METHOD FOR IDENTIFYING MISALIGNMENTS OF A SENSOR

Abstract

A method for identifying misalignments of a sensor of a sensor system of a motor vehicle, having at least one further sensor. The method includes ascertaining an associated position of an object in an overlap zone of at least two detection zones by way of the respective sensor which forms the overlap zone, comparing the ascertained positions with one another and, in the event of a deviation of the positions, identifying a misalignment. Alternatively or additionally, the movement of an object in a detection zone of a sensor can be tracked and used for identifying a misalignment.

Claims

1. A method for identifying misalignments of a sensor to be tested of a sensor system of a motor vehicle, having at least one further sensor, the sensor to be tested configured to detect objects in an associated detection zone during operation, and the detection zone of the sensor to be tested overlapping with a detection zone of the at least one further sensor, in an overlap zone, the method comprising the following steps: ascertaining, upon detection of an object in the overlap zone, a position of the object, using each of the sensor and the further sensor which form the overlap zone; comparing the ascertained positions with one another; identifying a misalignment of the sensor to be tested being in the event of a deviation of the ascertained positions.

2. A method for identifying misalignments of a sensor to be tested of a sensor system of a motor vehicle, having at least one further sensor, the sensor to be tested configured to detect objects in an associated detection zone during operation, the method comprising: tracking movement of an object toward the detection zone of the sensor to be tested, upon detection of the object in a detection zone of one of the at least one further sensors; determining a position of the object in the detection zone of the sensor to be tested at a specific point in time in advance based on the movement; ascertaining a position of the object at the specific point in time in the detection zone of the sensor to be tested; and identifying a misalignment of the sensor to be tested in the event of a deviation of the ascertained position from the position determined in advance.

3. The method as recited in claim 1, wherein: an associated extent of the object is ascertained for the identified object by each of the sensor to be tested and the further sensor; the ascertained extents are compared with one another, in the event of a deviation of the ascertained extents, the deviation of the extents is taken into account on identification of the misalignment.

4. The method as recited in claim 1, wherein: an associated orientation of the object is ascertained for the identified object, using each of the sensor to be tested and the further sensor; the ascertained orientations are compared with one another, in the event of a deviation of the ascertained orientations, the deviation of the orientations is taken into account on identification of the misalignment.

5. The method as recited claim 1, wherein the sensor to be tested and the further sensor are at a distance from one another.

6. The method as recited in claim 1, wherein, on identification of the misalignment, a value of the misalignment is determined and, and wherein, on identification of objects by the sensor to be tested having the misalignment, the value of the misalignment is taken into account as a correction.

7. The method as recited in claim 1, wherein, on identification of the misalignment, a value of the misalignment is determined and the misalignment is automatically corrected.

8. The method as recited in claim 1, wherein, on identification of the misalignment of the sensor to be tested, a function dependent on the objects identified by the sensor to be tested is disabled.

9. A motor vehicle, comprising: a sensor system having at least two sensors, each of the sensors configured to detecting an associated detection zone for objects in surroundings of the motor vehicle during operation; and a control device configured to identifying a misalignment of a sensor to be tested of the sensors of the sensor system, the detection zone of the sensor to be tested overlapping with a detection zone of at least one further sensor of the sensors, in an overlap zone, the control device configured to: ascertain, upon detection of an object in the overlap zone, a position of the object, using each of the sensor and the further sensor which form the overlap zone; compare the ascertained positions with one another; identify a misalignment of the sensor to be tested being in the event of a deviation of the ascertained positions.

10. The motor vehicle as recited in claim 1, wherein at least two of the sensors are arranged on different components of the motor vehicle and/or at a distance from one another.

Description

BRIEF DESCRIPTION OF THE DRAWING

[0041] FIG. 1 shows a plan view of a motor vehicle with a sensor system, according to an example embodiment of the present invention.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0042] In an exemplary embodiment, a sensor system 3, as is for example shown in FIG. 1, is used in a motor vehicle 1. The sensor system 3 comprises at least two sensors 2. In the exemplary embodiment shown, the sensor system 3 comprises purely by way of example four sensors 2. The respective sensor 2 detects objects 5 in an associated detection zone 4 indicated in FIG. 1. In the exemplary embodiment shown, the detection zones 4 of at least two of the sensors 2 overlap in an overlap zone 6. The respective detection zone 4 is in the surroundings of the motor vehicle 1. The sensors 2 thus detect objects 5 in the surroundings of the motor vehicle 1.

