DEVICE AND METHOD FOR ALIGNING A HEADLIGHT ADJUSTING DEVICE

Description

[0001] The present disclosure relates to an efficient and precise alignment of an adjustment test device for checking a headlight of a motor vehicle.

[0002] US 8,582,091 B2 describes a headlight aiming system comprising a headlight aiming device, a control system, an alignment system, and an image processing system. The headlight aiming device detects a light beam from a vehicle headlight and provides feedback on its position to the control system. The image processing system interacts with the headlight aiming device and, in particular, performs an initial alignment of the headlight with the optical system of the headlight aiming device. For this purpose, a marking is provided at a known location on the headlight, which is detected by a digital camera of the image processing system. The control system then positions the headlight so that it is directed precisely at the optical system of the headlight aiming device. Prior to headlight alignment, the vehicle is positioned in front of the headlight aiming system using position markers.

[0003] GB 2 307 312 A relates to a headlight aiming device for a production line, comprising a gantry on which a horizontally movable rack is mounted. A housing containing a camera is vertically movable on the rack. The housing also includes a Fresnel lens whose focal length is selected to project a headlight beam image onto the rear wall of the housing. To aim a headlight, beam images from an optimally aligned test headlight in various settings (low beam, high beam, etc.) are first captured and converted into digital grayscale images. These grayscale images are stored in a control unit and serve as a reference for adjusting an identical headlight using a grayscale algorithm.

[0004] The adjustment of a headlight aiming tester (SEP, also known as a headlight measuring device) can be achieved using photodiodes. A mobile robot, to which the aiming tester is attached, is moved along the entire horizontal axis of the front of a vehicle to determine the exact position of the headlight being checked. This process is typically quite time-consuming. Furthermore, backlighting can cause false readings and / or inaccuracies in the position determination.

[0005] Consequently, it is an objective of the present disclosure to address the aforementioned problems of the prior art and, in particular, to offer a method and a corresponding device which enable a time-efficient and precise alignment of a headlight aiming test device relative to a headlight.

[0006] To solve the aforementioned problem, the features of the independent claims are proposed. The dependent claims relate to preferred embodiments.

[0007] According to one aspect, a device for aligning a headlight aiming tester relative to a headlight (e.g., a low beam, a high beam, a fog light, etc.) of a (motor) vehicle is described. The headlight may be located at the front of the vehicle. The aiming tester may be part of a lighting robot for checking vehicle headlights. The device may also be a control device for the lighting robot.

[0008] The device is designed to capture at least one camera image of the vehicle's headlight or a camera image of the vehicle itself during an approach maneuver to the headlight aiming device (i.e., while the vehicle is moving towards the aiming device). The camera captures a temporal sequence of images for a corresponding sequence of points in time during the approach maneuver. The camera may be a stereo and / or 3D camera. In other words, the camera data may include stereoscopic and / or spatial information. This information enables particularly precise, fast, and efficient alignment of the aiming device relative to the vehicle's headlight being tested.

[0009] Alternatively or additionally, the device can be configured to use sensor data from a (laser) distance meter to determine the distance between the aiming device and the vehicle and / or the vehicle's headlight. Alternatively or additionally, the device can be configured to use the temporal sequence of (two-dimensional, 2D) camera images to determine the distance between the aiming device and the vehicle and / or the vehicle's headlight. A machine learning-based estimation method can be used to estimate the distance. Based on the determined distance, the aiming device can then be aligned relative to the vehicle's headlight being checked.

[0010] The device is configured to determine the position of the headlight within the camera image. The device is also configured to detect the vehicle's headlight within the camera image and / or determine its position within the camera image using an image analysis method, such as one based on machine learning and / or artificial intelligence. For example, a bounding box can be determined (in particular, placed) around the headlight. Furthermore, the position of the bounding box within the camera image can be determined. This position can correspond to a pixel position within the camera image.

[0011] The image analysis method can be designed to recognize one or more objects in a (2D or 3D) camera image and, if necessary, to determine the position of those objects. The image analysis method can thus include an object recognition method. The image analysis method (in particular, one or more neural networks within the image analysis method) can be trained using machine learning.

[0012] Alternatively or additionally, the device can be configured to determine the headlight type from a set of different headlight types using image analysis. The set of headlight types can include, for example, low beam, high beam, and / or fog lights. Thus, the headlight can be classified using image analysis.

