Method for aligning vehicle with ADAS calibration target, and ADAS calibration system

JP2023169124A5Pending Publication Date: 2025-12-09MAHLE INT GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023079810
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2022-05-16
Filing Date
2023-05-15
Publication Date
2025-12-09

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

To provide an accurate and reliable method for aligning a vehicle with an ADAS calibration target.SOLUTION: A method for aligning a vehicle 100 with an ADAS calibration target 4 attached to a support 2 bearing a camera 15 comprises: positioning the vehicle 100 within a service area 200; positioning the support 2 in front of the vehicle 100; moving the vehicle 100 with respect to the support 2 until a front license plate 101 of the vehicle 100 is detected by the camera 15; acquiring images by the camera 15 and identifying a corresponding alphanumeric code; reading out a model of the vehicle 100 univocally associated with the alphanumeric code; reading out a predetermined position associated with the model of the vehicle 100; and providing a driver with commands for driving the vehicle 100 to reache the predetermined position.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a method for aligning a vehicle with an ADAS calibration target and an ADAS calibration system.

Background Art

[0002] As is known, ADAS is an acronym for Advanced Driver Assistance Systems and is an electronic driving assistance system for vehicles that supports drivers for the purpose of enhancing safety and / or driving comfort.

[0003] At the technical level, ADAS systems are based on multiple sensors (cameras, radars, Lidars, etc.), and the multiple sensors can detect various information, and the various information is used as input data for smart algorithms that monitor the degree of autonomy of the vehicle.

[0004] Before the vehicle is put on the market, these sensors need to be directly calibrated by the manufacturer.

[0005] Among some calibration methods currently available, the applicant has recently developed an exclusive calibration method for optical sensors such as front cameras or radars mounted on vehicles. This method is disclosed in EP3588001 and US2019 / 0392610.

[0006] This calibration method can also be used when the vehicle is placed on an inclined plane. In fact, instead of performing a spatial adjustment of the screen with respect to the vehicle, it is sufficient to appropriately deform or adapt the image or video projected on the screen.

[0007] Most calibration devices are in the form of a movable cart, i.e., a support using wheels, in order to be placed in front of the vehicle in the test station.

[0008] Two laser meters are slidably mounted on a horizontal bar of a support to detect the distance between the support and the vehicle's two front or rear wheels. Specifically, the two laser meters are moved along the horizontal bar until they point to optical targets positioned on the vehicle's front or rear wheels using two clamps.

[0009] In fact, most ADAS calibration devices available on the market use systems that detect the vehicle's position relative to a target, and these systems commonly use clamps (mounted on the rim or tire) combined with optical targets, mirrors, or laser pointers.

[0010] Instead of clamps, certain solutions use targets placed on the ground adjacent to the wheels. This solution has the advantage of not requiring anything to be attached to the wheels, but it is less precise than using clamps.

[0011] In either solution, once the vehicle is positioned, it cannot be moved further. Consequently, adjustments to the relative position of the vehicle and target must be performed by moving the support structure. This requires time and attention from a human operator.

[0012] Furthermore, clamps are expensive and bulky items. Applying them always takes several minutes and requires some care to avoid scratching the often expensive rims.

[0013] There is a need for a solution that avoids the use of optical targets placed adjacent to clamps or wheels. [Overview of the project]

[0014] In this regard, the fundamental technical problem of the present invention is to propose a method for aligning a vehicle to an ADAS calibration target and an ADAS calibration system that overcome the shortcomings of the prior art described above.

[0015] In particular, an object of the present invention is to provide a method and ADAS calibration system for aligning a vehicle to an ADAS calibration target that is more accurate and reliable than conventional solutions.

[0016] Another objective of the present invention is to provide an ADAS calibration system that is more compact, easier to use, and faster to operate than known solutions.

[0017] The technical challenges and identified objectives are substantially achieved by a method for aligning a vehicle to an ADAS calibration target, the vehicle having a front license plate integrally mounted thereto, the front license plate having a substantially rectangular surface with an alphanumeric code uniquely associated with the vehicle, and the method is The steps include providing a support for the camera array and calibration target, The steps include positioning the vehicle within the service area, The steps include positioning the support near the service area so that the calibration target and camera array are in front of the vehicle, The steps include moving the vehicle relative to the support until the vehicle's front license plate is detected by the camera array, The steps include: acquiring an image using a camera array, and in response to the detection of the front license plate within that image, identifying the corresponding alphanumeric code; The steps include querying a first database to retrieve the vehicle model uniquely associated with an alphanumeric code, The steps include querying a second database to retrieve a predetermined location associated with the model of the vehicle being calibrated, The process includes the step of giving the driver instructions to drive the vehicle so that it reaches the predetermined position.