[0043] In the exemplary embodiment shown, the sensor system 3 has, purely by way of example, two sensors 2 which are mounted laterally on the motor vehicle 1 and oriented laterally in the direction of travel, which are hereinafter also denoted side sensors 13. The respective side sensor 13 ascertains a lateral zone as a detection zone 4 for objects 5. The respective side sensor 13 may be a radar sensor 7. The sensor system 3 additionally comprises two sensors 2 which are oriented forward in the direction of travel, which are hereinafter also denoted front sensors 14. One of the front sensors 2 is a radar sensor 7 and the other front sensor 2 a video sensor 8.

[0044] As is furthermore shown in FIG. 1, in the exemplary embodiment shown, the detection zone 4 of the respective side sensor 13 forms an associated overlap zone 6 with the detection zone 4 of the respective front sensor 14. In other words, the detection zone 4 of the respective side sensor 13 forms an associated overlap zone 6 with the detection zone 4 of the front sensor 14 configured as a radar sensor 7. In addition, the detection zone 4 of the respective side sensor 13 forms an associated overlap zone 6 with the detection zone 4 of the front sensor 14 configured as a video sensor 8. The detection zones 4 of the two front sensors 14 additionally form an associated overlap zone 6.

[0045] As is furthermore shown in FIG. 1, the detection zone 4 of the respective side sensor 13 also forms an overlap zone 6 with the detection zone 4 of the respective front sensor 14. These overlap zones 6 are thus formed by three sensors 2, namely in each case by one of the side sensors 13 and the two front sensors 14.

[0046] In the exemplary embodiment shown, the radar sensors 13 and the video sensor 8 are arranged, in particular mounted, on different components 10 of the motor vehicle 1. The radar sensors 7 are arranged on a bumper 11 and the video sensor 8 on a windshield 12 of the motor vehicle 1.

[0047] In order to identify a misalignment of a sensor 2a to be checked, account is taken of the overlap zone 6 of the sensor 2a to be tested with a further sensor 2b. In this case, an associated position of the object 5 is ascertained by way of the respective sensor 2a, 2b which forms the overlap zone 6. In other words, an associated position of the object 5 is ascertained, for example by triangulation, using data from the sensor 2a to be checked and using data from the further sensor 2b in the overlap zone 6. At least two ascertained positions for the object 5 are thus available. These ascertained positions are compared with one another. In the event of a deviation of the position ascertained by way of the sensor 2a to be tested from the at least one other ascertained position, in particular above a specified limit, a misalignment of the sensor 2a to be tested is identified. In the exemplary embodiment shown, it is assumed purely by way of example that, in the depiction of FIG. 1, it is the right-hand side sensor 13 which is the sensor 2a to be tested. Furthermore, in the exemplary embodiment shown, it is assumed purely by way of example that the front sensor 14 configured as a radar sensor 7 and/or the front sensor 14 configured as a video sensor 8 is/are the further sensor 2b.

[0048] As is likewise shown in FIG. 1, a misalignment can also be identified by an object 5 being identified in a detection zone 4 and the movement of the object 5 toward the detection zone 4 of a further sensor 2 being tracked. On the basis of the movement, a position of the object 5 in the detection zone 4 of the further sensor 2 at a specific point in time is determined in advance. At the specific point in time, the position of the object 5 in the detection zone 4 of the further sensor 2 is ascertained. In the event of a deviation of the ascertained position from the position determined in advance, a misalignment of one of the sensors 2 is identified. One of the sensors 2 may here be the sensor 2a to be checked. In other words, the sensor 2a to be checked may be that sensor 2 in whose detection zone 4 the object is identified and the movement of the object 5 toward the detection zone 4 of a further sensor 2b is tracked. The sensor 2a to be checked may likewise be that sensor 2 in whose detection zone 4 the position of the object 5 at the specific point in time is determined in advance by way of the further sensor 2b on the basis of the movement of the object 5. In the exemplary embodiment shown, an object 5 is identified at a point in time in the detection zone of the front sensor 14 configured as a radar sensor 7. In FIG. 1, the object 5 in the detection zone 4 of the front sensor 14 configured as a radar sensor 7 is therefore also denoted 5_t0. The movement of the object 5 toward the detection zone 4 of the side sensor 13, which is configured as a radar sensor 7 and shown on the right in the depiction of FIG. 1, is tracked. On the basis thereof, a position of the object 5 in the detection zone 4 of the left side sensor 13 configured as radar sensor 7 at a later specific point in time is determined in advance. According to this prior determination, the object 5 is thus located at the specific point in time in the detection zone 4 of the side sensor 13, which is configured as a radar sensor 7 and shown on the left in the depiction of FIG. 1. In FIG. 1, the object 5 in the detection zone 4 of the left side sensor 13 configured as a radar sensor 7 is therefore also denoted 5_t1. At the specific point in time, the position of the object 5 is furthermore ascertained by way of the left side sensor 13 configured as a radar sensor 7. If the ascertained position and the position determined in advance differ from one another, a misalignment is identified. In the exemplary embodiment shown, it is here assumed purely by way of example that the left side sensor 13 configured as a radar sensor 7 is the sensor 2a to be tested and the front sensor 14 configured as a video sensor 8 is the further sensor 2b.