[0013] The device is further configured to align the headlight aiming device relative to the vehicle's headlight, depending on the determined position of the headlight within the camera image (e.g., depending on the determined pixel position). The aiming device can be aligned relative to the headlight to ensure that the headlight has a specific reference position within the camera image (e.g., the center of the camera image). The alignment of the aiming device can be performed in a plane that is essentially perpendicular to the vehicle's direction of movement during the approach maneuver.

[0014] This document describes a device that uses the evaluation of a camera image to align the headlight aiming device relative to the headlight being tested during the approach maneuver. This enables a particularly time-efficient and precise headlight test.

[0015] As explained above, the device can be configured to repeatedly, and in particular periodically, (at a sequence of successive points in time) capture a camera image of the vehicle's headlight during the vehicle's approach maneuver and to determine the headlight's position within each camera image. The headlight aiming device can then be repeatedly aligned relative to the vehicle's headlight based on the headlight's position determined within each camera image.

[0016] By continuously adjusting the alignment of the adjustment test device during the approach maneuver, a particularly precise alignment of the adjustment test device can be achieved for a particularly precise check of the headlight.

[0017] The device can be configured to repeatedly adjust the alignment of the headlight aiming device relative to the vehicle's headlight during the vehicle's approach maneuver, such that upon completion of the maneuver, the headlight aiming device is aligned with the target position for performing the measurement. This target position can, for example, correspond to the reference position within the camera image when a specific distance between the aiming device and the headlight is reached. The aiming device can thus be gradually moved to the target position during the approach maneuver, enabling a particularly efficient and precise headlight check.

[0018] The device can be configured to determine (for a camera image at a specific point in time) that the determined position of the headlight aiming device deviates from the specified reference position within the camera image, in particular from the center of the camera image. The headlight aiming device can then be aligned (at that point in time) to bring the position of the headlight within the camera image subsequently captured during the approach maneuver closer to the reference position, and / or to ensure that the position of the headlight within the camera image subsequently captured during the approach maneuver corresponds to the reference position.

[0019] During the approach maneuver, the aiming device can be repeatedly moved towards the reference position within the captured camera image. This ensures, in a particularly efficient and reliable manner, that the aiming device is already in a specific target orientation for performing a measurement on the headlight upon completion of the approach maneuver.

[0020] The camera is preferably permanently attached to the headlight aiming device, so that a change in the orientation of the headlight aiming device leads to a corresponding change in the orientation of the camera relative to the vehicle's headlight. This allows for particularly precise alignment of the aiming device based on camera images.

[0021] The camera and / or headlight aiming device can be moved within a plane via one or more (electrically operated) guide elements to change the alignment of the camera and / or headlight aiming device relative to the vehicle's headlight. This plane can be essentially perpendicular to the vehicle's direction of movement during the approach maneuver. This allows for particularly precise and convenient automatic alignment of the aiming device.

[0022] The device can be configured to determine, based on the camera image (captured during the approach maneuver), whether the vehicle's headlight is emitting light or not. This can be achieved using an image analysis method. If it is determined that the vehicle's headlight is not emitting light, a notification can be issued to the user of the vehicle and / or the headlight aiming device. The notification can instruct the user to activate the vehicle's headlight. Alternatively, the user may be informed that the vehicle's headlight is defective. This can further increase the efficiency of the headlight testing procedure.

[0023] The device can be configured to determine whether the vehicle's headlight has reached a target distance to the headlight aiming device during the approach process. In particular, the device can be configured to determine distance information regarding the distance between the vehicle's headlight and the headlight aiming device based on at least one camera image (e.g., based on the stereoscopic and / or spatial information of the camera image and / or on the sequence of camera images). It can then be verified, in a particularly efficient and precise manner, based on this distance information, whether the vehicle's headlight has reached the target distance to the headlight aiming device for the headlight measurement process during the approach process.

[0024] Furthermore, the device can be configured to initiate (e.g., by issuing a notification) the termination of the vehicle's approach process upon determining that the target distance has been reached. Once the approach process is complete, the headlight aiming device can then directly perform the measurement for the vehicle's headlight (without requiring any further alignment of the aiming device). This enables a particularly efficient and precise measurement process.

[0025] The device can be configured to detect a number of different headlights of the vehicle within the camera image (captured during the approach maneuver). A first headlight can then be selected from this group (e.g., according to a specific, predefined selection criterion), and its position within the camera image can be determined. Furthermore, the headlight aiming device can be aligned relative to the first headlight based on its determined position within the camera image, in order to perform a measurement on that headlight.

[0026] During the approach maneuver, the first headlight can thus be aligned in order to be able to carry out the measurement process for the first headlight immediately upon completion of the approach maneuver.