[0018] According to one aspect of the present invention, the camera array is attached under the calibration target.

[0019] According to one embodiment of the present invention, the camera array comprises a single camera.

[0020] According to another embodiment of the present invention, the camera array comprises a pair of stereo cameras.

[0021] According to one aspect of the present invention, the detection of the front number plate is performed by color analysis or contour analysis.

[0022] The technical problems shown and the specified objectives are substantially achieved by an ADAS calibration system that calibrates at least one ADAS sensor of a vehicle to which a front number plate is integrally attached, the front number plate having a substantially rectangular surface with an alphanumeric code uniquely associated with the vehicle, the ADAS calibration system comprising: a movable support; a calibration device comprising a target attached to the support; a camera array attached to the support and configured to acquire an image of the front number plate of the vehicle; a control unit, [[ID=!]]which receives an image acquired from the camera array, recognizes the front number plate therein, identifies the corresponding alphanumeric code in the acquired image in response to the recognition of the front number plate in the acquired image, queries a first database to read the model of the vehicle uniquely associated with the alphanumeric code, queries a second database to read a predetermined position associated with the model of the vehicle to be calibrated, and a control unit configured to give a command to the driver to drive the vehicle so that the vehicle reaches the predetermined position.

[0023] It should be noted that there seems to be an incorrect tag in the original text, where [[ID=!]] should be . The above translation is based on the corrected content for a more accurate translation. According to one embodiment of the present invention, the target consists of a monitor configured to project an image or video.

[0024] According to one aspect of the present invention, the camera array is mounted under the monitor.

[0025] According to one embodiment of the present invention, the camera array comprises a single camera.

[0026] In particular, the camera is arranged such that its center is located at the center of the lower side of the monitor.

[0027] According to another embodiment of the present invention, the camera array comprises a pair of stereo cameras.

[0028] In particular, the pair of stereo cameras is arranged symmetrically with respect to the vertical symmetry axis of the monitor.

Brief Description of the Drawings

[0029] Further features and advantages of the present invention will become more apparent from the following non - limiting description of preferred but non - exclusive embodiments of a method for aligning a vehicle with an ADAS calibration target and an ADAS calibration system, as shown in the accompanying drawings. [Figure 1] One embodiment of an ADAS calibration system according to the present invention is schematically shown. [Figure 2] Another embodiment of an ADAS calibration system according to the present invention is schematically shown. [Figure 3] A plan view shows the various positions taken by the front number plate corresponding to the various positions taken by the vehicle during the method of aligning the vehicle with an ADAS calibration target according to the present invention. [Figure 4] Detection of the front number plate by the method of aligning a vehicle with an ADAS calibration target according to the present invention is shown.

Modes for Carrying Out the Invention

[0030] Referring to the drawing, number 1 indicates an ADAS calibration system for calibrating one or more ADAS sensors mounted on vehicle 100, where vehicle 100 is an automobile, particularly a passenger car, bus, heavy truck, road tractor, tractor trailer, articulated truck, agricultural machinery, work vehicle, or self-propelled vehicle.

[0031] For example, the ADAS sensor may include a CMOS or CCD type sensor of a television camera mounted on the vehicle 100, and / or a front radar.

[0032] Vehicle 100 has a front license plate 101 (hereinafter also referred to as "plate") that is integrally attached to the vehicle 100.

[0033] The front license plate 101 is mounted on the front center of the vehicle 100, preferably on the bumper.

[0034] As is known, the front license plate 101 has a substantially rectangular surface with an alphanumeric code 102.

[0035] The dimensions of a rectangular surface are usually determined by specific laws and regulations, depending on the country.

[0036] For example, in Italy, the rectangular surface of a front license plate with the number 101 measures 360mm x 110mm.

[0037] In particular, the front license plate 101 has an alphanumeric code 102 that is uniquely associated with the vehicle 100.

[0038] In fact, the alphanumeric codes of vehicles are stored in a dedicated national database along with data for other vehicles.

[0039] Assuming that the vehicle 100 has a longitudinal axis of symmetry AA, the front license plate 101 is theoretically mounted perpendicular to the longitudinal axis AA.

