DE102019214142A1 - Procedure for warning of mass movements on sloping terrain - Google Patents

Abstract

In a method (V) for warning of mass movements on sloping terrain (2) for a vehicle (1), the sloping terrain (2) is recorded by means of at least one camera (3) so that image data (B) are generated. Driving dynamics (F) of the vehicle (1) are determined. An optical flow (5) of at least one object (4; 4 *) located on the sloping terrain (2) is determined from the image data (B). It is evaluated whether the at least one object (4; 4 *) fulfills an epipolar condition (E). The at least one object (4; 4 *) is marked as a critical object in the image data if the at least one object (4; 4 *) does not meet the epipolar condition (E). A trajectory (I) of the vehicle (1) is determined from the driving dynamics (F). A trajectory (II) of the at least one critical object (4; 4 *) is determined on the basis of the image data (B). Based on the trajectory (I) of the vehicle (1) and the trajectory (II) of the at least one critical object (4; 4 *), a collision probability (K) between the vehicle (1) and the at least one critical object (4; 4 *) calculated. A warning (W) is issued when the collision probability (K) reaches or exceeds a threshold value (S).

Description

The present invention relates to a method for warning of mass movements on sloping terrain for a vehicle with the features according to claim 1, a control device with the features according to claim 6, a computer program product with the features according to claim 7, and a vehicle with the features according to claim 8th.

Vehicles that are z. B. move in the mountains, are exposed to the risk of rock failure or falling debris. A rock can come loose on sloping terrain, for example on a slope, and can fall onto the roadway along which the vehicle is moving. This represents a risk of collision for the vehicle if the danger is not recognized early.

From the publication by Pieri, Gabriele & Magrini, M & Moroni, Davide & Salvetti, Ovidio & Fantini, Andrea & Martino, Salvatore and Prestininzi, A. "Experimenting an embedded sensor network for early warning of natural risks due to fast failures along railways "from 2015 is a project for real-time monitoring of railway tracks to detect events that can endanger train traffic. These events can be caused, for example, by rock fractures or falling debris.

On the basis of the prior art, the present invention is based on the object of proposing a method by means of which rock fractures or other mass movements can be detected early in order to warn a driver of a vehicle.

Based on the aforementioned object, the present invention proposes a method for warning of mass movements on sloping terrain for a vehicle according to claim 1, a control device according to claim 6, a computer program product according to claim 7, and a vehicle according to claim 8. Further advantageous refinements and developments emerge from the subclaims.

In a method for warning of mass movements on sloping terrain for a vehicle, the sloping terrain is recorded by means of at least one camera, so that image data are generated. Driving dynamics of the vehicle are determined. An optical flow of at least one object located on the sloping terrain is determined from the image data. It is evaluated whether the at least one object fulfills an epipolar condition. The at least one object is marked as a critical object in the image data if the at least one object does not meet the epipolar condition. A trajectory of the vehicle is determined from the driving dynamics. A trajectory of the at least one critical object is determined on the basis of the image data. A collision probability between the vehicle and the at least one critical object is calculated on the basis of the trajectory of the vehicle and the trajectory of the at least one critical object. A warning is issued when the probability of a collision reaches or exceeds a threshold value.

The vehicle is a land vehicle and can be a car or a commercial vehicle (commercial vehicle), e.g. B. a truck, an agricultural or forestry vehicle, namely an agricultural machine, or a construction machine, an industrial truck, a snow groomer, a mining vehicle, a cleaning machine, or another commercial vehicle. The vehicle can be designed in such a way that it can carry out automated driving functions from level 2 of the SAE J3016 autonomy levels.

A sloping terrain is, for example, a slope, an embankment, a rock wall or the like. A mass movement is defined as any movement of one or more objects from a starting point to an end point, the movement not being caused by a drive device. A mass movement on sloping terrain is represented, for example, by rock fractures, rubble, falling rocks, avalanches, lava flows or the like. An object or several objects move in the direction of a ground plane due to the effects of gravity. This object or these objects topple or fall in the direction of the ground plane.