[0027] The device can further be configured to determine, based on the camera image, position information regarding the position of at least one second headlight relative to the first headlight (even during the approach maneuver). This position information can be determined for multiple second (i.e., additional) headlights. The position information can, for example, indicate the lateral offset in the horizontal and / or vertical direction between the first headlight and the second headlight (within the plane of the one or more guide elements of the light robot).

[0028] The device can be configured to determine position information based on stereoscopic and / or spatial information from one or more camera images. Alternatively or additionally, the position information can be determined based on the temporal sequence of camera images captured at the corresponding sequence of times during the vehicle's approach maneuver.

[0029] Thus, position information regarding the relative position of one or more additional headlights of the vehicle can be determined during the approach maneuver (based on one or more camera images).

[0030] Furthermore, the device can be configured to align the headlight aiming device relative to the second headlight (possibly alone and / or exclusively) after the measurement process for the first headlight, based on the position information, in order to perform the measurement process for the second headlight. The position information can thus be used after completion of the approach maneuver for particularly efficient and precise alignment of the aiming device to perform measurements for one or more additional headlights.

[0031] According to another aspect, a lighting robot for checking a vehicle's headlight is described. The lighting robot comprises a headlight aiming device configured to perform a measurement procedure to check the vehicle's headlight, and a camera configured to capture at least one camera image in relation to the vehicle's headlight (as the vehicle approaches the aiming device). The lighting robot further comprises one or more (electrically operated) guide elements configured to change the orientation of the headlight aiming device relative to the vehicle's headlight (e.g., within a plane perpendicular to the vehicle's direction of travel). The lighting robot also includes the (control) device described in this document.

[0032] According to another aspect, a method for aligning a headlight aiming device relative to a vehicle's headlight is described. The method includes capturing, using a (3D) camera, at least one image of the vehicle's headlight during the approach maneuver to the headlight aiming device, and determining the headlight's position within the camera image. For example, the position of the headlight's bounding box within the camera image can be determined. The method further includes aligning the headlight aiming device relative to the vehicle's headlight based on the determined position of the headlight within the camera image.

[0033] Another aspect described is a software (SW) program. The SW program can be configured to run on a processor and thereby execute the procedure described in this document.

[0034] Another aspect describes a storage medium. This storage medium can include a software program configured to run on a processor and thereby execute the procedure described in this document. Brief description of the characters

[0035] Figure 1: shows an exemplary light robot for checking a motor vehicle headlight; Figure 2a: shows an exemplary camera image of a vehicle approaching the light robot; Figure 2b: shows a motor vehicle with an activated headlight; Figure 2c: illustrates an exemplary distance measurement using camera data; Figure 2d: shows a vehicle with additional fog lights; Figure 2e: shows an exemplary sequence of camera images during a vehicle approach process; and Figure 3: shows a flowchart of an exemplary procedure for positioning a headlight aiming test device.

[0036] The following section describes in detail examples of the present revelation using illustrative figures. The characteristics of the examples can be combined in whole or in part, and the subject matter is not limited to the examples described.

[0037] Figur 1 Figure 1 shows an exemplary lighting robot 100 for checking one or more headlights 121, 122 of a motor vehicle 120. The vehicle 120 can be positioned by a driver in front of the adjustment testing device 107 of the lighting robot 100. The adjustment testing device 107 can be attached to one or more guide elements 103, 104 to allow horizontal and / or vertical adjustment of its position (relative to the ground on which the vehicle 120 is standing). The adjustment testing device 107 can be positioned in a defined manner (e.g., centrally) in front of a headlight 121, 122 of the vehicle 120 by means of the one or more guide elements 103, 104 of the lighting robot 100. The adjustment test device can then record and evaluate measurement data relating to the light emitted by the headlight 121, 122, e.g. to check and, if necessary, adjust the alignment of the headlight 121, 122.

[0038] As explained at the outset, this document deals with enabling a particularly (time-)efficient and precise positioning of the adjustment test device 107 in front of the one or more headlights 121, 122 of a vehicle 120. For this purpose, the device shown in Fig. 1 The illustrated light robot 100 is equipped with a camera 102, in particular a 3D and / or stereoscopic camera. The camera 102 can be arranged, in particular attached, to the one or more guide elements 103, 104 of the light robot 100 in a defined manner, e.g., with a specific horizontal and / or vertical offset, relative to the setting test device 107. Furthermore, the camera 102 can be configured to be moved together with the setting test device 107 by means of the one or more guide elements 103, 104, such that a change in the position of the camera 102 leads to a corresponding change in the position of the setting test device 107 (for any position changes).

