidar-only object does not meet any of the relevance criteria.

The relevance criterion(s) to be checked may, for example, originate from a group, the group comprising a first subgroup relating to the radar-only objects 38 and a second subgroup relating to lidar-only objects and fusion objects 42.

The first subgroup may include: a probability of existence of the corresponding radar object 38 is greater than a predetermined probability of existence threshold; the corresponding radar object 38 is included in the current radar object list; an amount of a cross-sectional area (Radar Cross Section (RCS)) of the corresponding radar object 38 is greater than a predetermined cross-sectional threshold, for example greater than 1 dBm ² , for example greater than 3 dBm ² , for example greater than 5 dBm ² ; information about a reliability of one or more measured values describing dynamic properties, for example the speed; the corresponding radar object 38 does not belong to a class that includes large objects, such as trucks or vans; the corresponding radar object 38 is stationary (it is very unlikely that there is no corresponding lidar object for a stationary radar object 38).

The second subgroup may comprise: a number of contour points 40 belonging to the corresponding lidar-only object is greater than a predetermined number threshold, for example greater than 2, for example greater than 3; a density of contour points 40 belonging to the corresponding lidar-only object is greater than a predetermined density threshold (if the distance of the corresponding lidar-only object to the motor vehicle 20 is less than a predetermined distance threshold, for example less than 2 m, for example less than 1 m, for example approximately 0.35 m); there is more than one contour point 40 to the corresponding lidar-only object.

The age of one of the lidar-only objects can be determined from historical lidar data by checking whether the corresponding lidar object is already encoded in previously acquired lidar data, in particular in previous lidar data that was acquired consecutively and immediately before the current lidar data. The age can be assumed to be longer the more previous lidar data the corresponding lidar object is encoded in. In other words, the age of one of the lidar-only objects can be assumed to be greater the more previous lidar object lists the corresponding lidar object is included in.

In a step S46, the position data representative of the current vehicle position of the motor vehicle 20 can be received. The position data can be received corresponding to the step S22 explained above.

In a step S48, the speed data can be received that is representative of a current speed of the motor vehicle 20. The speed data can be received corresponding to the step S24 explained above.

In a step S50, the region of interest 60, which is located on the roadway 32 in the direction of travel in front of the motor vehicle 20, can be determined depending on the position data and/or the speed data, for example position-based or speed-based, as explained in more detail below. The step S50 can be processed corresponding to the step S6 explained above. In particular, the step S50 can be processed corresponding to the steps S26, S28 and S30 explained above.

In a step S52, depending on the sensor data, it can be checked for each of the sensor objects in the fusion list whether the corresponding sensor object is located in the region of interest 60 or not. The step S52 can be processed corresponding to the step S8 or S32 explained above.

In an optional step S54, it can be determined whether one or more of the sensor objects located in the area of interest 60 pose a danger to the motor vehicle 20. The step S54 can be processed corresponding to the step S10 or S34 explained above.

If in the foregoing the first probability is assigned to the fusion objects 42, the second probability to the radar-only objects and/or the third probability to the lidar-only objects, then optionally only the fusion objects, only the radar-only objects or only the lidar-only objects whose assigned probability is greater than a predetermined probability threshold can be taken into account in steps S52 and/or S54.

If the fusion objects 42, the radar-only objects and/or the lidar-only objects are checked in the preceding to see whether they are relevant, for example based on the relevance criteria, then optionally only the fusion objects, only the radar-only objects or only the lidar-only objects that were assessed as relevant can be taken into account in steps S52 and/or S54.

In a step S56, the control signal for the braking device 24 of the motor vehicle 20 can be generated, for example corresponding to the steps S12 or S36 explained above.

Position-based determination of ROI

6 shows a plan view of an exemplary driving situation, of the motor vehicle 20 and of the area of interest 60 in front of the motor vehicle 20. The motor vehicle 20 has a width B. The motor vehicle 20 moves on the roadway 32, in particular on the first lane 34, in 6 upwards in the direction of travel. The region of interest 60 is delimited on its lateral outer sides by a first outer boundary 56 and a second outer boundary 58. The first outer boundary 56 is defined by interconnected first outer points 52 and the second outer boundary 58 is defined by interconnected second outer points 54.

The length I _ROI (in the longitudinal direction, perpendicular to the lateral direction) of the region of interest 60 (see 10 ) can be fixed or determined depending on the speed of the motor vehicle 20, for example by means of the method explained below for avoiding or reducing the collision of the autonomous motor vehicle 20 traveling on the roadway 32 with the object 50.

7 shows a flow chart of an embodiment of a method for position-based determination of the region of interest 60.

In a step S60, the position data can be received which are representative of the current vehicle position and optionally of an orientation of the motor vehicle 20, in other words a vehicle orientation of the motor vehicle 20. The step S60 can be processed corresponding to the step S22 or S46. The position data can be for a front position of a front 62 (see 9 ) of the motor vehicle 20. The region of interest 60 can subsequently be determined such that it extends from the front 62 in the current direction of travel of the motor vehicle 20.

wobei xv(t) die X-Position des Kraftfahrzeugs 20 zum Zeitpunkt t, yv(t) die Y-Position des Kraftfahrzeugs 20 zum Zeitpunkt t und ψv(t) der Gierwinkel bzw. die Fahrrichtung des Kraftfahrzeugs 20 zum Zeitpunkt t sind.The position of the motor vehicle 20 can be specified in a two-dimensional coordinate system in the form of an X and a Y coordinate and the vehicle orientation can be specified as an angle, for example as a yaw angle and/or direction of travel of the motor vehicle 20. The coordinate The coordinate system may, for example, refer to the motor vehicle 20, have the center Z or a center of the front 62 of the motor vehicle 20 as the origin, and/or be referred to as a motor vehicle coordinate system. Alternatively, a proprietary coordinate system or a real-world coordinate system may be used as the coordinate system. Thus, the position data may, for example, be present as: $${\overrightarrow{x}}_V\left(t\right)={\left(x_V\left(t\right),y_V\left(t\right),{\psi }_V\left(t\right)\right)}^{\text{T}}$$

where xv(t) is the X-position of the motor vehicle 20 at time t, yv(t) is the Y-position of the motor vehicle 20 at time t and ψv(t) is the yaw angle or the direction of travel of the motor vehicle 20 at time t.

In an optional step S62, the speed data can be received, for example corresponding to the step S24 or S48 explained above. If the speed data are received, the current speed of the motor vehicle 20 can be determined depending on the speed data and subsequently the length I _ROI (see 9 ) of the region of interest 60 can be determined depending on the speed of the motor vehicle 20. The length I _ROI can, for example, be determined depending on the speed and a collision time duration. The collision time duration can, for example, be determined using a method for determining the collision time duration, as explained in more detail below. Alternatively or additionally, the length I _ROI can be determined depending on the speed using a length lookup table in which suitable lengths I _ROI of the region of interest 60 are assigned to different speeds. The length I _ROI can, for example, extend from the front 62 of the motor vehicle 20 to a longitudinal end of the region of interest 60.

In a step S64, reference point positions of reference points 94, 96 (see 14 ) along the lane, for example the first lane 34. The reference points 94, 96 can be representative of a course of the lane 34, 36 in the direction of travel in front of the motor vehicle 20. For example, the reference points 94, 95 can be positioned along a center of the corresponding lane 34, 36. The reference point positions can be determined depending on the position data using a lane lookup table in which the reference point positions are assigned to the reference points 94, 96.

wobei x_L die X-Koordinate des entsprechenden Referenzpunkts der Fahrspur L, y_L die Y-Koordinate des entsprechenden Referenzpunkts der Fahrspur L, ψ_L die Fahrspurorientierung der Fahrspur L an dem entsprechenden Referenzpunkt und I_L die Länge der Fahrspur L bis zu dem entsprechenden Referenzpunkt sind. Die Fahrspur L kann beispielsweise die erste oder die zweite Fahrspur 34, 36 sein, wobei in diesem Zusammenhang der Index L der Fahrspur 34, 36 beispielsweise gleich 1 für die erste Fahrspur 34 oder 2 für die zweite Fahrspur 36 sein kann.The reference point positions can each be specified in a two-dimensional coordinate system in the form of an X and a Y coordinate, and an orientation of the corresponding lane 34, 36, in other words a lane orientation, can be specified as an angle. The coordinate system can be, for example, a proprietary coordinate system or a real-world coordinate system, wherein the position of the motor vehicle 20 and the reference point positions are either specified in the same coordinate system or are transferred to the same coordinate system. Thus, the reference point positions can be, for example, present as: $${\overrightarrow{x}}_L=\left(\begin{array}{cccc}{x}_L& y_L& {\psi }_L& {\mathcal{l}}_L\end{array}\right)$$

where x _L is the X coordinate of the corresponding reference point of the lane L, y _L is the Y coordinate of the corresponding reference point of the lane L, ψ _L is the lane orientation of the lane L at the corresponding reference point and I _L is the length of the lane L up to the corresponding reference point. The lane L can be, for example, the first or the second lane 34, 36, wherein in this context the index L of the lane 34, 36 can be, for example, equal to 1 for the first lane 34 or 2 for the second lane 36.

The motor vehicle 20 can, for example, have a separate lane lookup table for each lane 34, 36 on which the motor vehicle 20 moves, so that the motor vehicle 20 can determine the area of interest 60 on a position-based basis in each lane 34, 36. The corresponding lane lookup tables can be stored in the motor vehicle 20, for example in the memory unit 26 of the control unit 25, or made available to the motor vehicle 20 via a communication connection, for example via the Internet. The lane lookup tables can be created in advance depending on the real lanes 34, 36, for example by measuring the corresponding lanes 34, 36 and extracting the relevant measurement results and storing them in the corresponding lane lookup tables. The lane lookup tables can also have information about one or more curves, for example corresponding curvature data.

The reference point positions of the reference points 94, 96 can be specified, for example, by GPS coordinates in the lane lookup table. The reference point positions and optionally the reference points 94, 96 can be provided with indices in the lane lookup table and ordered accordingly. For example, the lane lookup table can have an index for each reference point and a reference point position of the reference point assigned to the index, for example within a row of the Lane lookup table. An order of the indices can correspond to an order of the reference points 94, 96 on the lane 34, 36 in the real world, with the length I _L of the lane up to the corresponding reference point increasing from reference point to reference point. The length I _L of the lane up to the corresponding reference point can then also be specified as a number of reference points, taking into account the distance between the reference points. Likewise, the length I _ROI of the region of interest can then be specified as a number of reference points.

The lane 34, 36 for which the region of interest 60 is to be determined may be a part of the roadway 32, which has the two or more lanes 34, 36. The region of interest 60 may then be determined to be located in the lane 34, 36 in which the motor vehicle 20 is currently traveling, for example in the first lane 34. Alternatively, the region of interest 60 may be determined to be located in the lane 34, 36 that is adjacent to the lane 34, 36 in which the motor vehicle 20 is currently traveling, for example in the second lane 36.