The sloping terrain is recorded by means of the at least one camera, so that image data are generated. The vehicle thus has the at least one camera. Of course, the vehicle can have more than one camera. The at least one camera can be designed as a mono camera or as a stereo camera. The at least one camera is arranged on the vehicle in such a way that it can capture an area in the direction of travel in front of the vehicle. In addition, the at least one camera is sufficiently directed over the horizon. The at least one camera thus continuously records the sloping terrain in an area in front of the vehicle. As a result, image data are continuously generated.

Driving dynamics of the vehicle are also determined. The driving dynamics are made up of longitudinal dynamics, transverse dynamics, lifting movement, yawing movement, pitching movement, rolling movement and vibrations together. This determination takes place by means of the vehicle's own sensors, which are known from the prior art. For example, the vehicle can have commercially available acceleration sensors, yaw rate sensors, vibration sensors, steering angle sensors, etc. as driving dynamics sensors. The driving dynamics are continuously determined.

An optical flow of at least one object located on the sloping terrain is determined from the image data. The at least one object is arranged on the sloping terrain at the beginning of the generation of the image data. The at least one object can be any object located at a point on the sloping terrain. The at least one object can, for. B. a rock, a stone, a tree, a branch o. Ä. Be. Of course, there can be several objects on the sloping terrain, for example a collection of objects, e.g. B. snow, mud, sand, rubble, lava, gravel or the like.

The determination of the optical flow is a method for tracking the movement of a point from a first recording to a second recording and then to further recordings. Here the at least one object is tracked based on the image data from the camera. In other words, the optical flow represents a group of motion vectors of the at least one object. If the vehicle has a stereo camera that is able to take two recordings of the same area at the same time, the optical flow of the at least one object can can be determined immediately. This also applies when using two or more mono cameras that are synchronized in time.

However, if the vehicle only has a mono camera, the optical flow is determined using recordings that were generated at two different points in time. For this it is necessary that, in addition to the image data, the driving dynamics are included in the determination of the optical flow. A translation and a rotation of the vehicle that existed between the generation of the recordings at the two different points in time are determined from the driving dynamics. If the translation and the rotation of the vehicle between two points in time and the installation position of the at least one camera are known, the translation and the rotation of the at least one camera can be calculated. In other words, the movement of the vehicle can be related to the image data. This allows the optical flow between the two recordings to be determined at these points in time. Due to the epipolar geometry, in particular the calculation and evaluation of the fundamental matrix, the movement of the vehicle and the camera can be related to the image data.

On the basis of the optical flow, it is evaluated whether the at least one object fulfills an epipolar condition. The evaluation is carried out by means of a control device which the vehicle has. The control device is connected to the at least one camera so that data and signals can be exchanged. The connection can be wired or wireless. The control device can have a memory device in which data can be stored. Furthermore, the control device can be connected to the sensors that determine the driving dynamics.

The epipolar condition is specified by means of the epipolar geometry. Using the epipolar geometry, the coordinates of an object can be calculated from two recordings if the orientation of the camera or cameras is known. The projection centers of the cameras or the camera, depending on the shape of the camera, and the recorded object span an epipolar plane. The epipolar plane intersects the two recordings in a straight line. Another straight line, which connects the two projection centers, penetrates an image plane of the respective recording at a point which is referred to as the epipole. The epipolar condition states that the optical flow of any static object is through the epipole. Measurement inaccuracies must of course be taken into account. That is to say that the epipolar condition is fulfilled for the at least one object when the optical flow of the at least one object runs through the epipole.

If, however, the optical flow of the at least one object does not run through the epipole, taking into account the measurement inaccuracies, the at least one object is not static but dynamic. This means that the at least one object performs a movement which is recorded by the at least one camera. This represents a mass movement. The at least one object is therefore subsequently marked as a critical object in the image data. This marking can either only be internal to the system or can be output to a driver of the vehicle by means of an output device. A system-internal marking means that if it is determined that it is a critical object, the object will continue to be observed.