[0039] The camera 102 preferably has the same detection direction as the adjustment testing device 107. In particular, during a measurement process, the camera 102 is directed towards the one or more headlights 121, 122 of a vehicle 120 to be checked. Thus, camera data 105 relating to the one or more headlights 121, 122 to be checked can be acquired by the camera 102 and provided to a (control) device 101 of the light robot 100. The camera data 105 can comprise a temporal sequence of camera images, each showing at least a section of the vehicle 120.

[0040] The light robot 100, in particular the (control) device 101, can be configured to communicate with the vehicle 120 via a (possibly wireless) communication link 106, e.g. to transmit instructions to the driver of the vehicle 120 regarding how the vehicle 120 should approach the light robot 100, and / or when the vehicle 120 should be brought to a standstill (in order to end the approach process).

[0041] The light robot 100 can be trained to acquire camera data 105 during an approach process in which the vehicle 120 approaches the alignment test device 107 and the camera 102 of the light robot 100. The camera data 105, in particular a camera image from the camera data 105, can be evaluated to detect that the vehicle 120 is approaching the light robot 100. Furthermore, one or more headlights 121, 122 of the approaching vehicle 120 can be detected using an image analysis method. The image analysis method can, for example, include object recognition based on machine learning and / or artificial intelligence (AI). Based on the one or more detected headlights 121, 122, the alignment test device 107 can then be positioned during the approach of the vehicle 120 to enable the most time-efficient and precise positioning of the alignment test device 107.

[0042] In the Figuren 2a bis 2e Different aspects of the positioning and / or alignment of the adjustment test device 107 during the approach process of the vehicle 120 are shown. Fig. 2a Figure 1 shows an example camera image 200 of an approaching vehicle 120 with two headlights 121, 122. In the example shown, the center 201 of the camera image 200 is located midway between the two headlights 121, 122. Using an image analysis method, the one or more headlights 121, 122 of the vehicle 120 can be identified and, if necessary, represented as bounding boxes 131, 132. Furthermore, the positions of the one or more headlights 121, 122 (or the corresponding bounding boxes 131, 132) can be determined.

[0043] As in Fig. 2b As shown, the device 101 can be configured to detect, by analyzing a camera image 200, whether a headlight 121, 122 of the vehicle 120 is emitting light or not. In the example shown, it can be detected, for instance, that the first headlight 121 is emitting light and that the second headlight 122 is not. Feedback regarding the status of the one or more headlights 121, 122, such as activated, deactivated, or defective, can then be given to the driver of the vehicle 120. For example, the driver can be prompted to activate the one or more headlights 121, 122 if, based on the camera data 105, it is detected that the one or more headlights 121, 122 are inactive. This further increases the efficiency of the measurement method.

[0044] The device 101 can be configured to determine distance information 200, based on one or more camera images, with respect to the longitudinal distance 211 between the vehicle 120 and the light robot 100, and / or distance information 212, with respect to the lateral distance 212 between two different headlights 121, 122 of the vehicle 120. The distance information can be determined, in particular, when using a 3D and / or stereoscopic camera 102, based on the camera data 105.

[0045] For example, as exemplified in Fig. 2c As shown, a first distance a between camera 102 and headlight 121, as well as a second distance b between camera 102 and headlight 122, are determined. Using the following formula Δ x = a 2 + b 2 − 2 ab cos γ From this, the transverse distance Δ can be calculated. x212 between the two spotlights 121, 122 are determined, where y is the angle between the two beams between the camera 102 and the respective spotlight 121, 122.

[0046] The longitudinal distance 211 between the light robot 100 and the spotlights 121, 122 can be determined, for example, based on the camera data 105 (from a 3D and / or stereoscopic camera 102). Alternatively or additionally, the longitudinal distance 211 can be determined using a (laser) distance sensor and / or based on 2D camera data 105 using an AI and / or machine-learned estimation algorithm.

[0047] The device 101 can be configured to repeatedly, and in particular periodically, determine distance information regarding the longitudinal and / or lateral distance 211, 212 during an approach maneuver of the vehicle 120. The approach maneuver can then be controlled depending on the distance information. For example, the driver of the vehicle 120 can be prompted to bring the vehicle 120 to a standstill (to end the approach process) when, based on the distance information, it is recognized that a certain target longitudinal distance between the vehicle 120 and the one or more headlights 121, 122 has been reached. This enables a particularly efficient and precise measurement process.

[0048] As exemplified in Fig. 2d As shown, the measures described in this document can be used for one or more different types of headlights. The type of a headlight 121, 122 can be determined based on the camera data 105 using an image analysis method (based on machine learning and / or AI). For example, the described measures can be used to check one or more fog lights 220.