In a step S66, the region of interest 60 can be determined depending on the reference points 94, 96 and depending on the width B of the motor vehicle 20. For example, depending on the specified reference points 94, 96, a driving line 90, 92 (see 14 ) on the lane 34, 36, wherein the driving line 90, 92 can be determined by connecting the reference points 94, 96. The reference point positions of the reference points 94, 96 can be specified in the lane lookup table, for example, such that the driving line 90, 92 runs in the middle of the corresponding lane 34, 36. The driving line 90, 92 can, for example, run parallel to the center line 35 of the lane 32 and/or parallel to one of the two side edges 37, 39 of the lane 32. For example, the reference points 94, 96 can be selected such that they are at a distance along the driving line 90, 92 of, for example, 1 cm to 20 cm, for example from 5 cm to 15 cm, for example 10 cm. The width B of the motor vehicle 20 is normally fixed depending on an actual width B of the motor vehicle 20. The width B can, for example, be stored in the memory unit 26.

In detail, the region of interest 60 can be determined depending on the reference points 94, 96 and depending on the width B of the motor vehicle 20 by defining the first outer boundary 56 of the region of interest 60 on one side of the driving line 90, 92 and the second outer boundary 58 of the region of interest 60 on the other side of the driving line 90, 92, wherein the outer boundaries 56, 58 each have a predetermined distance from the driving line 90, 92, and wherein the outer boundaries (56, 58) each run parallel to the driving line (90, 92).

In particular, the outer boundaries 56, 58 can be determined by determining a first outer point 52 for the first outer boundary 56 and a second outer point 54 for the second outer boundary 58 for each of the reference points 94, 96. The outer points 52, 54 can be determined such that they each have the predetermined distance from the corresponding reference point 94, 96 and that they each lie on a straight line that runs through the corresponding reference point 94, 96 and that is perpendicular to a tangent to the driving line 90, 92 at the corresponding reference point 94, 96. The first outer points 52 can then be connected to one another to form the first outer boundary 56, and the second outer points 54 can be connected to one another to form the second outer boundary 58.

The predetermined distance can correspond in each case to at least half the width B of the motor vehicle 20. The predetermined distance can, for example, be predetermined such that it corresponds to the width B of the motor vehicle 20 plus an additional tolerance range, wherein the tolerance range can lie in a range, for example, from 1 cm to 50 cm, for example 5 cm to 25 cm, for example approximately 10 cm, i.e. approximately 5 cm on each side of the motor vehicle 20. In other words, the width of the region of interest 60 can be predetermined such that it is slightly larger than the width B of the motor vehicle 20. The additional tolerance range can contribute to the safety of the motor vehicle 20.

If the area of interest 60, as with reference to the 6 and 7 explained, is determined position-based, it can be particularly advantageous to use the following with reference to 8th explained method to check whether one or more of the objects 50 are located in the area of interest 60. This applies, for example, because the outer points 52, 54 can be used for this check, which are available anyway in the position-based determination.

8th shows a flow chart of an embodiment of a method for determining whether the object 50 is located in the region of interest 60 in the direction of travel in front of the motor vehicle 20, which is particularly advantageous when the area of interest 60 is determined position-based.

In a step S67, the sensor data can be received, for example corresponding to one of the steps S2 or S20 explained above.

In a step S68, the region of interest 60 can be determined based on position, as explained above.

In a step S69, two or more triangles can be determined which together cover the region of interest 60. The triangles can be determined, for example, using the outer points 52, 54. In other words, the region of interest 60 can be divided into triangles which together form the region of interest 60, and the check as to whether the object 50 is located in the region of interest 60 or not can be carried out separately for each of these triangles. The triangles can have first triangles and second triangles, wherein one of the triangles of the region of interest 60, in particular one of the second triangles, in 6 is shown.

The first triangles can each be determined using two first outer points 52 on the first outer boundary 56 and a second outer point 54 on the second outer boundary 58. The first outer points 52 can be indexed using the running variable i, which is a natural number, and the second outer points 54 can be indexed using the running variable j, which is a natural number. Thus, the first triangles can be determined, for example, using the j-th second outer point 54 and using the i-th and (i+1)-th first outer points 56. In contrast, the second triangles can each be determined using two of the second outer points 54 on the second outer boundary 58 and one of the first outer points 52 on the first outer boundary 56. Thus, the second triangles can be determined, for example, using the i-th first outer point 52 and using the j-th and (j+1)-th second outer points 54, as in 6 shown on one of the triangles.

How many triangles are determined based on the outer points 52, 54 in order to cover the region of interest 60 can be determined based on the length I _ROI of the region of interest. The first and second outer points 52, 54 for determining the triangles can be selected such that the triangles do not overlap one another and completely cover the region of interest 60.

In a step S70, it can be checked sequentially, i.e. for each of the determined triangles one after the other, whether one or more of the objects 50 are located in one or more of the triangles, until all triangles have been checked.

Subsequently, it can optionally be checked whether one or more of the objects in the area of interest 60 pose a danger to the motor vehicle, for example for the entire area of interest 60 or again for each of the triangles separately.

Speed-based ROI determination

9 shows a vehicle model of the motor vehicle 20 for speed-based determination of the region of interest 60. The motor vehicle 20 has a front 62 and a rear 64. A coordinate system related to the motor vehicle 20, in other words a motor vehicle coordinate system, has a longitudinal axis that corresponds to a longitudinal axis V of the motor vehicle. The motor vehicle coordinate system is selected such that the center Z of the motor vehicle 20, also referred to as the vehicle reference point, lies at the origin of the motor vehicle coordinate system.

The motor vehicle 20 can have four-wheel steering. In this case, the front and rear axles of the motor vehicle 20, in particular the corresponding wheels 23, can steer independently of one another. In this case, solely taking into account a longitudinal speed v _long and an angular speed or yaw rate ω _ego of the motor vehicle 20 to determine the region of interest 60 is not useful, since this does not take into account the sideways movement of the motor vehicle 20 during the special driving maneuvers.

• - ein Seitenwinkel δ_c, der den Winkel zwischen dem virtuellen Fahrrad und der longitudinalen Achse V beschreibt;
• - ein Lenkwinkel δ_s, der den Lenkwinkel des virtuellen Fahrrads mit Vorderradlenkung beschreibt;
• - eine Länge I_F, die einen Abstand zwischen den Vorder- und Hinterrädern des virtuellen Fahrrads darstellt, wobei die Länge I_F gleich dem Abstand zwischen dem Zentrum Z und der Vorderachse des Kraftfahrzeugs 20 gewählt wird.

An elegant approach to also take the sideways movement into account arises from the way Forces parameterizes the vehicle movement, for example in “A generic vehicle controller”, by AJ de Graaf, Technical Report (unnumbered), Frog Navigation Systems BV, September 2003. Unfinished; and/or in “Steering pole”, AJ de Graaf, Technical Report (unnumbered), 2getthere BV, July 2012. In particular, Forces defines the concept of a “virtual steering wheel”, which, in the context of determining the region of interest 60, can be interpreted as a kinematic bicycle model of a virtual bicycle with front wheel steering. The rear wheel center of the virtual bicycle coincides with the vehicle reference point P, which in the case of the motor vehicle 20 can be chosen as the geometric vehicle center. This concept is in 9 which shows a plan view of the motor vehicle 20 and the vir current bicycle. From 9 It can be seen that a position and attitude of the virtual bicycle in the vehicle coordinate system with the longitudinal axis V and with the center Z of the motor vehicle 20 is determined by the following variables:

• - a lateral angle δ _c describing the angle between the virtual bicycle and the longitudinal axis V;
• - a steering angle δ _s , which describes the steering angle of the virtual bicycle with front wheel steering;
• - a length I _F representing a distance between the front and rear wheels of the virtual bicycle, the length I _F being chosen to be equal to the distance between the center Z and the front axle of the motor vehicle 20.

For a given length I _F of the frame of the virtual bicycle, the angles δ _c and δ _s are such that an instantaneous center of rotation (ICR) of the virtual bicycle corresponds to that of the motor vehicle 20, whereby a simple representation of the movement of the motor vehicle 20 can be achieved, which provides a basis for determining the region of interest 60.

It should be noted that although this concept can describe the sideways motion of the motor vehicle 20 using the side angle δ _c , it cannot describe a changing side angle δ _c . Therefore, a derivative of the side angle δ _c is assumed to be equal to zero. This assumption implies that the angular velocity of the motor vehicle 20, also referred to as the yaw rate ω _ego , is identical to the yaw rate ω of the virtual bicycle.

The longitudinal velocity v _long and the lateral velocity v _lat of the center Z as well as the yaw rate ω _ego are available to the control unit 25. The side angle δ _c is then calculated from $${\delta }_{\text{c}}=\text{atan2}\left({\upsilon }_{\text{lat}},{\upsilon }_{\text{long}}\right)$$

und wobei die Grenze dann allgemein angegeben werden kann mit $${\delta }_{\text{lim}}\left(\upsilon \right)=\{\begin{array}{cc}{\delta }_{\text{max}},& \upsilon \ge {\upsilon }_{\text{lim}}\\ \frac{{\delta }_{\text{max}}}{{\upsilon }_{\text{lim}}}\upsilon ,& \upsilon <{\upsilon }_{\text{lim}},\end{array}$$

und wobei v_lim << 1 m/s eine empirisch oder mittels Simulation abstimmbare Geschwindigkeitsgrenze und δ_max der maximale Lenkwinkel des Kraftfahrzeugs 20 ist.At a low speed v _ego of the motor vehicle 20, the calculation of the side angle δ _c may be inaccurate due to measurement noise in practice or numerical inaccuracies in a corresponding simulation. This problem can be mitigated by setting the side angle δ _c to zero as the speed decreases. For example, an upper limit δ _lim (v) and a lower limit -δ _lim (v) of the side angle δ _c can be specified as a function of the speed v of the motor vehicle 20, where the speed v of the virtual bicycle is given by $$\upsilon =\sqrt{{\upsilon }_{\text{long}}^2+{\mathrm{<figure-callout\; id="2"\; label="\upsilon \; lat"\; filenames="DE102022212897A1\_0068.png,DE102022212897A1\_0069.png"\; state="\{\{state\}\}">\upsilon </figure-callout>}}_{\text{lat}}^{}}$$

and the limit can then be generally specified as $${\delta }_{\text{lim}}\left(\upsilon \right)=\{\begin{array}{cc}{\delta }_{\text{Max}},& \upsilon \ge {\upsilon }_{\text{lim}}\\ \frac{{\delta }_{\text{Max}}}{{\upsilon }_{\text{lim}}}\upsilon ,& \upsilon <{\upsilon }_{\text{lim}},\end{array}$$

and where v _lim << 1 m/s is a speed limit that can be adjusted empirically or by simulation and δ _max is the maximum steering angle of the motor vehicle 20.

$${\delta }_c=\{\begin{array}{ll}-{\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_c<-{\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\tilde{\delta }}_{\text{c}},\hfill & -{\delta }_{\text{lim}}\left(\upsilon \right)\le {\tilde{\delta }}_{\text{c}}\le {\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_c>{\delta }_{\text{lim}}\left(\upsilon \right).\hfill \end{array}$$

Thus, the side angle δ _c can be determined as follows: $${\tilde{\delta }}_c=\text{atan2}\left({\upsilon }_{\text{lat}},{\upsilon }_{\text{long}}\right)$$

$${\delta }_c=\{\begin{array}{ll}-{\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_c<-{\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\tilde{\delta }}_{\text{c}},\hfill & -{\delta }_{\text{lim}}\left(\upsilon \right)\le {\tilde{\delta }}_{\text{c}}\le {\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_c>{\delta }_{\text{lim}}\left(\upsilon \right).\hfill \end{array}$$

wobei v der Geschwindigkeitsvektor des Zentrums Z ist, ω der Winkelgeschwindigkeitsvektor und r_vs ein Vektor der Länge I_F des Rahmens in Längsrichtung des virtuellen Fahrrads ist.To obtain the steering angle δ _s , the speed v _s of a center S of the front wheel of the virtual bicycle can be determined. If v _s is interpreted as the speed vector of this center S in the real number space, the result is $${\upsilon }_{\text{s}}=\upsilon +\omega \times r_{\text{vs}}$$

where v is the velocity vector of the center Z, ω is the angular velocity vector and r _vs is a vector of the length I _F of the frame in the longitudinal direction of the virtual bicycle.