The motion vectors of the optical flow that do not meet the epipolar condition can be grouped into groups. The at least one object can be represented by one or more groups of motion vectors.

If the marking of the critical object is output, this can be done, for example, by means of a colored highlighting, which is displayed by the output device. The output device can be designed as a display, for example. The vehicle thus has an output device. The output device is connected to the at least one camera. This connection is such that data and signals can be exchanged. The output device can be connected to the at least one camera via the control device, or can be connected directly to the at least one control device. For example, the image data that are generated by the at least one camera can, by means of the output device, for. B. be issued to a driver of the vehicle.

The trajectory of the vehicle is then determined from the driving dynamics. This z. B. the direction of movement, the speed and the acceleration of the vehicle are known. The trajectory of the at least one critical object is determined on the basis of the image data. This takes place, for example, taking into account the optical flow that has already been determined. This z. B. the direction of movement, the speed and the acceleration of the at least one critical object is known.

A collision probability between the vehicle and the at least one critical object is calculated on the basis of the trajectory of the vehicle and the trajectory of the at least one critical object. This calculation is carried out by means of the control device. The collision probability indicates whether and how high the probability is that a collision with the at least one critical object will occur if the vehicle continues to travel unchanged. Additional factors can be taken into account, such as a braking distance of the vehicle, safety distances to be observed, road surface conditions with regard to static friction, reaction times of the driver of the vehicle, reaction times or control limits of an actuator system of the vehicle, or the like.

If the likelihood of a collision reaches or exceeds a threshold value, a warning is issued, e.g. B. by means of the output device or another warning device. The warning can be acoustic, visual, haptic or a combination of these options.

The threshold value does not represent a maximum or minimum probability of collision. The threshold value is, for example, set at the factory and stored in the storage device of the control device. As an alternative to this, the threshold value can be determined automatically by means of the control device, depending on the current situation. For example, this threshold value can be selected to be lower in the case of unfavorable road conditions, for example slippery, wet, dirty road, which lead to a low coefficient of static friction than in favorable road conditions (e.g. dry road).

It is advantageous that the driver can be warned at an early stage as soon as a critical object and thus a mass movement on the sloping terrain is detected. This makes it possible to avoid a collision between the vehicle and the critical object. In addition, the driver is not forced to constantly observe the route he is driving along the sloping terrain in order to identify a mass movement himself. This increases the safety of the driver.

According to a further development, there is also an intervention in the longitudinal dynamics of the vehicle, so that a collision between the vehicle and the critical object is avoided. According to a further development, the intervention in the longitudinal dynamics is designed as an acceleration intervention and / or as a braking intervention and / or a steering intervention.

In other words, as soon as the collision probability has reached or exceeded the threshold value, a collision avoidance maneuver is carried out. For this purpose, the control device controls z. B. a braking system of the vehicle, so that a braking maneuver is carried out to avoid the collision. This represents the braking intervention. The control device is therefore connected to the braking system of the vehicle so that data and signals can be exchanged. The vehicle is braked by the braking intervention, which changes the trajectory and thus the longitudinal dynamics of the vehicle. The braking intervention can be designed such that the vehicle comes to a standstill after the braking intervention. Alternatively, the braking intervention can be designed in such a way that the vehicle reduces its speed but still moves. This reduces the likelihood of a collision so that it is below the threshold value.