[0049] During an approach maneuver, position information 221 regarding the position of the fog light 220 relative to a (main) headlight 121 of the vehicle 120 can be determined based on camera data 105. The position information 221 can indicate how the fog light 220 is positioned relative to the main headlight 121 on the vehicle 120 (e.g., with respect to an offset along the transverse axis and / or along the vertical axis of the vehicle 120). The position information 221 for the fog light 220 can be determined during the approach maneuver. Similarly, position information 221 for one or more additional headlights 122 of the vehicle 100 can be determined (also during the approach maneuver).

[0050] The position information 221 for one or more additional headlights 122, 220 can be used during a measurement process for an additional headlight 122, 220 to position the aiming device 107 in front of the additional headlight 122, 220 in order to check this additional headlight 122, 220. In particular, the aiming device 107 can be aligned with the respective additional headlight 122, 220 directly after the measurement process for the main headlight 121 based on the position information 221 (determined during the approach process) (without having to locate the additional headlight 122, 220 beforehand). This can be done accordingly for the measurement processes for all additional headlights 122, 220.By determining the position information 221 beforehand, the relative distance between the respective additional headlight 122, 220 and the main headlight 121 is known and can therefore be used for the accelerated execution of the measurement processes for the one or more additional headlights 122, 220.

[0051] Fig. 2e Figure 1 shows an exemplary sequence of camera images 200 that were captured during the approach process of vehicle 120. The topmost camera image 200 shows the entire front of vehicle 120, and based on this camera image 200 (using an image analysis method, in particular an object recognition method), a set of one or more headlights 121, 122, 220 of vehicle 120 can be determined. Furthermore, one headlight 121 can be selected from the set of headlights 121, 122, 220 as the main headlight for the further approach process. For the zero, one or more additional headlights 122, 220 from the set of headlights 121, 122, 220, position information 212, 221 can then be determined, which in each case shows the relative position of the additional headlight 122, 122 relative to the main headlight 121.The position information 212, 221 can be determined on the basis of a camera image 200 and / or on the basis of a temporal sequence of camera images 200.

[0052] During the approach maneuver of the vehicle 120, the camera 102 and the alignment test device 107 can be moved by the one or more guide elements 103, 104 depending on the position of the main headlight 121 within the individual camera images 200, in particular such that the main headlight 121 is located at a specific reference position 201 within the individual camera images 200, e.g. in the center of the respective camera image 200. This is exemplified in the sequence of camera images 200 (from top to bottom) in Fig. 2e As shown, in the uppermost camera image 200, the main headlight 121 is not yet positioned at the reference position 201 within the camera image 200. The camera 102 and the adjustment testing device 107 can then be moved by the one or more guide elements 103, 104 to ensure that the main headlight 121 is positioned at the reference position 201 within the following camera images 200. During the approach process, the position of the camera 102 and the adjustment testing device 107 can then be repeatedly adjusted to ensure that the main headlight 121 is positioned at the reference position 201 within each of the respective camera images 200.

[0053] Thus, the adjustment testing device 107 can be positioned in front of the main headlight 121 during the approach maneuver of the vehicle 120, so that the measurement of the main headlight 121 can begin without delay after the approach maneuver has been completed.

[0054] Following the measurement of the main headlight 121, the position information 212, 221 (determined during the approach maneuver) for another headlight 122, 220 from the set of headlights 121, 122, 220 can be used to position the aiming device 107 in front of the other headlight 122, 220 using the one or more guide elements 103, 104. Typically, no further data is required for this, so the repositioning of the aiming device 107 can be carried out in a particularly efficient manner. The same procedure can be followed for all other headlights 122, 220.

[0055] A method and a corresponding device 101 are described that make it possible to detect one or more headlights 121, 122, 220 of a vehicle 120 and to determine their position. The one or more headlights 121, 122, 220 are detected as the vehicle 120 approaches the test or measuring device 107, and not only when the vehicle 120 is positioned in front of the light robot 100, in particular in front of the adjustment test device 107. This allows for a time saving in terms of measuring the light output of a headlight 121, 122, 220.

[0056] The light robot 100 described in this document includes a camera 102 for detecting and determining the position of one or more headlights 121, 122, 220 of a vehicle 120. By using a 3D-capable camera 102 (or alternatively or additionally by using a (laser) rangefinder or by estimating the distance from a 2D camera image 200 using a machine learning-based estimation method), the spatial position of the one or more headlights 121, 122, 220 can also be determined precisely. When a vehicle 120 approaches the alignment test device 107, it can be detected (based on the camera data 105) whether one or more headlights 121, 122, 220 are not illuminated. Furthermore, the distance 212 between the individual headlights 121, 122 can be determined via the 3D-capable camera 102 as the vehicle 120 approaches the adjustment testing device 107.