Assuming a coordinate system consisting of the longitudinal and lateral axes of the vehicle coordinate system and a Z-axis perpendicular to the vehicle plane, v _s can be expressed by its three components, i.e. $${\upsilon }_{\text{s}}^{\text{T}}=\left[\begin{array}{ccc}{\upsilon }_{\text{s}\text{,long}}& {\upsilon }_{\text{s}\text{,lat}}& {\upsilon }_{\text{s}\text{,2}}\end{array}\right]$$

$${\omega }^{\text{T}}=\left[\begin{array}{ccc}0& 0& \omega \end{array}\right]$$

$$r_{\text{vs}}^{\text{T}}=\left[\begin{array}{ccc}\mathcal{l}\text{cos}{\delta }_{\text{c}}& \mathcal{l}\text{sin}{\delta }_{\text{c}}& 0\end{array}\right]$$

In this coordinate system it also follows that $${\upsilon }^{\text{T}}=\left[\begin{array}{ccc}{\upsilon }_{\text{long}}& {\upsilon }_{\text{lat}}& 0\end{array}\right]$$

$${\omega }^{\text{T}}=\left[\begin{array}{ccc}0& 0& \omega \end{array}\right]$$

$$r_{\text{vs}}^{\text{T}}=\left[\begin{array}{ccc}\mathcal{l}\text{cos}{\delta }_{\text{c}}& \mathcal{l}\text{sin}{\delta }_{\text{c}}& 0\end{array}\right]$$

und die Geschwindigkeit v_s des Vorderrads des virtuellen Fahrrads ergibt sich zu $${\upsilon }_{\text{s}}=\Vert {\upsilon }_{\text{s}}\Vert =\sqrt{{\upsilon }_{\text{s}\text{,long}}^2+{\upsilon }_{\text{s}\text{,lat}}^2}$$

The result is $$\left[\begin{array}{c}{\upsilon }_{\text{s}\text{,long}}\\ {\upsilon }_{\text{s}\text{,lat}}\\ {\upsilon }_{\text{s}\text{,z}}\end{array}\right]=\left[\begin{array}{c}{\upsilon }_{\text{long}}\\ {\upsilon }_{\text{lat}}\\ 0\end{array}\right]+\omega \mathcal{l}\left[\begin{array}{c}-\text{sin}{\delta }_{\text{c}}\\ \text{cos}{\delta }_{\text{c}}\\ 0\end{array}\right]$$

and the speed v _s of the front wheel of the virtual bicycle is $${\upsilon }_{\text{s}}=\Vert {\upsilon }_{\text{s}}\Vert =\sqrt{{\upsilon }_{\text{s}\text{,long}}^2+{\upsilon }_{\text{s}\text{,<figure-callout id="2" label="lat" filenames="DE102022212897A1\_0068.png,DE102022212897A1\_0069.png" state="{{state}}">lat</figure-callout>}}^2}$$

Since the virtual bicycle by definition performs a purely kinematic movement, an orientation of the front wheel of the virtual bicycle corresponds to the direction of v _s . Therefore, the steering angle δ _s can be considered as the angle between v and v _s , which can be determined as follows $${\delta }_{\text{s}}=\text{atan2}\left({\upsilon }_{\text{s}\text{,lat}},{\upsilon }_{\text{s}\text{,long}}\right)-{\delta }_{\text{c}}$$

$${\delta }_{\text{s}}=\{\begin{array}{ll}-{\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_{\text{s}}<-{\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\tilde{\delta }}_{\text{s}},\hfill & -{\delta }_{\text{lim}}\left(\upsilon \right)\le {\tilde{\delta }}_{\text{s}}\le {\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_{\text{s}}>{\delta }_{\text{lim}}\left(\upsilon \right),\hfill \end{array}$$

Außerdem kann der Seitenwinkel δ_c entweder mit oder ohne die im Vorhergehenden erwähnten Grenzen bestimmt werden.This results in $${\tilde{\delta }}_{\text{s}}=\text{atan2}\left({\upsilon }_{\text{s}\text{,lat}},{\upsilon }_{\text{s}},{\upsilon }_{\text{s}\text{,long}}\right)-{\delta }_{\text{c}}$$

$${\delta }_{\text{s}}=\{\begin{array}{ll}-{\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_{\text{s}}<-{\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\tilde{\delta }}_{\text{s}},\hfill & -{\delta }_{\text{lim}}\left(\upsilon \right)\le {\tilde{\delta }}_{\text{s}}\le {\delta }_{\text{lim}}\left(\upsilon \right)\hfill \\ {\delta }_{\text{lim}}\left(\upsilon \right),\hfill & {\tilde{\delta }}_{\text{s}}>{\delta }_{\text{lim}}\left(\upsilon \right),\hfill \end{array}$$

In addition, the side angle δ _c can be determined either with or without the limits mentioned above.

However, there may also be control units 25 for which the lateral angle δ _c , the steering angle δ _s and/or the speed v _s of the front wheel of the virtual bicycle are known, for example due to a safety function of the control unit, so that these values are available directly and do not have to be determined as explained above.

Thus, the region of interest 60 can be completely determined by the speed v = ||v||, the yaw rate ω = ||ω|| and the orientation of the virtual bicycle with respect to the motor vehicle 20, which corresponds to the azimuth angle δ _c . Thus, the speed v can be specified using the component v _s in the longitudinal direction of the virtual bicycle as follows: $$\upsilon ={\upsilon }_{\text{s}}\text{cos}{\delta }_{\text{s}}$$

Similarly, the yaw rate ω can be determined based on the component v _s caused by rotation of the virtual bicycle and given as $$\omega =\frac{{\upsilon }_{\text{s}}\text{sin}{\delta }_{\text{s}}}{\mathcal{l}}$$

For this and the following calculations, it is assumed that the motor vehicle 20 is traveling in a forward direction, namely in the first direction of travel. Traveling in a reverse direction requires a small adjustment of the calculations, as explained below.

The region of interest 60 is determined by the expected route of the motor vehicle 20, represented by the virtual bicycle. To this end, the region of interest 60 covers an area that is "curved" along an arc, with a width that increases with increasing distance from the motor vehicle 20, and the determined, finite length I _ROI . A center of the region of interest 60 coincides with the instantaneous center of rotation ICR of the motor vehicle 20.

10 shows a schematic diagram for explaining the speed-based determination of the region of interest 60, in particular an example of the region of interest 60, together with a number of relevant variables. In 10 Two relevant coordinate systems are also shown: first, a coordinate system P of the virtual bicycle, the origin of which lies at the center of the rear axle of the virtual bicycle and is aligned with the longitudinal axis of the bicycle, and second, a rotated coordinate system S of the virtual bicycle, which is aligned tangentially to a center line of the region of interest 60 and the origin of which lies at a center of a handlebar of the virtual bicycle.

The region of interest 60 only applies to objects 50 located in front of the motor vehicle 20, i.e. ^P x _t ≥ l _F , where ^P x _t is the x-coordinate of the object 50 in the coordinate system P and the length I _F is the distance between the center of the rear wheel and the center of the front wheel of the virtual bicycle, the length I _F thus corresponding to the length of the frame of the virtual bicycle. To determine whether one of the objects 50 is located in the region of interest 60, a target distance d _t from the motor vehicle 20 to the object 50 along the centerline of the region of interest 60 with respect to the origin of the coordinate system S and a target radius R _t from the center of rotation ICR to the object 50 can be determined.

wobei R_f ein Radius eines Kreises mit dem Ursprung im Rotationszentrum ICR durch die Mitte des Vorderrads des virtuellen Fahrrades ist und α_t > 0 der Winkel zwischen einer vom Rotationszentrum ICR durch die Vorderradmitte und einer Linie vom Rotationszentrum ICR, die auf das Objekt 50 zeigt, ist. Daraus ergibt sich $$R_{\text{f}}=\sqrt{R_{\text{ICR}}^2+{\mathcal{l}}^2}$$

wobei RIC_R ein Radius einer Bewegung des Hinterrads des virtuellen Fahrrads um das Rotationszentrum ICR ist, und $${\alpha }_{\text{t}}=\text{atan}2\left({}^{\text{S}}x{}_{\text{t}},R_{\text{f}}-\text{sgn}\left(\omega \right){}^{\text{S}}y{}_{\text{t}}\right)$$

ist, wobei (^Sx_t, ^Sy_t) die Zielposition im gedrehten Koordinatensystem S und ω die Gierrate des Fahrrads sind, die der Gierrate ω_ego des Kraftfahrzeugs 20 entspricht. Die Drehrichtung des virtuellen Fahrrads wird durch die Vorzeichenfunktion der Gierrate ω berücksichtigt, die im Gegenuhrzeigersinn positiv ist.The target distance d _t can be specified as $$d_{\text{t}}={\alpha }_{\text{t}}{R}_{\text{e}}$$

where R _f is a radius of a circle with the origin at the rotation center ICR through the center of the front wheel of the virtual bicycle and α _t > 0 is the win angle between a line from the rotation center ICR through the front wheel center and a line from the rotation center ICR pointing to the object 50. This results in $$R_{\text{e}}=\sqrt{{\mathrm{<figure-callout\; id="2"\; label="R\; ICR"\; filenames="DE102022212897A1\_0068.png,DE102022212897A1\_0069.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\text{ICR}}^{}+{\mathcal{<figure-callout\; id="2"\; label="l"\; filenames="DE102022212897A1\_0068.png,DE102022212897A1\_0069.png"\; state="\{\{state\}\}">l</figure-callout>}}^2}$$

where RIC _R is a radius of movement of the rear wheel of the virtual bicycle around the rotation center ICR, and $${\alpha }_{\text{t}}=\text{atan}2\left({}^{\text{S}}x{}_{\text{t}},R_{\text{e}}-\text{sgn}\left(\omega \right){}^{\text{S}}y{}_{\text{t}}\right)$$

where ( ^S x _t , ^S y _t ) is the target position in the rotated coordinate system S and ω is the yaw rate of the bicycle, which corresponds to the yaw rate ω _ego of the motor vehicle 20. The direction of rotation of the virtual bicycle is taken into account by the sign function of the yaw rate ω, which is positive in the counterclockwise direction.