As an alternative or in addition to this, the control device can control a drive system of the vehicle so that acceleration is carried out in order to avoid the collision. This represents the acceleration intervention. The control device is therefore connected to the drive system of the vehicle so that data and signals can be exchanged. The drive system of the vehicle provides drive energy to propel the vehicle. The vehicle is accelerated by the acceleration intervention, which reduces the trajectory and thus the longitudinal dynamics of the vehicle will be changed. This reduces the likelihood of a collision. Either only the acceleration intervention can be carried out, or this can be combined with the braking intervention. For example, an acceleration intervention can first take place in order to avoid the at least one critical object and then a braking intervention can take place in order to reduce the speed of the vehicle again. For example, a braking intervention can first take place and then an acceleration intervention if the braking intervention is not sufficient to lower the probability of a collision in such a way that it is below the threshold value.

As an alternative or in addition to this, the control device can control a steering system of the vehicle so that a steering maneuver is carried out in order to avoid the collision. This represents the steering intervention. The control device is therefore connected to the steering system of the vehicle so that data and signals can be exchanged. As a result of the steering intervention, a steering angle is set on the steerable wheels of the vehicle, so that the at least one critical object can be avoided. The at least one critical object is thus bypassed. In other words, the trajectory and thus the longitudinal dynamics of the vehicle are adapted. This reduces the likelihood of a collision so that it is below the threshold value. The steering intervention can be combined with the braking intervention or with the acceleration intervention. For example, the vehicle can first be braked in order to then carry out a steering intervention and be accelerated. For example, the vehicle can be braked or accelerated and a steering angle can be set at the same time.

According to a further-developing embodiment, an area in which the at least one critical object will move is additionally marked and output. This is done by means of the output device. For example, the image data that are generated by the at least one camera are displayed to the driver by means of the output device. In addition, starting from the trajectory that was determined for the at least one critical object, an area in which the at least one critical object will move is marked in the output image. The marking can, for example, be a colored highlighting of the affected area or a "graying out" of the unaffected area. As a result, the driver of the vehicle can be shown a zone in which the probability of a collision is increased and which should therefore be avoided.

According to a further developing embodiment, the at least one critical object is additionally observed over a period of time for plausibility checking, it being evaluated whether the at least one critical object does not meet the epipolar condition over the entire period of time. As soon as a critical object has been identified, it is observed over a predetermined period of time. The time span can, for example, be set at the factory and stored in the memory device of the control device. For example, the period of time can be one or more seconds, for example 2s, 5s, 10s, 20s, 30s, 40s, 50s or 60s, or several minutes, e.g. B. 2 min, 3 min or 4 min. In other words, the fulfillment of the epipolar condition is checked over several recordings of the at least one camera and is thereby verified.

A common minimum requirement for processing the image data is two consecutive frames, i.e. H. the processing rate is 1s per number of recorded frames per second of the at least one camera. The slower the vehicle moves, the longer the time between the processed frames must be so that sufficient changes are noticeable from one recording to the other. The time span can therefore also be less than 1s, for example the time between the number of frames per second of the at least one camera.

If the at least one critical object does not always meet the epipolar condition over this period of time, it is a plausibility-checked critical object. If the at least one critical object initially does not meet the epipolar condition over this period of time and then fulfills it again, it can be assumed that it is not a critical object. For example, the movement of the at least one critical object can have been stopped. Again, for example, it can be an error in the image data that was generated by means of the at least one camera.

The plausibility check thus includes a consolidation over the period. In addition, the plausibility check can also include a check that the critical object is actually approaching the ground level, i.e. H. whether it falls.

If the plausibility check thus shows that the at least one critical object represents a critical object checked for plausibility, the method steps described above are initiated and carried out. However, if the plausibility check shows that the at least one critical object does not represent a critical object that has been checked for plausibility, the method steps described above are not initiated.

The control device can be connected to at least one camera of the vehicle and to an output device of the vehicle, the control device having means that carry out the method that has already been described in the previous description. Connectable means that the control device is connected to the at least one camera of the vehicle and to the at least one output device of the vehicle when the control device is used in the vehicle. The control device therefore has interfaces by means of which these connections can be established. The interfaces are designed in such a way that data and signals can be exchanged via them. The control device can also be connectable to a drive system of the vehicle, to a braking system of the vehicle, to a steering system of the vehicle, to sensors for detecting the driving dynamics of the vehicle, to a warning device of the vehicle and / or to other vehicle systems. The control device can, for. B. be designed as an ECU or domain ECU.