[0040] The ADAS calibration system 1 is positioned in front of a vehicle 100 in a service area 200 and includes a support 2, preferably a movable support 2, i.e., a support 2 with wheels 12.

[0041] The ADAS calibration system 1 includes a calibration device 3 attached to a support 2.

[0042] The calibration device 3 includes a target 4, which is preferably a monitor configured to project an image or video.

[0043] In particular, the monitor 4 has a rectangular display surface. The monitor 4 is attached to the support 2 such that the first pair of sides of its rectangular display surface are parallel to the ground on which the support 2 is placed, and the second pair of sides of its rectangular display surface are perpendicular to the ground.

[0044] The first pair of sides is preferably longer than the second pair of sides.

[0045] The ADAS calibration system 1 further comprises a camera array 5 integrally attached to a support 2.

[0046] For example, camera array 5 is mounted at the bottom or beneath monitor 4.

[0047] According to one embodiment of the present invention, the camera array 5 comprises a single camera 15, for example, a monochrome camera.

[0048] The center of camera 15 is considered to be the origin of the spatial coordinate system (x, y, z).

[0049] In particular, the display surface of monitor 4 lies on a plane defined by the coordinates x and y of the spatial coordinate system (x, y, z), or on a plane parallel to plane x and y.

[0050] The remaining axis is axis z, which coincides with the optical axis of camera 15.

[0051] In particular, the camera 15 is positioned such that its origin is located in the center of the bottom edge of the display surface of the monitor 4, and the bottom edge is aligned with axis X.

[0052] According to another embodiment of the present invention, the camera array 5 comprises a pair of stereo cameras 5a, 5b for establishing binocular vision.

[0053] In this embodiment, the origin of the spatial coordinate system (x, y, z) coincides with the center of the bottom edge of the display surface of monitor 4.

[0054] The two cameras 5a and 5b are positioned symmetrically with respect to the origin along the lower edge of the display surface of monitor 4.

[0055] In both embodiments, the camera array 5 is configured to acquire an image of the front center of the vehicle 100 located in the service area 200.

[0056] In particular, camera array 5 is configured to acquire an image of the front license plate 101.

[0057] The ADAS calibration system 1 further comprises a control unit 6 configured to receive images from the camera array 5 and recognize the front license plate 101 within them.

[0058] In particular, in response to the recognition of the front license plate 101 in the image acquired by the camera array 5, the control unit 6 is configured to identify the corresponding alphanumeric code 102 engraved on the front license plate 101.

[0059] Furthermore, the control unit 6 is further configured to control the projection of images or videos onto the monitor 4 in accordance with known methods, such as those disclosed in the present applicant's documents EP3588001 and US2019 / 0392610.

[0060] Preferably, the control unit 6 is mounted on the support 2, for example, behind the monitor 4.

[0061] The method for aligning a vehicle to an ADAS calibration target according to the present invention is described below.

[0062] Vehicle 100 will initially be deployed within service area 200.

[0063] The ADAS calibration system 1 is positioned near the service area 200 and moves the support 2 so that, in particular, the target (e.g., monitor 4) and camera array 5 are positioned in front of the vehicle 100.

[0064] For simplicity, the method will be further described with reference to an embodiment comprising a single camera 15.

[0065] Camera 15 begins acquiring an image of the front center of vehicle 100 located in service area 200.

[0066] The control unit 6 processes the image acquired by the camera 15 and displays the image on the monitor 4 by highlighting the region of interest as a rectangle or square (for example, a green square) on the monitor 4.

[0067] In particular, the region of interest is the area where the camera should appear when the vehicle 100 approaches the support 2.

[0068] This region of interest is appropriate because it does not require processing very large images (and therefore more computational power), and at the same time, it does not consider optical artifacts that are far removed from the region of interest itself and, while not of physical significance, could interfere with the algorithm.

[0069] The human operator seated in the driver's seat of vehicle 100 is guided by messages displayed on monitor 4 to slowly approach support 2, taking care to keep the steering wheel stable so that vehicle 100 moves in a straight line (regardless of direction) as it approaches support 2. For simplicity, the approach movement is shown in Figure 3, referring to the movement of the front license plate 101.

[0070] As soon as the front license plate 101 enters the field of view of the camera 15, the front license plate 101 is detected by the control unit 6, the corresponding image is decoded, and this makes it possible to identify the alphanumeric code 102.