$$\begin{array}{cc}\left[\begin{array}{c}{}^{\text{S}}x{}_{\text{t}}\\ {}^{\text{S}}y{}_{\text{t}}\end{array}\right]=R\left({\phi }_{\text{f}}\right)\left[\begin{array}{c}{}^{\text{P}}x{}_{\text{t}}-\mathcal{l}\\ {}^{\text{P}}y{}_{\text{t}}\end{array}\right],& \omega <0\left(\text{clockwise}\right),\end{array}$$

wobei (^Px_t, ^Py_t) die Position des Objekts 50 im Koordinatensystem P ist, und $${\phi }_{\text{f}}=\text{arctan}\left(\frac{\mathcal{l}}{R_{\text{ICR}}}\right)$$

der Winkel zwischen der Linie von dem Rotationszentrum ICR durch die Mitte des Hinterrads des virtuellen Fahrrads und der Linie von dem Rotationszentrum ICR durch die Mitte des Vorderrads des virtuellen Fahrrads ist, und R(Θ) eine Rotationsmatrix ist gemäß $$R\left(\theta \right)=\left[\begin{array}{cc}\text{cos}\theta & -\text{sin}\theta \\ \text{sin}\theta & \text{cos}\theta \end{array}\right]$$

The coordinates of the target position in the coordinate system S are equal $$\begin{array}{cc}\left[\begin{array}{c}{}^{\text{S}}x{}_{\text{t}}\\ {}^{\text{S}}y{}_{\text{t}}\end{array}\right]=R\left(-{\phi }_{\text{e}}\right)\left[\begin{array}{c}{}^{\text{P}}x{}_{\text{t}}-\mathcal{l}\\ {}^{\text{P}}y{}_{\text{t}}\end{array}\right],& \omega \ge 0\left(\text{anti-clockwise}\right)\end{array}$$

$$\begin{array}{cc}\left[\begin{array}{c}{}^{\text{S}}x{}_{\text{t}}\\ {}^{\text{S}}y{}_{\text{t}}\end{array}\right]=R\left({\phi }_{\text{e}}\right)\left[\begin{array}{c}{}^{\text{P}}x{}_{\text{t}}-\mathcal{l}\\ {}^{\text{P}}y{}_{\text{t}}\end{array}\right],& \omega <0\left(\text{clockwise}\right),\end{array}$$

where ( ^P x _t , ^P y _t ) is the position of the object 50 in the coordinate system P, and $${\phi }_{\text{e}}=\text{arctan}\left(\frac{\mathcal{l}}{R_{\text{ICR}}}\right)$$

is the angle between the line from the rotation center ICR through the center of the rear wheel of the virtual bicycle and the line from the rotation center ICR through the center of the front wheel of the virtual bicycle, and R(Θ) is a rotation matrix according to $$R\left(\theta \right)=\left[\begin{array}{cc}\text{cos}\theta & -\text{sin}\theta \\ \text{sin}\theta & \text{cos}\theta \end{array}\right]$$

mit $${\tilde{\phi }}_{\text{f}}=\text{sgn}\left(\omega \right){\phi }_{\text{f}}$$

so dass man eine geeignete Formulierung für die Koordinaten des Objekts 50 erhält.This results in $$\left[\begin{array}{c}{}^{\text{S}}x{}_{\text{t}}\\ {}^{\text{S}}y{}_{\text{t}}\end{array}\right]=R^{\top }\left({\tilde{\phi }}_{\text{e}}\right)\left[\begin{array}{c}{}^{\text{P}}x{}_{\text{t}}-\mathcal{l}\\ {}^{\text{P}}y{}_{\text{t}}\end{array}\right]$$

with $${\tilde{\phi }}_{\text{e}}=\text{sgn}\left(\omega \right){\phi }_{\text{e}}$$

so that one obtains a suitable formulation for the coordinates of the object 50.

wobei v die Longitudinalgeschwindigkeit des virtuellen Fahrrads ist. Da das virtuelle Fahrrad per Definition vorwärtsfährt, gilt, dass v ≥ 0 ist.A remaining unknown variable required for the calculation of d _t is the radius R _ICR with the center of rotation ICR, which can be determined as follows $$R_{\text{ICR}}=\frac{\upsilon }{\left|\omega \right|}$$

where v is the longitudinal velocity of the virtual bicycle. Since the virtual bicycle moves forward by definition, v ≥ 0.

In addition to d _t , the target radius R _t with respect to the center of rotation ICR is also required to determine whether one of the objects 50 is located in the region of interest 60. This target radius R _t is equal to $$R_{\text{t}}=\sqrt{{\left({}^{\text{S}}x{}_{\text{t}}\right)}^2+{\left(R_{\text{e}}-\text{sgn}\left(\omega \right){}^{\text{S}}y{}_{\text{t}}\right)}^2}.$$

At this point, the target distance d _t along the centerline of the region of interest 60 and the target radius R _t are known. However, the above calculations may encounter problems when the motor vehicle 20 is traveling straight (v > 0, ω = 0) and is stationary (v = ω = 0). These two cases are considered separately below.

When driving straight ahead, R _ICR tends to infinity. Since not all control units 25 can handle this, a maximum value R _ICR,max can be introduced for the target radius R _t , which can be set to 10 ⁶ m, for example, which almost corresponds to driving straight ahead. Since a simple limit on R _ICR would still mean that the R _ICR would have to be calculated first, the upper limit can be introduced in another way, as explained below.

First, a concept of “measured yaw rate” ω _m can be introduced, which can either be measured or determined and can be specified as $${\omega }_{\text{m}}=\frac{{\upsilon }_{\text{s}}\text{sin}{\delta }_{\text{s}}}{\mathcal{l}}$$

wobei msgn(x) (mit x = ω_m) eine modifizierte Version der Signum-Funktion bezeichnet: $$\text{msgn}\left(x\right)=\{\begin{array}{cc}\mathrm{1,}& x\ge 0\\ -\mathrm{1,}& x<\mathrm{0,}\end{array}$$

Infolgedessen ist eine Obergrenze R_ICR,max für R_ICR eingeführt, wodurch numerische oder rechnerische Probleme bei der Geradeausfahrt vermieden werden können.A lower limit for the absolute value of the measured yaw rate ω _m can then be set: $$\omega =\{\begin{array}{ll}{\omega }_{\text{m}},\hfill & \left|{\omega }_{\text{m}}\right|\le \frac{\upsilon }{R_{\text{ICR},\text{Max}}}\hfill \\ \text{msgn}\left({\omega }_{\text{m}}\right)\frac{\upsilon }{R_{\text{ICR},\text{Max}}},\hfill & \left|{\omega }_{\text{m}}\right|<\frac{\upsilon }{R_{\text{ICR},\text{Max}}},\hfill \end{array}$$

where msgn(x) (with x = ω _m ) denotes a modified version of the Signum function: $$\text{msgn}\left(x\right)=\{\begin{array}{cc}\mathrm{1,}& x\ge 0\\ -\mathrm{1,}& x<\mathrm{0,}\end{array}$$

As a result, an upper limit R _ICR,max for R _ICR is introduced, which can avoid numerical or computational problems when driving straight ahead.

When braking to a standstill, two further problems may arise with the above calculations. First, at standstill, ie v = 0, ω _m = 0 must hold and thus ω = 0, leaving R _ICR undetermined. Second, both the speed v and the yaw rate ω become very small when braking to a standstill or at low speeds. Consequently, the controller 25 may be very sensitive to steering operations as well as to measurement noise at the speed v and/or ω _m when calculating R _ICR .

wodurch das erste Problem gelöst ist.It can be prevented that R _ICR remains indeterminate at standstill by introducing a minimum speed v _ROI,min > 0, which results in $$\omega =\{\begin{array}{ll}{\omega }_{\text{m}},\hfill & \left|{\omega }_{\text{m}}\right|\ge \frac{{\upsilon }_{\text{ROI},\text{min}}}{R_{\text{ICR},\text{Max}}}\hfill \\ \text{msgn}\left({\omega }_{\text{m}}\right)\frac{{\upsilon }_{\text{ROI},\text{min}}}{R_{\text{ICR},\text{Max}}},\hfill & \left|{\omega }_{\text{m}}\right|<\frac{{\upsilon }_{\text{ROI},\text{min}}}{R_{\text{ICR},\text{Max}}}\hfill \end{array}$$

which solves the first problem.

The second problem, i.e. the sensitivity of the measurements to noise at very low speeds, can be mitigated to some extent by setting R _ICR to R _ICR,max at low speeds. Furthermore, a speed-dependent lower limit R _ICR,lb (v) for R _ICR can be introduced. This gives $$R_{\text{ICR}}=\{\begin{array}{ll}{R}_{\text{ICR},\text{Max}},\hfill & \frac{\upsilon }{\left|\omega \right|}\ge R_{\text{ICR},\text{Max}}\hfill \\ \frac{\upsilon }{\left|\omega \right|},\hfill & R_{\text{ICR},\text{lb}}\left(\upsilon \right)\le \frac{\upsilon }{\left|\omega \right|}<R_{\text{ICR},\text{Max}}\hfill \\ R_{\text{ICR},\text{lb}}\left(\upsilon \right),\hfill & \frac{\upsilon }{\left|\omega \right|}<R_{\text{ICR},\text{lb}}\left(\upsilon \right),\hfill \end{array}$$

Dennoch kann es sein, dass die Grenzen des interessierenden Bereichs 60 bei niedrigen Geschwindigkeiten aufgrund der Empfindlichkeit der Berechnung des R_ICR bei Lenkbewegungen immer noch relativ ungenau werden können. Um eine gleichmäßige Reaktion auf diese Lenkvorgänge zu erzielen, kann die gemessene Gierrate ω_m mit einem Tiefpassfilter gefiltert werden, beispielsweise gemäß dem folgenden diskreten Zeitfilter erster Ordnung: $${\omega }_m\left(k+1\right)=e^{-t_s/{\tau }_{\omega }}{\omega }_m\left(k\right)+\left(1-e^{-t_s/{\tau }_{\omega }}\right){\omega }_{\text{m},\text{r}}\left(k\right)$$

wobei ω_m,r eine "rohe" gemessene Gierrate, t_s eine Abtastzeit des Steuergeräts 25 und τ_ω eine Zeitkonstante sind, und k ∈ Z die diskrete Zeit bezeichnet. Die gleiche Art von Filter kann auch auf die Longitudinalgeschwindigkeit v_long angewandt werden.Furthermore, R _ICR,min can be chosen as a minimum radius, for example equal to a minimum turning radius of the motor vehicle 20, whereby R _ICR,lb (v) can be given by $$R_{\text{ICR},\text{lb}}\left(\upsilon \right)=\{\begin{array}{ll}{R}_{\text{ICR},\text{min}},\hfill & \upsilon \ge {\upsilon }_{\text{ROI},\text{min}}\hfill \\ R_{\text{ICR},\text{Max}}-\frac{R_{\text{ICR},\text{Max}}-R_{\text{ICR},\text{min}}}{{\upsilon }_{\text{ROI},\text{min}}}\hfill & \upsilon <{\upsilon }_{\text{ROI},\text{min}},\hfill \end{array}$$