The control device has means that carry out the method that has already been described in the previous description. These means can be implemented, for example, as a computer program product that runs on the control device. On the basis of the calculated probability of a collision between the vehicle and the at least one critical object, the control device controls the output device so that the warning is output. This activation takes place when the probability of a collision reaches or exceeds the threshold value, as also already described.

The computer program product comprises instructions which, when the program is executed by the control device already described, execute the method which has also already been described. The computer program product can comprise a program code that contains these commands. The program code can be embodied, for example, on a data carrier or as a downloadable data stream.

The vehicle has the control device that has already been described. In addition, the vehicle has the output device and the at least one camera, the control device being connected to the camera and to the output device. This has also already been described. The vehicle is set up to carry out automated driving functions. This has already been described in the previous description.

• 1 eine schematische Darstellung eines Fahrzeugs neben einem abfallenden Gelände nach einem Ausführungsbeispiel,
• 2 eine weitere schematische Darstellung des Fahrzeugs aus 1,
• 3 eine schematische Darstellung eines Verfahrens für das Fahrzeug aus 1 und aus 2.

Various exemplary embodiments and details of the invention are described in more detail with the aid of the figures explained below. Show it:

• 1 a schematic representation of a vehicle next to a sloping terrain according to an embodiment,
• 2 a further schematic representation of the vehicle 1 ,
• 3 a schematic representation of a method for the vehicle 1 and from 2 .

1 shows a schematic representation of a vehicle 1 next to a sloping terrain 2 according to an embodiment. The sloping terrain 2 is a slope. The vehicle 1 moves on a road in the direction of travel that is adjacent to the sloping terrain 2 runs.

The vehicle 1 instructs a camera 3 on, which is designed as a stereo camera. Alternatively, the vehicle can 1 multiple cameras 3 exhibit. The camera 3 is designed in such a way that it has a horizon in front of the vehicle 1 can capture. This is by means of the field of view 6th the camera 3 shown. Within the field of view 6th the camera captures the sloping terrain 2 and objects on it 4th , 4 * , 9 continuously. This is in 2 shown more clearly.

The vehicle 1 also has a control device 8th on that with the camera 3 connected is. The control device 3 is so with the camera 3 connected so that data and signals can be exchanged. For example, the image data received by the camera 3 generated when this is the sloping terrain 2 detected to the control device 8th to get redirected.

The vehicle 1 also has an output device 7th on. This is with the control device 8th connected. The connection is such that data and signals can be exchanged. The output device 7th is used to capture the image data from the camera 3 via the control device 8th are forwarded to the output device to output. The output device is also used 7th to do this, a warning to a driver of the vehicle 1 output when the object 4th as a critical object 4th has been classified. This procedure is used in 3 explained in more detail. The output device 7th is for example as a display within a passenger compartment of the vehicle 1 educated.

2 shows a further schematic representation of the vehicle 1 out 1 . It is made clear here how by means of the evaluation of the from the camera 3 generated image data by means of the control device 8th it is determined which of the objects 4th , 4 * , 9 , the on the sloping terrain 2 located as a critical object 4th , 4 * is identified.

On the sloping terrain 2 there are several static objects 9 . For a better overview, only three of these are static objects 9 provided with a reference number. These show a horizontal optical flow 5 because they don't move. For the sake of clarity, there are only three vectors that make up the optical flow 5 represent, provided with reference numerals. The optical flow 5 of static objects 9 runs through an epipole that goes through the camera 3 is given. The static objects 9 thus meet the epipolar condition. This is done by means of the image data received by the camera 3 are generated. The camera 3 creates two recordings, from their image data the optical flow 5 of the respective static objects 9 is determined. This is in 3 explained in more detail.