[0057] As the vehicle 120 approaches the aiming device 107, the position of a (main) headlight 121 can be continuously determined. The robot 100 can align the camera 102 mounted on the movable aiming device 107 by horizontal and vertical movements so that the first (main) headlight 121 to be measured is always in the center 201 of the respective camera image 200 as the vehicle 120 approaches the light robot 100.

[0058] When the vehicle 120 approaches the adjustment testing device 107, feedback can be given to the driver of the vehicle 120 by measuring the distance 211 between the (main) headlight 121 and the adjustment testing device 107, when the desired target distance has been reached.

[0059] Because it is possible to detect whether individual or all headlights 121, 122, 220 are not illuminated as the vehicle 120 approaches the test stand 100, the operator of the test stand 100 can be notified of this at an early stage. The message regarding a headlight 121, 122, 220 could, for example, be "Turn on headlights" or "Headlight defective." This allows the test procedure to be optimized.

[0060] As a vehicle 120 approaches, the distances 212 between the individual headlights 121, 122, 220 can be determined. This also applies to the low beam and / or high beam 121, 122 and / or to the one or more fog lights 220. Once these distances 212 are known, the aiming device 107 can be moved directly to the next headlight 122, 220 after measuring the first (main) headlight 121, without having to search for the next headlight 122, 220 again. This speeds up the testing process.

[0061] Because the aiming device 107 is continuously aligned with the first headlight 121 to be measured as the vehicle 120 approaches, measurement of this headlight 121 can begin immediately as soon as the vehicle 120 has moved to the desired target distance. This speeds up the testing process.

[0062] By being able to measure spatial distances 211, 212, the user can be given feedback when the vehicle 120 approaches, when the desired target distance between the one or more headlights 121, 122, 220 and the adjustment testing device 107 has been reached.

[0063] By using a camera 102 and evaluating the camera images 200 with methods from the field of computer vision and machine learning, it is possible to detect whether one or more headlights 121, 122, 220 are included in the respective camera image 200, and their 2D position within the respective camera image 200 can be determined. This method works independently of the vehicle type of the respective vehicle 120 and does not require a database of possible vehicle types and their respective headlight positions.

[0064] The detection and position determination of one or more headlights 121, 122, 220 can be carried out while the vehicle 120 is approaching the light robot 100. This allows the headlight aiming test device 107 to align itself with the first headlight 121 to be tested while the vehicle 120 is approaching, so that measurement of this headlight 121 can begin immediately once the vehicle 120 has reached the desired measuring distance. This speeds up the testing process.

[0065] Furthermore, with the aid of a 3D-capable camera 102, the relative distances 122, 221 between the individual headlights 121, 122, 220 can be determined as the vehicle 120 approaches, so that after measuring the first headlight 121, it is already known how the light robot 100 must proceed to position itself in front of the one or more other headlights 122, 220. Therefore, no further search for the one or more additional headlights 122, 220 is necessary, and the testing process can be further accelerated.

[0066] In particular, when using a 3D-capable camera 102, the (longitudinal) distance 211 of the vehicle 120 or of the one or more headlights 121, 122 to the headlight aiming test device 107 can be determined, which allows the user to receive feedback when the vehicle 120 approaches the light robot and the desired distance to the headlight aiming test device 107 has been reached.

[0067] By using a camera 102 and machine learning and / or AI, headlights 121, 122, 220 that are switched off or not illuminated due to a defect can also be detected. Existing image data of headlights in various states and of vehicles can be used for training, and / or training runs can be carried out with real vehicles or using video screens. Therefore, feedback can be given to the user as the vehicle 120 approaches the light robot 100, and the inspection process can be accelerated if no headlights 121, 122, 220 are illuminated, indicating that the lights are switched off, or if only one or more individual headlights 121 are illuminated, for example, the low beam 121 on the right side but not on the left, indicating that a headlight 122 is defective.

[0068] Fig. 3Figure 1 shows a flowchart of a (possibly computer-implemented) procedure 300 for aligning a headlight aiming test device 107 relative to a headlight 121, 122, 220 of a (motor) vehicle 120 (e.g., a car, a truck, a bus, a motorcycle, etc.). The aiming test device 107 can be part of a light robot 100.