However, the boundaries of the region of interest 60 may still become relatively inaccurate at low speeds due to the sensitivity of the calculation of the R _ICR to steering movements. To achieve a smooth response to these steering movements, the measured yaw rate ω _m may be filtered with a low-pass filter, for example according to the following first order discrete time filter: $${\omega }_m\left(k+1\right)=e^{-t_s/{\tau }_{\omega }}{\omega }_m\left(k\right)+\left(1-e^{-t_s/{\tau }_{\omega }}\right){\omega }_{\text{m},\text{r}}\left(k\right)$$

where ω _{m,r is} a "raw" measured yaw rate, t _{s is} a sampling time of the controller 25, τ _ω is a time constant, and k ∈ Z denotes the discrete time. The same type of filter can also be applied to the longitudinal velocity v _long .

$${}^{\text{P}}y{}_{\text{l}}=\text{sgn}\left(\omega \right)\left(R_{\text{ICR}}-\left(R_{\text{f}}-\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{cos}\phi \right)$$

$${}^{\text{P}}x{}_{\text{r}}=\left(R_{\text{f}}+\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{sin}\phi$$

$${}^{\text{P}}y{}_{\text{r}}=\text{sgn}\left(\omega \right)\left(R_{\text{ICR}}-\left(R_{\text{f}}+\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{cos}\phi \right)$$

wobei d ∈ [0, I_ROI(v)] die Bogenentfernung entlang der Mittellinie des interessierenden Bereichs 60 ist, gemessen vom Vorderrad des virtuellen Fahrrads, und wobei der Winkel φ zwischen einer Linie vom Rotationszentrum ICR durch einen Punkt an der linken oder rechten Außenbegrenzung 56, 58 und einer Linie vom Rotationszentrum durch die Mitte des Hinterrads des virtuellen Fahrrads ist. D. h., der Winkel φ kann angeben werden durch $$\phi =\frac{d}{R_{\text{f}}}+{\phi }_{\text{f}}$$

The boundaries of the region of interest 60 can be described by mathematically describing its outer boundaries 56, 58. This can be used, for example, to visualize the region of interest 60. The first outer boundary 56, which in the figures is the right outer boundary of the region of interest 60, can be described by the parameterized coordinates ( ^P x _r , ^P y _r ) and the second outer boundary 58, which in the figures is the left outer boundary of the region of interest 60, can be described by the parameterized coordinates ( ^P x _l , ^P y _l ), each in the coordinate system P of the virtual bicycle. Thus, the outer boundaries 56, 58 can be specified by $${}^{\text{P}}x{}_{\text{l}}=\left(R_{\text{e}}-\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{sin}\phi$$

$${}^{\text{P}}y{}_{\text{l}}=\text{sgn}\left(\omega \right)\left(R_{\text{ICR}}-\left(R_{\text{e}}-\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{cos}\phi \right)$$

$${}^{\text{P}}x{}_{\text{r}}=\left(R_{\text{e}}+\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{sin}\phi$$

$${}^{\text{P}}y{}_{\text{r}}=\text{sgn}\left(\omega \right)\left(R_{\text{ICR}}-\left(R_{\text{e}}+\text{sgn}\left(\omega \right)\frac{1}{2}{\omega }_{\text{ROI}}\left(d\right)\right)\text{cos}\phi \right)$$

where d ∈ [0, I _ROI (v)] is the arc distance along the centerline of the region of interest 60, measured from the front wheel of the virtual bicycle, and where the angle φ between a line from the center of rotation ICR through a point on the left or right outer boundary 56, 58 and a line from the center of rotation through the center of the hin terrain of the virtual bicycle. That is, the angle φ can be given by $$\phi =\frac{d}{R_{\text{e}}}+{\phi }_{\text{e}}$$

11 shows a diagram of a flow chart of an embodiment of a method for speed-based determination of the region of interest. The region of interest 60 is located on the roadway 32 in the direction of travel in front of the motor vehicle 20 and is expected to be driven over by the motor vehicle 20 in the near future.

In a step S71, the speed data representative of a current speed v _ego of the motor vehicle 20 can be received, for example corresponding to the step S24.

In a step S72, yaw data can be received that is representative of the yaw rate ω _ego of the motor vehicle 20. The yaw rate ω _ego of the motor vehicle 20 can be generated, for example, by means of a yaw rate sensor, a magnetic sensor, for example a compass, GPS, and/or by means of an inertial measurement unit (IMU), transmitted to the control unit 25 and received by the control unit 25.

In a step S72, the side angle δ _c can be determined depending on the speed data, for example using the formulas explained above, where the side angle δ _c is the angle between a frame of a given virtual bicycle and the longitudinal axis V of the motor vehicle 20. The virtual bicycle can be specified such that the rear axle of the virtual bicycle runs through the geometric center Z of the motor vehicle 20 and that the length l _F of the frame of the virtual bicycle corresponds to a distance from the center Z to the front axle of the motor vehicle 20.

The current speed v _ego of the motor vehicle 20 can be encoded in the speed data in the form of a lateral speed v _lat and a longitudinal speed v _long of the center Z. The side angle δ _c can then be determined depending on the lateral speed v _lat and the longitudinal speed v _long of the center Z. For example, the side angle δ _c can be determined by means of an inverse function of the angular function tangent depending on the lateral speed v _lat and the longitudinal speed v _long , as explained above.

In a step S74, the region of interest 60 can be determined depending on the azimuth angle δ _c and the yaw rate ω or ω _ego , for example using the formulas explained above.

$$\left|R_{\text{t}}-R_{\text{f}}\right|\le \frac{1}{2}{\omega }_{\text{ROI}}\left(d_{\text{t}}\right)$$

wobei I_ROI(v) die, optional geschwindigkeitsabhängige, Länge, insbesondere die Bogenlänge, des interessierenden Bereichs 60 ist und ω_ROI(d_t) die, optional entfernungsabhängige Bogenbreite an der Position des Objekts 50 ist. Die Länge l_ROI(v) kann beispielsweise bestimmt werden auf der Grundlage des Anhaltewegs des Kraftfahrzeugs 20 bei maximaler Verzögerung. Bei geringem Abstand vom Kraftfahrzeug 20 zum Objekt 50 kann ω_ROI(d_t) beispielsweise durch eine Breite der entsprechenden Fahrspur 34, 36 oder durch die Breite B des Kraftfahrzeugs 20 bestimmt werden. Bei größeren Abständen vom Kraftfahrzeug 20 zum Objekt 50 kann die Ermittlung des interessierenden Bereichs 60 zunehmend ungenauer werden, was durch einen breiteren interessierenden Bereich 60 kompensiert werden kann. Der interessierende Bereich 60 kann näherungsweise als links/rechts-symmetrisch in Bezug auf einen voraussichtlichen Weg des Kraftfahrzeugs 20 angenommen werden.It is then easy to check whether one of the objects 50 is located in the speed-based region of interest 60. For example, the object 50 can be classified as being located in the region of interest 60 if the following conditions are met: $$d_{\text{t}}\le {\mathcal{l}}_{\text{ROI}}\left(\upsilon \right)$$

$$\left|R_{\text{t}}-R_{\text{e}}\right|\le \frac{1}{2}{\omega }_{\text{ROI}}\left(d_{\text{t}}\right)$$

where I _ROI (v) is the, optionally speed-dependent, length, in particular the arc length, of the region of interest 60 and ω _ROI (d _t ) is the, optionally distance-dependent arc width at the position of the object 50. The length l _ROI (v) can, for example, be determined on the basis of the stopping distance of the motor vehicle 20 at maximum deceleration. If the distance from the motor vehicle 20 to the object 50 is small, ω _ROI (d _t ) can, for example, be determined by a width of the corresponding lane 34, 36 or by the width B of the motor vehicle 20. At greater distances from the motor vehicle 20 to the object 50, the determination of the region of interest 60 can become increasingly less accurate, which can be compensated for by a wider region of interest 60. The region of interest 60 can be approximately assumed to be left/right symmetrical with respect to an expected path of the motor vehicle 20.

If the motor vehicle 20 is traveling in reverse, the formulas explained above can be used unchanged to determine the region of interest 60 and/or to check whether one of the objects 50 is located in the region of interest 60, wherein only the sign of the measured yaw rate ω _m has to be changed.

Track detector

12 shows an example of two of the previously explained driving lines 90, 92 for the motor vehicle 92. The first driving line 90 is marked by the first reference points 94 and the second driving line 92 is marked by the second reference points 96. The first driving line 90 is representative of the first lane 34. The second driving line 92 is representative of the second lane 36. The motor vehicle 20 can move according to a velocity vector 98 in the direction of travel, for example along the first travel line 90.

13 shows a flow chart of an embodiment of a method for determining the probability with which the motor vehicle 20 is located on one of the lanes 34, 36, for example on the first lane 34.

In a step S80, the position data that are representative of the current vehicle position of the motor vehicle 20 can be received, for example corresponding to the steps S22, S46 or S60 explained above. As explained above, the position data can be in the form of X and Y coordinates. If the position data are representative of the vehicle orientation of the motor vehicle 20, the position data can include the current vehicle orientation, in particular in the form of the yaw angle.

In a step S82, a reference point position of one of the reference points 94, 96 along the lane 34, 36 can be determined. The reference point position can be determined depending on the position data using a first lane lookup table that is assigned to the corresponding lane 34, 36 and in which the reference points 94, 96 are assigned the corresponding reference point positions. The reference points 94, 96 can be representative of a course of the corresponding lane 34, 36 in the direction of travel in front of the motor vehicle 20. The reference point position and the corresponding reference point 94, 96 can be determined such that the reference point position is closest to the current vehicle position of the motor vehicle 20, compared to the reference point positions of the other reference points 94, 96 in the first lane lookup table.

The first reference points 94 can be stored in the first lane lookup table together with their reference point positions, for example in the form of X and Y coordinates, with the lane orientations at the corresponding first reference points 94 and optionally with the lengths l _L of the first lane 34 at the corresponding first reference points 94. The second reference points 96 can be stored in a second lane lookup table together with their reference point positions, for example in the form of X and Y coordinates, with the lane orientations at the corresponding second reference points 96 and optionally with the lengths l _L of the second lane 36 at the corresponding second reference points 96.

If the lane orientation of the corresponding lane 34, 36 at the corresponding reference points 94, 96 is assigned to the reference points 94, 96 in the lane lookup tables, then in step S82, in addition to the reference point position, the lane orientation at the determined reference point position that is assigned to the corresponding reference point 94, 96 and that is closest to the current motor vehicle position can be determined using the first lane lookup table.

The first and possibly further lane lookup tables can be created in advance depending on the real lanes 34, 36, for example by measuring the corresponding lanes 34, 36 and extracting the relevant measurement results and storing them in the corresponding lane lookup tables. The lane lookup tables can also contain information about one or more curves, for example corresponding curvature data, wherein the curvature data can be stored in the form of the lane orientations in the corresponding lane lookup table.

In a step S84, a distance between the determined reference point position and the current vehicle position can be determined. For example, the reference point position and the current vehicle position can be in Cartesian coordinates of the same coordinate system and the distance can be determined using simple, known mathematical methods. For example, the distance can be determined and/or specified as a Euclidean distance.