[0049] Before the positions are compared with one another, it is conveniently ascertained whether the two sensors 2a, 2b are detecting the same object 5. For this purpose, in particular geometric and/or optical features of the object 5 may be ascertained by way of the respective sensor 2a, 2b and compared with one another.

[0050] On identification of the misalignment, an extent of the object 5 may be ascertained by way of the respective sensor 2a, 2b and compared with one another. In particular, this means that an associated extent of the object 5 is ascertained for the object 5 by way of the respective sensor 2a, 2b. At least two ascertained extents of the object 5 are thus available. These extents are compared with one another. In the event of a deviation of the extent ascertained by way of the sensor 2a to be tested from the at least one other ascertained extent, in particular above a specified limit, account is taken of the deviation of the extents on identification of the misalignment. In the simplest case this can be achieved by the deviation of the ascertained extents meaning a misalignment. In other words, a misalignment is identified in the event of a deviation of the ascertained extents. Alternatively or additionally, the deviation of the ascertained extents can be taken into account as a weighting on identification of the misalignment.

[0051] It is likewise possible, on identification of the misalignment by way of the respective sensor 2a, 2b, to take account of an orientation of the object 5, for example a longitudinal extent of the object 5. In particular, this means that an associated orientation of the object 5 is ascertained for the identified object 5 by way of the respective sensor 2a, 2b. At least two ascertained orientations are thus available which are compared with one another. In the event of a deviation of the orientation ascertained by way of the sensor 2a to be tested from the at least one other ascertained orientation, account is taken of the deviation of the orientations on identification of the misalignment. In the simplest case, a misalignment is identified in the event of a deviation of the ascertained orientations, in particular above a specified limit. The deviation of the ascertained orientations can likewise be taken into account as a weighting on identification of a misalignment.

[0052] If a misalignment of the sensor 2a to be tested is identified, this can be taken into account in various ways. For example, a value of the misalignment can be determined and, on identification of objects 5 by the sensor 2a to be tested, taken into account as a correction. It is alternatively possible to eliminate the misalignment automatically. For this purpose, on identification of a misalignment, a value of the misalignment can be determined and the misalignment automatically eliminated by an alignment. For this purpose, the sensor system 3 and/or the motor vehicle 1 may have appropriate actuators which are not shown. It is also possible, on identification of a misalignment of the sensor 2a, to disable a function in the motor vehicle 2 which is dependent on objects 5 detected by the sensor 2a to be tested. In particular, these are functions of at least partially autonomous driving and/or automatic actuation of components of the motor vehicle 1, for example automatic braking, high-beam assist and the like.

[0053] The described method is carried out automatically. A control device 9 is used for this purpose. The control device 9 is shown in greatly simplified form in FIG. 1. The control device 9 is communicatively connected to the sensors 2 and configured to carry out the method. The control device 9 may for this purpose contain an algorithm, in particular a computer program product.

[0054] The misalignment is preferably identified by way of two sensors 2a, 2b which are arranged on different components 10 of the motor vehicle 1 and/or at a distance from one another. This in particular thus prevents a mechanical application of force, in particular damage to the component 10 and/or replacement of component 10, from resulting in misalignment of both sensors 2a, 2b or at least avoids this risk.

[0055] Accordingly, a misalignment is preferably identified by an object 5 in the overlap zone 6 formed by the detection zone 4 of the front sensor 14 configured as a video sensor 8 and the detection zone 4 of the side sensor 13, which is configured as a radar sensor 7 and shown on the left in the depiction of FIG. 1. It can thus be identified whether it is the front sensor 14 configured as a video sensor 8 or the side sensor 13 shown on the left in the depiction of FIG. 1, which is misaligned. Sensors 2 of different types are thus furthermore used for identifying the misalignment.

[0056] A misalignment is likewise preferably identified by an object 5 in the detection zone 4 of the front sensor 14 configured as a video sensor 8 and in the detection zone 4 of one of the side sensors 13 and the detection zone 4 of the front sensor 14 configured as a radar sensor 7. It can thus be highly reliably identified whether one of the front sensors 14 or the side sensor 13 is misaligned.