[0069] Method 300 comprises capturing 301, during an approach maneuver of the vehicle 120 to the headlight aiming device 107 and using a camera 102 (of the light robot 100), at least one camera image 200 of the headlight 121, 122, 220 of the vehicle 120. Thus, a camera image 200 is already captured while the vehicle 120 is still moving towards the aiming device 107.

[0070] Method 300 further comprises determining 302 the position of the spotlight 121, 122, 220 within the camera image 200. For this purpose, an image analysis method (based on machine learning and / or AI) can be used. The position of the spotlight 121, 122, 220 can indicate the coordinates of the spotlight 121, 122, 220 (e.g., the center point or a corner point of the spotlight 121, 122, 220) within the image plane of the camera image 200.

[0071] Furthermore, the method 300 comprises aligning the headlight aiming device 107 relative to the headlight 121, 122, 220 of the vehicle 120, depending on the determined position of the headlight 121, 122, 220 within the camera image 200. Thus, a relative positioning of the aiming device 107 relative to the headlight 121, 122, 220 (to be checked) can be achieved (already during the approach maneuver), in particular such that the aiming device 107 has a target position for carrying out the measurement process for the headlight 121, 122, 220 at the end of the approach maneuver. This enables a particularly efficient and precise inspection of a headlight 121, 122, 220.

Claims

1. Device (101) for aligning a headlamp setting test device (107) relative to a headlamp (121, 122, 220) of a vehicle (120); wherein the device (101) is connected in terms of signal technology to a camera (102) and is configured - during an approach maneuver of the vehicle (120) to the headlamp setting test device (107), to detect a temporal sequence of camera images (200) of the headlamp (121, 122, 220) of the vehicle (120) for a corresponding sequence of points in time on the basis of the camera (102); - on the basis of a machine-learned and / or artificial intelligence-based image analysis method, to detect the headlamp of the vehicle within a camera image and to determine a position of the headlamp (121, 122, 220) within the camera image (200); and - to align the headlamp setting test device (107) relative to the headlamp (121, 122, 220) of the vehicle (120) depending on the determined position of the headlamp (121, 122, 220) within the camera image (200).

2. Device (101) according to Claim 1, wherein the device (101) is configured, during the approach maneuver of the vehicle (120), repeatedly, in particular periodically, - to detect a camera image (200) of the headlamp (121, 122, 220) of the vehicle (120) on the basis of the camera (102); - to determine the position of the headlamp (121, 122, 220) within the respective camera image (200); and - to align the headlamp setting test device (107) relative to the headlamp (121, 122, 220) of the vehicle (120) depending on the respectively determined position of the headlamp (121, 122, 220) within the respective camera image (200).

3. Device (101) according to Claim 2, wherein the device (101) is configured to repeatedly adapt the alignment of the headlamp setting test device (107) relative to the headlamp (121, 122, 220) of the vehicle (120) during the approach maneuver of the vehicle (120) in such a way that, when the approach maneuver of the vehicle (120) is ended, the headlamp setting test device (107) has a target alignment for carrying out a measurement process for the headlamp (121, 122, 220) of the vehicle (120).

4. Device (101) according to one of the preceding claims, wherein the device (101) is configured - to determine that the determined position of the headlamp setting test device (107) deviates from a reference position (201) within the camera image (200), in particular from a center point of the camera image (200); and - to align the headlamp setting test device (107) in order - to approximate the position of the headlamp (121, 122, 220) to the reference position (201) within a camera image (200) subsequently detected during the approach maneuver; and / or - to cause the position of the headlamp (121, 122, 220) to correspond to the reference position (201) within a camera image (200) subsequently detected during the approach maneuver.

5. Device (101) according to one of the preceding claims, wherein the device (101) is configured - to detect a set of different headlamps (121, 122, 220) of the vehicle (120) within the camera image (200); - to select a first headlamp (121) from the set of headlamps (121, 122, 220); - to determine the position of the first headlamp (121) within the camera image (200); and - to align the headlamp setting test device (107) relative to the first headlamp (121) depending on the determined position of the first headlamp (121) within the camera image (200) in order to carry out a measurement process for the first headlamp (121).

6. Device (101) according to Claim 5, wherein the device (101) is configured - to determine position information (212, 221) relating to a position of at least one second headlamp (122, 220) relative to the first headlamp (121) on the basis of the camera image (200); and - to align the headlamp setting test device (107) relative to the second headlamp (122, 220) following the measurement process for the first headlamp (121) on the basis of the position information (212, 221) in order to carry out a measurement process for the second headlamp (122, 220).