If the position data are representative of the vehicle orientation of the motor vehicle 20 and in step S82 the lane orientation at the determined reference point position is determined using the first lane lookup table in addition to the reference point position, then in step S84 a deviation of the current vehicle orientation from the determined lane orientation can be determined in addition to the distance.

wobei xv(t) und x_L(t) die im Vorhergehenden eingeführten Vektoren zum Zeitpunkt t sind, die die Fahrzeugposition und -orientierung bzw. die Referenzpunktposition und Fahrspurorientierung angeben.If one assumes that the output values of the sensors used and/or the corresponding control unit are normally distributed, the distance and the deviation can be specified according to a Gaussian distribution as $${\overrightarrow{x}}_V\left(t\right)-{\overrightarrow{x}}_L\left(t\right)\sim N\left(\cdot |0,\mathrm{\Sigma }\right)$$

where xv(t) and x _L (t) are the vectors introduced above at time t, which represent the driving vehicle position and orientation or the reference point position and lane orientation.

In einem Schritt S86 kann die Wahrscheinlichkeit, mit der sich das Kraftfahrzeug 20 auf der entsprechenden Fahrspur 34, 36 befindet, abhängig von dem Abstand ermittelt werden. Beispielsweise kann die Wahrscheinlichkeit anhand einer vorgegebenen Abstandnachschlagetabelle, in der verschiedenen Abständen entsprechende Wahrscheinlichkeiten zugeordnet sind, ermittelt werden. Die Abstandnachschlagetabelle kann im Vorfeld empirisch oder mittels Simulation ermittelt werden und beispielsweise auf der Speichereinheit 26 gespeichert werden. Die Wahrscheinlichkeit, mit der sich das Kraftfahrzeug 20 auf der entsprechenden Fahrspur 34, 36 befindet, kann beispielsweise verwendet werden, um den interessierenden Bereich 60 positionsbasiert zu ermitteln, wie im Vorhergehenden erläutert.It follows that the squared norm of the state deviations with respect to a precision matrix Σ ^-1 is distributed according to the chi-square distribution with k = 3 degrees of freedom, which is why $${\delta }_{\mathrm{\Sigma }}^2\left({\overrightarrow{x}}_V,{\overrightarrow{x}}_L\right)\doteq {\left({\overrightarrow{x}}_V,{\overrightarrow{x}}_L\right)}^{\text{T}}{\mathrm{\Sigma }}^{-1}\left({\overrightarrow{x}}_V,{\overrightarrow{x}}_L\right)\sim {\chi }^2\left(\cdot ;k\right)$$

In a step S86, the probability with which the motor vehicle 20 is located in the corresponding lane 34, 36 can be determined depending on the distance. For example, the probability can be determined using a predetermined distance lookup table in which corresponding probabilities are assigned to different distances. The distance lookup table can be determined in advance empirically or by means of simulation and stored, for example, on the storage unit 26. The probability with which the motor vehicle 20 is located in the corresponding lane 34, 36 can, for example, be used to determine the region of interest 60 based on position, as explained above.

If the deviation of the vehicle orientation from the lane orientation is determined in step S84, the probability can be determined using the predetermined distance lookup table, wherein in this context different distances and/or different deviations are assigned corresponding probabilities in the distance lookup table.

wobei γ_Σ ein Konfidenzwert ist, der als Wahrscheinlichkeit, mit der sich das Kraftfahrzeug 20 auf der Fahrspur 34, 36 befindet, interpretiert und verwendet werden kann.Alternatively, the distance and the deviation can be considered together in the form of $${\gamma }_{\mathrm{\Sigma }}\left({\overrightarrow{x}}_V,{\overrightarrow{x}}_L\right)\doteq 1-{\displaystyle {\int }_0^{{\delta }_{\mathrm{\Sigma }}^2\left({\overrightarrow{x}}_V,{\overrightarrow{x}}_L\right)}{\chi }^2\left(x;k\right)}\text{d}x$$

where γ _Σ is a confidence value that can be interpreted and used as the probability that the motor vehicle 20 is in the lane 34, 36.

Optionally, in step S86, the determined probability with which the motor vehicle 20 is located in the lane 34, 36 can be compared with a predetermined lane threshold value. If the determined probability is less than the predetermined lane threshold value, the motor vehicle 20 can be classified as not being in the corresponding lane 34, 36. If the probability is equal to or greater than the predetermined lane threshold value, the motor vehicle 20 can be classified as being in the corresponding lane 34, 36. The predetermined lane threshold value can be specified, for example, to be greater than 80%, for example greater than 85%, for example greater than 90%, and can be a maximum of 100%.

Optionally, the motor vehicle 20 can be configured to travel on the corresponding lane 34, 36 either in a first direction of travel or in a second direction of travel opposite to the first direction of travel, wherein the first lane lookup table can be representative of the course of the first lane 34 in the first direction of travel and a second lane lookup table can be representative of the course of the first lane 34 in the second direction of travel. In this context, the current direction of travel of the motor vehicle 20 can be determined, for example, based on the position data, in particular the current vehicle orientation, in particular the current yaw angle. Based on the direction of travel, the first or the second lane lookup table can then be selected to determine the reference point position that is closest to the current vehicle position of the motor vehicle 20. In particular, the reference point position closest to the current vehicle position of the motor vehicle 20 can be determined from the first lane lookup table when the current direction of travel corresponds to the first direction of travel, and the reference point position can be determined from the second lane lookup table when the current direction of travel corresponds to the second direction of travel, wherein in the second lane lookup table the reference point positions are assigned to the first reference points 94 which are representative of the course of the first lane 34 in the second direction of travel.

If the roadway 32 has the first lane 34 and at least the second lane 36 directly next to the first lane 34, the method described above for determining the probability that the motor vehicle 20 is in the lane 34, 36 of the roadway 32 can be used to check whether the motor vehicle 20 is in the second lane 36. For example, the reference point position that is closest to the current vehicle position of the motor vehicle 20 can be determined using a third lane lookup table, wherein in the third lane lookup table the reference point positions are assigned to the second reference points 96 that are representative of a course of the second lane 36 in the second direction of travel. Furthermore, a fourth lane lookup table can be stored in which the reference point positions are assigned to the second reference points 96 that are representative of the course of the second Lane 36 in the first direction of travel is representative.

Thus, the reference point position closest to the current vehicle position of the motor vehicle 20 may be searched not only in the first lane lookup table, but also in the second, third and/or fourth lane lookup table. For example, the reference point position closest to the current vehicle position of the motor vehicle 20 may be searched in the fourth lane lookup table if the motor vehicle 20 is traveling in the first direction of travel and the distance of the current vehicle position to the closest reference point position from the first lane lookup table is greater than the predetermined lane threshold. If the distance of the current vehicle position to the closest reference point position from the fourth lane lookup table is then less than the predetermined lane threshold, the motor vehicle 20 is classified as being in the second lane 36. However, if the distance of the current vehicle position to the closest reference point position from the first lane lookup table is already smaller than the predetermined lane threshold value, consultation of the fourth lane lookup table can be dispensed with and the motor vehicle 20 can be classified as being in the first lane 34.

Alternatively, the first reference point 94 closest to the current vehicle position from the first lane lookup table and the second reference point 96 closest to the current vehicle position from the fourth lane lookup table may be searched for and the corresponding distances to the current vehicle position may be determined. If the distance between the current vehicle position and the determined first reference point 94 is less than the distance between the current vehicle position and the determined second reference point 96, the motor vehicle 20 may be classified as being in the first lane 34. If the distance between the current vehicle position and the determined first reference point 94 is greater than the distance between the current vehicle position and the determined second reference point 96, the motor vehicle 20 may be classified as being in the second lane 36. However, if both distances are greater than the predetermined lane threshold, the motor vehicle 20 may be classified as being in neither the first nor the second lane 34, 36.

If the check to determine which lane 34, 36 the motor vehicle 20 is currently in is determined using the confidence value γ _Σ explained above and thus both the distance between the position of the motor vehicle 20 and the closest reference point 94, 96 and the deviation between the current vehicle orientation and the lane orientation in the closest reference point 94, 96 are taken into account in the check, then, for example, the confidence value γ _Σ can be determined for all lanes 34, 36 and directions of travel, in particular using the first to fourth lane lookup tables. The motor vehicle 20 can then be classified as being in the lane 34, 36 for which the confidence value γ _Σ is highest. Optionally, the motor vehicle 20 may be classified as not being in any of the lanes 34, 36 if the highest confidence value γ _Σ is less than the predetermined lane threshold.

Daraus ergibt sich als Kovarianz im Beobachtungsraum $${\displaystyle \sum =\left(\begin{array}{ccc}\text{cos}{\psi }_V& \text{sin}{\psi }_V& 0\\ -\text{sin}{\psi }_V& \text{cos}{\psi }_V& 0\\ 0& 0& 1\end{array}\right){\displaystyle {\sum }_V\left(\begin{array}{ccc}\text{cos}{\psi }_V& -\text{sin}{\psi }_V& 0\\ \text{sin}{\psi }_V& \text{cos}{\psi }_V& 0\\ 0& 0& 1\end{array}\right)}}$$

Falls erkannt werden soll, dass das Kraftfahrzeug 20 nur dann als sich auf einer der Fahrbahnen 34, 36 befinden klassifiziert werden soll, wenn die Abstände und/oder Abweichungen innerhalb festgelegter Grenzen liegen, so kann von einem z-Score z°=°1 ausgegangen werden. Damit ergibt sich der Fahrspurschwellenwert beispielsweise zu $${\mathrm{\gamma }}_{thr}=1-{\displaystyle {\int }_0^{z^2}{\mathrm{\chi }}^2\left(x;3\right)\text{d}x=1-}{\displaystyle {\int }_0^1{\mathrm{\chi }}^2\left(x;3\right)\text{d}x=0.8012519}$$

The lane threshold value can be determined, for example, by first determining how far the motor vehicle 20 can deviate from the lane 34, 36 in terms of its position and orientation in order to just be classified as being in the corresponding lane 34, 36. For example, corresponding threshold values can be determined empirically and/or by simulation in advance and stored in the control unit 25. For example, a maximum distance of the X coordinate and the Y coordinate of the position of the motor vehicle 20 from the X coordinate or the Y coordinate of the reference point position of the corresponding reference point 94, 96 can be specified as 0.2 m each, and a maximum deviation of the current vehicle orientation from the lane orientation at the corresponding reference point 94, 96 can be specified as 0.108 rad. This results in the confidence matrix $${\mathrm{\Sigma }}_V=\left(\begin{array}{ccc}{0.2}^2& 0& 0\\ 0& {0.2}^2& 0\\ 0& 0& {0.108}^2\end{array}\right)$$

This results in the covariance in the observation space $${\displaystyle \sum =\left(\begin{array}{ccc}\text{cos}{\psi }_V& \text{<figure-callout id="0" label="sin ψ V" filenames="DE102022212897A1\_0065.png,DE102022212897A1\_0068.png" state="{{state}}">sin</figure-callout>}{\psi }_V{}_{}& 0\\ -\text{sin}{\psi }_V& \text{cos}{\psi }_{\mathrm{<figure-callout\; id="0"\; label="V"\; filenames="DE102022212897A1\_0065.png,DE102022212897A1\_0068.png"\; state="\{\{state\}\}">V</figure-callout>}}& 0\\ 0& 0& 1\end{array}\right){\displaystyle {\sum }_V\left(\begin{array}{ccc}\text{cos}{\psi }_V& -\text{<figure-callout id="0" label="sin ψ V" filenames="DE102022212897A1\_0065.png,DE102022212897A1\_0068.png" state="{{state}}">sin</figure-callout>}{\psi }_V{}_{}& 0\\ \text{sin}{\psi }_V& \text{cos}{\psi }_{\mathrm{<figure-callout\; id="0"\; label="V"\; filenames="DE102022212897A1\_0065.png,DE102022212897A1\_0068.png"\; state="\{\{state\}\}">V</figure-callout>}}& 0\\ 0& 0& 1\end{array}\right)}}$$

If it is to be recognized that the motor vehicle 20 should only be classified as being on one of the lanes 34, 36 if the distances and/or deviations are within specified If the lane threshold is within the specified limits, a z-score of z°=°1 can be assumed. This results in the lane threshold value, for example, $${\mathrm{\gamma }}_{tHr}=1-{\displaystyle {\int }_0^{{\mathrm{<figure-callout\; id="2"\; label="z"\; filenames="DE102022212897A1\_0068.png,DE102022212897A1\_0069.png"\; state="\{\{state\}\}">z</figure-callout>}}^2}{\mathrm{\chi }}^2\left(x;3\right)\text{d}x=1-}{\displaystyle {\int }_0^1{\mathrm{\chi }}^2\left(x;3\right)\text{d}x=0.8012519}$$

gilt, kann dann das Kraftfahrzeug als sich auf der entsprechenden Fahrspur 34, 36 befindend klassifiziert werden.If $${\mathrm{\gamma }}_{\Sigma }\left({\overrightarrow{x}}_V,{\overrightarrow{x}}_L\right)>{\mathrm{\gamma }}_{tHr}$$

applies, the motor vehicle can then be classified as being in the corresponding lane 34, 36.