In contrast to the static objects 9 , assigns the object 4th , 4 * no horizontal optical flow 5 on. At the object 4th , 4 * it is the same object 4th , 4 * , which is shown from the perspective of a first recording and from the perspective of a second recording. The object 4th , 4 * exhibits an optical flow 5 which does not run through the epipole that runs through the camera 3 is given. The object 4th , 4 * thus does not meet the epipolar condition. It is therefore a critical object 4th , 4 * of which one is warned. This is done in 3 explained in more detail.

3 shows a schematic representation of a method V for the vehicle 1 out 1 and from 2 . In a first step 101 becomes the sloping terrain 2 by means of the camera 3 captured, so that the image data B. to be generated. The sloping terrain 2 is only within the field of view 6th of the vehicle 1 detected. At least two consecutive recordings are made. The image data B. are sent to the control device 8th forwarded.

In a second step 102 becomes the driving dynamics F of the vehicle 1 determined. This means that by means of the vehicle's own sensors, longitudinal dynamics, transverse dynamics, lifting movement, yawing movement, pitching movement, rolling movement and vibrations of the vehicle 1 be determined. These sensors are not shown. The data on driving dynamics F are sent to the control device 8th forwarded.

In a third step 103 is by means of the control device 8th from the image data B. the optical flow 5 of the object 4th ; 4 * , which is located on the sloping terrain 2 located, determined. In determining the optical flow 5 can depending on the shape of the camera 3 the data on driving dynamics F are included. This is necessary if, for. B. a mono camera or several mono cameras can be used. The driving dynamics F become a translation and a rotation of the vehicle 1 determined that existed between the creation of the individual recordings. When the translation and rotation of the vehicle 1 known can be the optical flow 5 be determined as the movement of the vehicle 1 on the image data B. can be obtained.

In a fourth step 104 is by means of the control device 8th based on the optical flow 5 and based on the image data B. evaluated whether the at least one object 4th ; 4 * the epipolar condition E. Fulfills. When the epipolar condition E. is fulfilled, it is not a critical object 4th , 4 * . The procedure V is thereupon for this very object 4th , 4 * not continued and can start over for another object. Meeting the epipolar condition E. is in 3 denoted by E +. The epipolar condition E. states that the optical flow of any static object is through the epipole. Of course, measurement inaccuracies must be taken into account.

When the epipolar condition E. however, is not fulfilled, it is a critical object 4th , 4 * . Failure to meet the epipolar condition E. is in 3 marked with E. This critical object 4th , 4 * will then be in a fifth step 105 by means of the control device 8th as a critical object 4th , 4 * marked. This marking takes place in the image data B. that are issued. For example, the critical object 4th , 4 * highlighted in color.

In a sixth step 106 becomes from the driving dynamics F by means of the control device 8th a trajectory I. of the vehicle 1 determined. This vehicle trajectory I. indicates in which direction, with which speed, under which acceleration and with which steering angle the vehicle 1 moves on the street.

In a seventh step 107 is based on the image data B. a trajectory II of the at least one critical object 4th ; 4 * estimated. This object trajectory II indicates in which direction, with which speed, under which acceleration, and with which steering angle the critical object 4th , 4 * along the sloping terrain 2 emotional.

In an eighth step 108 is based on the trajectory I. of the vehicle 1 and the trajectory II of the at least one critical object 4th ; 4 * a probability of collision K between the vehicle 1 and the at least one critical object 4th ; 4 * by means of the control device 8th calculated. In addition, a time-to-collision can be estimated. This probability of collision K it is checked whether this is a threshold value S. reached or exceeded. The likelihood of collision K indicates how high the probability is that the vehicle will continue to travel unchanged 1 to a collision with the critical object 4th , 4 * comes. The threshold S. can for example be set at the factory and in the memory device of the control device 8th be deposited. Alternatively, the threshold S. depending on the current situation by means of the control device 8th can be set automatically. For example, this threshold can be S. In the case of unfavorable road conditions, for example slippery roads, wet, dirty roads, which lead to a low coefficient of static friction, lower values are selected than in favorable road conditions (e.g. dry road surfaces).