7. Device (101) according to Claim 6, wherein the position information (212, 221) indicates a transverse offset in the horizontal and / or vertical direction between the first headlamp (121) and the second headlamp (122, 220).

8. Device (101) according to one of Claims 6 to 7, wherein the device (101) is configured to determine the position information (212, 221) - on the basis of stereoscopic and / or spatial information from the camera image (200); and / or - on the basis of a temporal sequence of camera images (200) which are detected at a corresponding sequence of points in time during the approach maneuver of the vehicle (120).

9. Device (101) according to one of the preceding claims, wherein the device (101) is configured - to determine that the headlamp (121, 122, 220) of the vehicle (120) has reached a target distance from the headlamp setting test device (107) within the scope of the approach process; - in response to the determination, to cause the approach process of the vehicle (120) to be ended; and - after the approach process is ended, to cause the headlamp setting test device (107) to carry out a measurement process for the headlamp (121, 122, 220) of the vehicle (120).

10. Device (101) according to one of the preceding claims, wherein the device (101) is configured - to determine distance information relating to a distance (211) between the headlamp (121, 122, 220) of the vehicle (120) and the headlamp setting test device (107) on the basis of the at least one camera image (200); and - to check, on the basis of the distance information, whether or not the headlamp (121, 122, 220) of the vehicle (120) has reached a target distance from the headlamp setting test device (107) within the scope of the approach process in order to carry out a measurement process for the headlamp (121, 122, 220).

11. Device (101) according to one of the preceding claims, wherein the device (101) is configured - to determine, on the basis of the camera image (200), whether or not the headlamp (121, 122, 220) of the vehicle (120) emits light; and - if it is determined that the headlamp (121, 122, 220) of the vehicle (120) does not emit light, to output an indication to a user of the vehicle (120) and / or of the headlamp setting test device (107), in particular - to cause the user to activate the headlamp (121, 122, 220) of the vehicle (120); and / or - to inform the user that the headlamp (121, 122, 220) of the vehicle (120) is defective.

12. Device (101) according to one of the preceding claims, wherein the device (101) is configured, on the basis of a machine-learned and / or artificial intelligence-based image analysis method, - to detect the headlamp (121, 122, 220) of the vehicle (120) within the camera image (200); and / or - to determine the position of the headlamp (121, 122, 220) of the vehicle (120) within the camera image (200); and / or - to determine a type of the headlamp (121, 122, 220) from a set of different headlamp types; wherein the set of headlamp types in particular comprises: low beam, high beam and / or fog lamps.

13. Device (101) according to one of the preceding claims, wherein - the camera (102) comprises a stereo and / or 3D camera; and / or - the camera (102) is fixedly connected to the headlamp setting test device (107), such that a change in the alignment of the headlamp setting test device (107) leads to a corresponding change in an alignment of the camera (102) relative to the headlamp (121, 122, 220) of the vehicle (120); and / or - the camera (102) and / or the headlamp setting test device (107) are movable within a plane via one or more guide elements (103, 104) in order to change the alignment of the camera (102) and / or of the headlamp setting test device (107) relative to the headlamp (121, 122, 220) of the vehicle (120).

14. Light robot (100) for checking a headlamp (121, 122, 220) of a vehicle (120); wherein the light robot (100) comprises - a headlamp setting test device (107) which is designed to carry out a measurement process for checking the headlamp (121, 122, 220) of the vehicle (120); - a camera (102) which is configured to detect a temporal sequence of camera images (200) relating to the headlamp (121, 122, 220) of the vehicle (120) for a corresponding sequence of points in time; - one or more guide elements (103, 104) which are configured to change an alignment of the headlamp setting test device (107) relative to the headlamp (121, 122, 220) of the vehicle (120); and - a device (101) according to one of the preceding claims.

15. Method (300) for aligning a headlamp setting test device (107) relative to a headlamp (121, 122, 220) of a vehicle (120); wherein the method (300) comprises - detecting (301), during an approach maneuver of the vehicle (120) to the headlamp setting test device (107) and on the basis of a camera (102), a temporal sequence of camera images (200) of the headlamp (121, 122, 220) of the vehicle (120) for a corresponding sequence of points in time; - detecting the headlamp (121, 122, 220) on the basis of a machine-learned and / or artificial intelligence-based image analysis method within a camera image (200) and determining (302) a position of the headlamp (121, 122, 220) within the camera image (200); and - aligning (303) the headlamp setting test device (107) relative to the headlamp (121, 122, 220) of the vehicle (120) depending on the determined position of the headlamp (121, 122, 220) within the camera image (200).