TTC and length ROI

wobei i repräsentativ für eine Fahrsituation ist und beispielsweise die Werte A für eine erste Fahrsituation, B für eine zweite Fahrsituation und C für eine dritte Fahrsituation annehmen kann und wobei $$\mathrm{\Delta }{t}_c$$

die Zeitdauer ist, die vergeht, bis es zu einer Kollision mit einem der Objekte 50 kommt. Das Diagramm gemäß 14 und insbesondere die Verläufe der geschwindigkeitsabhängigen Zeitschwellenwerte können beispielsweise mittels den einzelnen Fahrsituation entsprechenden Simulationen oder empirisch im Vorfeld ermittelt werden, beispielsweise von einem Hersteller des Steuergeräts 25 und/oder des Kraftfahrzeugs 20. 14 shows a diagram in which several exemplary curves of speed-dependent time threshold values are shown as a function of the speed of the motor vehicle 20, wherein the time threshold values are indicated in the form of $${\overline{\mathrm{\Delta }t}}_{c,i}$$

where i is representative of a driving situation and can, for example, take the values A for a first driving situation, B for a second driving situation and C for a third driving situation and where $$\mathrm{\Delta }{t}_c$$

is the time that passes until a collision with one of the objects 50 occurs. The diagram according to 14 and in particular the courses of the speed-dependent time threshold values can be determined in advance, for example by means of simulations corresponding to the individual driving situation or empirically, for example by a manufacturer of the control unit 25 and/or the motor vehicle 20.

Thus, for one or more speeds v _ego of the motor vehicle 20, one or more first time threshold values for the first driving situation A, one or more second time threshold values for the second driving situation B, and/or one or more third time threshold values for the third driving situation C can be determined. The first driving situation A can, for example, relate to the fact that the motor vehicle 20 is approaching the object 50 and the object 50 is stationary. The second driving situation B can, for example, relate to the fact that the object 50 is a motor vehicle 20 coming towards the motor vehicle 20, whereby in the present case, when simulating the second driving situation B, only speeds v _ego of the motor vehicle 20 up to 5.56 m/s were taken into account, which is why the corresponding curve in the diagram according to 14 ends at this value. The third driving situation C can, for example, refer to the fact that the object 50 is a vulnerable road user, also referred to as a "Vulnerable Road User (VRU)", who is coming towards the motor vehicle 20. A fourth driving situation can, for example, refer to the fact that the object 50 is a pedestrian who is crossing the roadway 32, wherein this fourth driving situation is assumed in this description to be the same as the first driving situation A, since the corresponding simulations resulted in the same values for both driving situations.

Based on the 14 A time threshold lookup table can be generated from the diagram shown, in particular from the corresponding data. In the time threshold lookup table, the various speeds v _ego can be assigned the respective first time threshold if this first time threshold is greater than the corresponding second and third time thresholds for the corresponding speed v _ego ; the various speeds v _ego can be assigned the respective second time threshold in the time threshold lookup table if this second time threshold is greater than the corresponding first and third time thresholds; and the various speeds v _ego can be assigned the respective third time threshold in the time threshold lookup table if this third time threshold is greater than the corresponding first and second time thresholds. The various speeds v _ego can thus be assigned the largest of the corresponding time thresholds.

15 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines speziellen Verfahrens zum Vermeiden oder Vermindern der Kollision des autonomen Kraftfahrzeugs 20 mit dem Objekt 50, das auf der Fahrbahn 32 fährt.The largest of the time threshold values at the given speed v _ego can thus be selected as the total time threshold value at a given speed v _ego of the motor vehicle 20, for example according to $${\overline{\mathrm{\Delta }t}}_{\text{c}}\left({\upsilon }_{\text{ego}}\right)=\underset{i\in S}{\text{Max}}{\overline{\mathrm{\Delta }t}}_{\text{c}\text{,}i}\left({\upsilon }_{\text{ego}}\right)$$

15 shows a flow chart of an embodiment of a special method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50 traveling on the roadway 32.

In a step S80, the sensor data can be received which are representative of the fact that the object 50 is located in front of the motor vehicle 20 in the direction of travel. The step S80 can be processed, for example, corresponding to the steps S2, S20 or S67.

In a step S82, the speed data representative of the current speed v _ego of the motor vehicle 20 can be received. The current speed v _ego can be determined, for example, from the radar data or retrieved from the storage unit 26.

wobei ein Abstand d(t) von dem Kraftfahrzeug zu dem Objekt 50 angegeben werden kann mit $$d\left(t\right)=q_{\text{tgt}}\left(t\right)-q_{\text{ego}}\left(t\right)$$

wobei q_tgt(t) die Position des Objekts 50 zum Zeitpunkt t und q_ego(t) die Position des Kraftfahrzeugs 20 zum Zeitpunkt t beschreibt. In diesem Zusammenhang ist anzumerken, dass q_ego(t) dem vorstehend erläuterten Positionsvektor $\overrightarrow{x_V}\left(t\right)$

ohne die Fahrzeugorientierung, insbesondere ohne den Gierwinkel des Kraftfahrzeugs 20, entspricht. Die Position q_ego(t) des Objekts 50 kann anhand der Sensordaten und der Positionsdaten ermittelt werden. Die Position q_ego(t) des Objekts 50 kann einem voraussichtlichen Aufprallpunkt an dem Objekt 50 entsprechen. Die Position q_ego(t) des Kraftfahrzeugs 20 kann sich beispielsweise auf eine Position der Mitte des vorderen Stoßfängers des Kraftfahrzeugs 20 beziehen. Der Parameter d_s kann sich auf einen vorgegebenen Sicherheitsabstand beziehen, der zu dem Objekt 50 eingehalten werden soll, wenn das Kraftfahrzeug 20 nach dem Bremsen zum Stillstand gekommen ist. Mit d_s wird ein virtueller Stoßfänger für das Kraftfahrzeug 20 eingeführt, von dem der Zeitschwellenwert nicht beeinflusst wird, weswegen nachfolgend d_s = 0 gesetzt wird.In a step S84, the time period TTC can be determined which elapses until the motor vehicle 20 collides with the object 50, depending on the sensor data and the speed data. The time period TTC can be defined as Δt _c (t) in the form $$\mathrm{\Delta }{t}_{\text{c}}\left(t\right):=\{\begin{array}{ll}\mathrm{0,}\hfill & d\left(t\right)<d_{\text{s}}\hfill \\ \frac{d\left(t\right)-d_{\text{s}}}{\left|\dot{d}\left(t\right)\right|},\hfill & d\left(t\right)\ge d_{\text{s}},\dot{d}\left(t\right)<0\hfill \\ \infty ,\hfill & \text{otherwise}\hfill \end{array}$$

where a distance d(t) from the motor vehicle to the object 50 can be specified with $$d\left(t\right)=q_{\text{tgt}}\left(t\right)-q_{\text{ego}}\left(t\right)$$

where q _tgt (t) describes the position of the object 50 at time t and q _ego (t) describes the position of the motor vehicle 20 at time t. In this context, it should be noted that q _ego (t) corresponds to the position vector explained above $\overrightarrow{x_V}\left(t\right)$

without the vehicle orientation, in particular without the yaw angle of the motor vehicle 20. The position q _ego (t) of the object 50 can be determined based on the sensor data and the position data. The position q _ego (t) of the object 50 can correspond to a probable point of impact on the object 50. The position q _ego (t) of the motor vehicle 20 can, for example, refer to a position of the center of the front bumper of the motor vehicle 20. The parameter d _s can refer to a predetermined safety distance that should be maintained from the object 50 when the motor vehicle 20 has come to a standstill after braking. With d _s, a virtual bumper for the motor vehicle 20 is introduced, which does not influence the time threshold, which is why d _s = 0 is set below.

Consequently, the time period TTC or Δt _c (t) indicates the time interval from a current point in time t to a moment t+Δt _c (t) at which the safety distance d _s is violated, assuming that both the motor vehicle 20 and the object 50 maintain a constant speed and that they are on a collision course.

In step S86, the speed-dependent time threshold value Δt _c,i depending on the speed data, in particular depending on the speed v _ego of the motor vehicle 20. The speed-dependent time threshold value Δt For example _{, c,i} can be determined using the time threshold lookup table.

In a step S87, it can be checked whether the determined time period TTC is less than the determined time threshold value. If the condition of step S87 is met, processing can be continued again in step S80. If the condition of step S87 is not met, processing can be continued in a step S88.

In a step S88, the control signal for the braking device 24 of the motor vehicle 20 can be generated.

If the sensor data are representative of the fact that two or more of the objects 50 are located in the direction of travel in front of the motor vehicle 20, the time period TTC that elapses until the motor vehicle 20 collides with the corresponding object 50 can be determined for each of the objects 50, depending on the sensor data and the speed data. The control signal for the braking device 24 can then be generated if one of the determined time periods TTC is less than the determined time threshold value.