Is the probability of a collision K less than the threshold S. (K <S), the procedure becomes V for the critical object 4th , 4 * not continued. It can then be carried out again for another object.

Reaches or exceeds the probability of a collision K however the threshold S. (K≥S), is in a ninth step 109 a warning W. by means of the output device 7th issued. The control device controls for this purpose 8th the output device 7th at. The warning W. can be done visually, acoustically, haptically or as a combination of these possibilities.

The examples shown here are only selected as examples. For example, in addition to the warning, the control device can be used to adapt the longitudinal dynamics of the vehicle. For example, a braking intervention, a steering intervention and / or an acceleration intervention can take place. As a result, the trajectory of the vehicle is adapted in such a way that the collision probability is reduced and is below the threshold value.

List of reference symbols

1

vehicle

2

sloping terrain

3

camera

4th

Object ( 1 . Admission)

4 *

Object ( 2 . Admission)

5

Optical flow

6th

Field of view

7th

Dispenser

8th

Control device

9

static object

101

first step

102

second step

103

Third step

104

fourth step

105

fifth step

106

sixth step

107

seventh step

108

eighth step

109

ninth step

B.

Image data

E.

Epipolar condition

K

Probability of collision

S.

Threshold

V

Procedure

W.

warning

I.

Vehicle trajectory

II

Trajectory of the object

Claims (8)

Method (V) for warning of mass movements on sloping terrain (2) for a vehicle (1), wherein - the sloping terrain (2) is recorded by means of at least one camera (3) so that image data (B) are generated, - a driving dynamics (F) of the vehicle (1) is determined, - an optical flow (5) of at least one object (4; 4 *) located on the sloping terrain (2) is determined from the image data (B), - it is evaluated whether the at least one object (4; 4 *) fulfills an epipolar condition (E), - The at least one object (4; 4 *) is marked as a critical object in the image data if the at least one object (4; 4 *) does not meet the epipolar condition (E), - A trajectory (I) of the vehicle (1) is determined from the driving dynamics (F), - On the basis of the image data (B), a trajectory (II) of the at least one critical object (4; 4 *) is determined, - Based on the trajectory (I) of the vehicle (1) and the trajectory (II) of the at least one critical object (4; 4 *) a collision probability (K) between the vehicle (1) and the at least one critical object (4; 4 *) is calculated, - A warning (W) is issued when the collision probability (K) reaches or exceeds a threshold value (S). Procedure (V) according to Claim 1 , whereby an intervention in the longitudinal dynamics of the vehicle (1) also takes place, so that a collision between the vehicle (1) and the critical object (4; 4 *) is avoided. Procedure (V) according to Claim 2 , the intervention in the longitudinal dynamics being designed as an acceleration intervention and / or as a braking intervention and / or a steering intervention. Method (V) according to one of the preceding claims, wherein an area in which the at least one critical object (4; 4 *) will move is additionally marked and output. Method (V) according to one of the preceding claims, wherein the at least one critical object (4; 4 *) is additionally observed over a period of time for plausibility checking, it being evaluated whether the at least one critical object (4; 4 *) has the epipolar Condition (E) not met over the entire period. Control device (8) for a vehicle (1), wherein the control device (8) can be connected to at least one camera (3) of the vehicle (1) and to an output device (7) of the vehicle (1), wherein the control device (8) Has means which carry out the method (V) according to one of the preceding claims. A computer program product comprising instructions which, when the program is executed by a control device (8), are executed according to Claim 6 , the procedure according to one of the Claims 1 to 5 To run. Vehicle (1) having a control device (8) according to Claim 6 , an output device (7) and at least one camera (3), the control device (8) being connected to the camera (3) and to the output device (7).

Images