16 shows a diagram in which several exemplary lengths I _ROI of the region of interest 60 are shown as a function of the speed v _ego of the motor vehicle 20. The diagram according to 16 and in particular the courses of the lengths I _ROI can be determined, for example, by means of simulations corresponding to the individual driving situation or empirically in advance, for example by the manufacturer of the control unit 25 and/or the motor vehicle 20.

wobei d _l einem Auslöseabstand in der entsprechenden Fahrsituation entspricht, i wieder die Fahrsituation A, B, C angibt, und der Zusatzterm „0,8v_ego“ vorgesehen ist, der verhindert, dass das Bremsen des Kraftfahrzeugs 20 aufgrund eines der Objekte 50 genau in dem Moment ausgelöst wird, in dem das entsprechende Objekt 50 in den interessierenden Bereich 60 eintritt, was eine kleine Verzögerung bei der Auslösung des Bremsvorgangs verursachen könnte. Dieser Zusatzterm, insbesondere sein Betrag, kann beispielsweise empirisch oder mittels Simulation ermittelt und dann vorgeben werden. Somit ergeben sich zu den verschiedenen Geschwindigkeiten v_ego je drei verschiedene Längen I_ROI des interessierenden Bereichs, insbesondere erste Längen I_ROI,1 für die erste Fahrsituation A, zweite Längen I_ROI,2 für die zweite Fahrsituation B und dritte Längen I_ROI,3 für die Fahrsituation C.The lengths I _ROI of the region of interest 60 for one of the driving situations i can be determined, for example, according to $$\begin{array}{cc}{\mathcal{l}}_{\text{ROI}\text{,}i}\left({\upsilon }_{\text{ego}}\right)={\overline{d}}_i\left({\upsilon }_{\text{ego}}\right)+0.8{\upsilon }_{\text{ego}},& i\in S\end{array}$$

where d _l corresponds to a triggering distance in the corresponding driving situation, i again indicates the driving situation A, B, C, and the additional term "0.8v _ego " is provided, which prevents the braking of the motor vehicle 20 due to one of the objects 50 from being triggered at the exact moment in which the corresponding object 50 enters the area of interest 60, which could cause a small delay in the triggering of the braking process. This additional term, in particular its amount, can be determined empirically or by means of Simulation and then specified. This results in three different lengths I _ROI of the area of interest for the different speeds v _ego , in particular first lengths I _ROI,1 for the first driving situation A, second lengths I _ROI,2 for the second driving situation B and third lengths I _ROI,3 for the driving situation C.

Based on the 16 A length lookup table can be generated from the diagram shown, in particular on the basis of the corresponding data. In the length lookup table, the various speeds v _ego can be assigned the respective first length l _ROI,1 if this first length l _ROI,1 is greater than the corresponding second and third lengths l _ROI,2 , l _ROI,3 at the corresponding speed v _ego ; the various speeds v _ego can be assigned the respective second length l _ROI,2 if this second length l _ROI,2 is greater than the corresponding first and third lengths l _ROI,1 , l _ROI,3 at the corresponding speed v _ego ; and the various speeds v _ego can be assigned the respective third length l _ROI,3 if this third length l _ROI,3 is greater than the corresponding first and second lengths l _ROI,1 , l _ROI,2 at the corresponding speed v _ego . The various speeds v _ego length lookup table can thus be assigned the largest of the corresponding lengths length l _ROI, i in each case.

17 zeigt ein Ablaufdiagramm eines Ausführungsbeispiels eines speziellen Verfahrens zum Vermeiden oder Vermindern der Kollision des autonomen Kraftfahrzeugs 20 mit dem Objekt 50.For determining the region of interest 60, in particular for determining its length l _ROI , the greatest length l _ROI,i in the driving situations i at the speed v _ego of the motor vehicle 20 can thus be used, for example according to $${\mathcal{l}}_{\text{ROI}}\left({\upsilon }_{\text{ego}}\right)=\underset{i\in S}{\text{Max}}{\mathcal{l}}_{\text{ROI}\text{,}i}\left({\upsilon }_{\text{ego}}\right)$$

17 shows a flow chart of an embodiment of a special method for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50.

A step S90 can be processed corresponding to the step S80.

In a step S91, the position data representative of the current vehicle position of the motor vehicle 20 can be received, for example corresponding to the step S22.

In a step S91, the speed data can be received, for example corresponding to the step S24.

In a step S93, the time period TTC can be determined, for example corresponding to the step S84.

In a step S94, the time threshold value can be determined, for example corresponding to step S86.

In a step S95, the length I _ROI of the region of interest 60, which is located on the roadway 32 in the direction of travel in front of the motor vehicle 20, can be determined depending on the time period TTC and the speed data, for example using the length lookup table. Alternatively or additionally, the length I _ROI of the region of interest 60 can be determined by determining a trigger distance depending on the speed v _ego of the motor vehicle 20 and the determined time threshold value, wherein the trigger distance is the distance between the object 50 and the motor vehicle 20 that the object 50 has from the motor vehicle 20 when the time period TTC falls below the corresponding time threshold value. The length I _ROI can be selected so that it is greater than the trigger distance.

In a step S96, the region of interest 60 with the determined length I _ROI can be determined depending on the position data, in particular position-based, as explained above.

In a step S97, it can be checked whether the object 50 is located in the region of interest 60, depending on the sensor data, for example corresponding to the step S66 and/or corresponding to the step S66 with reference to 8th explained procedures.

In a step S99, the control signal for the braking device 24 can be generated, for example corresponding to one of the steps S12, S36 or S56.

The general method explained above for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50, the special methods explained above for avoiding or reducing the collision of the autonomous motor vehicle 20 with the object 50 and/or the other methods explained above can be combined with each other as desired. For example, the general method can serve as a basic framework and the special methods can optionally serve as building blocks of the basic framework, for example depending on the specific application. For example, in the case of the method described with reference to 2 The method explained above can be used to check whether the region of interest 60 is position-based or speed-based determination, for example as in the case of 3 Alternatively or additionally, the procedure described in 2 explained method, the sensor data are fused, as in the case of the one with reference to 5 Alternatively or additionally, the procedure described in 2 explained methods, the region of interest 60 can be determined position-based, for example as in the case with reference to 8th Alternatively or additionally, the procedure described in 2 It can be checked whether the object 50 is located in the region of interest 60 or not, for example as in the method described with reference to 9 Alternatively or additionally, the procedure described in 2 explained methods, the region of interest 60 can be determined based on speed, for example as in the case with reference to 11 Alternatively or additionally, the procedure described in 2 It can be checked whether the motor vehicle 20 is currently in the lane 34, 36, for example as in the case with reference to 13 Alternatively or additionally, the procedure described in 2 The time period TTC can be determined using the method explained, for example as in the case of the method referred to 15 Alternatively or additionally, the procedure described in 2 The length I _ROI of the region of interest 60 can be determined using the method explained above, for example as in the method described with reference to 17 explained procedures.

In addition, it should be noted that "having" does not exclude other elements or steps and "a", "an" or "an" does not exclude a plurality. It should also be noted that features or steps that have been described with reference to one of the above embodiments can also be used in combination with other features or steps of other embodiments described above. Reference signs in the claims are not to be regarded as a limitation.

Reference symbols

20

Motor vehicle

21

Passenger cabin

22

door

23

Wheels

24

Braking device

25

Control unit

26

Storage unit

27

processor

30

Display

32

roadway

34

first lane

35

Centerline

36

second lane

37

first page margin

38

Radar object

39

second page margin

40

Contour point

42

Fusion object

44

Velocity vector radar-only object

46

Velocity vector lidar-only object

48

Velocity vector fusion object

50

object

52

first outer point

54

second outer point

56

first outer boundary

58

second outer boundary

60

area of interest

62

front

64

Rear

90

first route

92

second route

94

first reference points

96

second reference points

98

Velocity vector motor vehicle

V

longitudinal axis of the motor vehicle

L

Index of lanes

vlong

Longitudinal velocity of the center

vego

Speed of the motor vehicle

ωego

Yaw rate of the vehicle

ωm

measured yaw rate

IROI

Length of the area of interest

lF

Length of the frame of the virtual bike

lL

Length of the corresponding lane to the corresponding reference point

TTC

Duration

Z

geometric center of the motor vehicle

engl

Target distance to object

RICR

Radius R _ICR around rotation center

j, i

Running variables, natural numbers

S2-S99

Steps two to ninety-nine

Claims (8)

Method for determining a probability with which an autonomous motor vehicle (20) is located in a lane (34, 36) of a roadway (32), the method comprising: Receiving position data that are representative of a current vehicle position of the motor vehicle (20); Determining at least one reference point position of a reference point (94, 96) along the lane (34, 36) using a first lane lookup table that is assigned to the lane (34, 36) and in which reference points (94, 96) are assigned corresponding reference point positions, depending on the position data, wherein the reference points (94, 96) are representative of a course of the lane (34, 36) in the direction of travel in front of the motor vehicle (20) and wherein the determined reference point position is closest to the current vehicle position of the motor vehicle (20); Determining a distance of the determined reference point position to the current vehicle position; and Determining the probability that the motor vehicle (20) is in the lane (34) depending on the distance. Procedure according to Claim 1 , wherein the probability with which the motor vehicle (20) is in the lane (34, 36) is compared with a predetermined lane threshold value; the motor vehicle (20) is classified as not being in the lane (34, 36) if the probability is less than the lane threshold value; and the motor vehicle (20) is classified as being in the lane (34, 36) if the probability is equal to or greater than the lane threshold value. Method according to one of the preceding claims, wherein the position data are representative of a current motor vehicle orientation of the motor vehicle (20); in the first lane lookup table, the reference points (94, 96) are each assigned a lane orientation of the lane (34, 36) at the corresponding reference points (94, 96); the lane orientation at the reference point position that is assigned to the corresponding reference point (94, 96) and that is closest to the current motor vehicle position is determined using the first lane lookup table; a deviation of the current motor vehicle orientation from the determined lane orientation is determined; and the probability with which the motor vehicle (20) is in the lane (34, 36) is determined depending on the deviation. Method according to one of the preceding claims, wherein the motor vehicle (20) is configured to drive on the lane (34, 36) optionally in a first direction of travel or in a second direction of travel opposite to the first direction of travel; the first lane lookup table is representative of the course of the lane (34, 36) in the first direction of travel; a current direction of travel of the motor vehicle (20) is determined; the reference point position that is closest to the current vehicle position of the motor vehicle (20) is determined using the first lane lookup table if the current direction of travel corresponds to the first direction of travel; and the reference point position that is closest to the current vehicle position of the motor vehicle (20) is determined using a second lane lookup table if the current direction of travel corresponds to the second direction of travel, wherein in the second lane lookup table reference point positions are assigned to reference points (94, 96) that are representative of a course of the lane (34, 36) in the second direction of travel. Method according to one of the preceding claims, wherein the roadway (32) has the lane (34) as the first lane (34) and at least one second lane (36) directly next to the first lane (34); and the reference point position that is closest to the current vehicle position of the motor vehicle (20) is determined using a third lane lookup table, wherein reference point positions in the third lane lookup table are assigned to reference points (94, 96) that are representative of a course of the second lane (36). Control device (25) for determining a probability with which an autonomous motor vehicle (20) is on a lane (34, 36) of a vehicle lane (32), the control unit (22) comprising: a memory unit (26) for storing position data and at least one lane lookup table which is assigned to the lane (34, 36) and in which reference points (94, 96) are assigned corresponding reference point positions, wherein the reference points (94, 96) are representative of a course of the lane (34, 36) in the direction of travel in front of the motor vehicle (20); and a processor (27) which is configured to execute the method according to one of the preceding claims. Computer program for determining a probability with which an autonomous motor vehicle (20) is located on a lane (34, 36) of a roadway (32), wherein the computer program has instructions which cause the method according to one of the Claims 1 until 5 is processed when it is sent from a control unit (22) to Claim 6 be executed. Computer-readable storage medium on which a computer program is Claim 7 is stored.

Images