[0071] The detection of the front license plate 101 may be performed according to a known algorithm, for example, based on color analysis (for example, Italian license plates consist of a white rectangle sandwiched between two small blue rectangles, each having a specific shape ratio) or contour analysis (the so-called "Canny" family of algorithms).

[0072] Next, the method includes the step of querying a first database to retrieve the model of the vehicle 100 associated with the alphanumeric code 102.

[0073] In fact, as already mentioned above, the alphanumeric code 102 is uniquely associated with vehicle 100, so the model vehicle can be retrieved from a first database, which may be a national database of automobiles.

[0074] Once the vehicle model is known, a second database is queried to retrieve predetermined locations associated with the said model in order to perform proper calibration of the ADAS sensors mounted on the vehicle 100.

[0075] The predetermined position is defined relatively as the distance between the center of the front license plate 101 and the center of the camera array 5.

[0076] In an embodiment with a single camera 15, the center is the center of the camera 15 (i.e., the origin of the spatial coordinate system (x, y, z)).

[0077] Preferably, the second database is an internal database that can be stored in memory configured to communicate with the control unit 6. The memory may be attached to the support 2 or located separately from the support 2.

[0078] Next, the method proceeds to the step of instructing a human operator to drive the vehicle 100 to reach a predetermined location.

[0079] When vehicle 100 arrives at a predetermined location, calibration of the ADAS sensors mounted on vehicle 100 can be initiated. This can be done according to known techniques.

[0080] In particular, once vehicle 100 reaches a predetermined position, it remains stationary.

[0081] At the outset of this disclosure, it was assumed that the front license plate 101 was substantially perpendicular to the longitudinal axis AA of the vehicle 100. However, deviations from orthogonality may exist, for example, caused by an impact to the front license plate 101.

[0082] Therefore, the operator seated in the driver's seat of vehicle 100 is required to approach the support 2 slowly while taking care to keep the steering wheel in a stable position.

[0083] Therefore, the center of the front license plate 101 will follow a trajectory T that gives the direction of the longitudinal axis AA of the vehicle 100, appropriately interpolated (interpolated as Ti in Figure 3).

[0084] This information regarding the trajectory T and the apparent dimensions of the front number plate 101 are used to determine four basic parameters that should be set in the ADAS calibration system to perform the calibration of the vehicle 100.

[0085] This is done according to known calculation methods, as will be explained below for completeness.

[0086] By knowing the parameters of camera 15 (especially the pixel size in microns and its focal length) and obtaining the apparent dimensions of the front license plate 101 (height × width, which may be corrected by a slight rotation of the plate itself) from the image processed as described above, it is possible to determine the distance from camera 15 to the front license plate 101 using the following formula.

[0087] (1) D = f × N × p / D0 Here, D is the distance from the camera's focal point (origin) to the front license plate "101".

[0088] N is the number of pixels (height or width) in the acquired image of front license plate 101.

[0089] P represents the micron size of the camera pixel (usually on the order of 1-3 microns). D0 is the actual size (height or width) of the front license plate 101 in accordance with national regulations.

[0090] Equation (1) is valid only if the front license plate 101 is perpendicular to axis z. Otherwise, there is uncertainty given by the inclination of the front license plate 101 with respect to axis z, which can be eliminated in a mathematically rigorous manner by using an embodiment with a pair of stereo cameras 5a, 5b.

[0091] However, if the tilt angle is small, this fact can be ignored, and the front license plate 101 can be considered perpendicular to the ADAS calibration system anyway.

[0092] For example, if monitor 4 and vehicle 100 (and therefore plate 101) are tilted relative to each other by 5°, the apparent width of the plate as seen from the camera will decrease by a coefficient cos(5°), or 0.4%.

[0093] According to equation (1), this uncertainty is reflected in the distance calculation. When the typical distance from a structure is 1m, an uncertainty of 4mm arises in the distance calculation, which is within an acceptable range for most automobile manufacturers.

[0094] As an alternative, binocular vision can be used, as mentioned above.

[0095] In an embodiment equipped with a pair of stereo cameras 5a and 5b, the image of the front license plate 101 is used as an optical target without any dimensional assumptions. This embodiment makes it possible to obtain a more accurate determination even if there is a vertical displacement in the position of the front license plate 101.

[0096] The other steps of this method are substantially the same as those described for a single camera.

[0097] The features of the method for aligning a vehicle to an ADAS calibration target and the ADAS calibration system according to the present invention, as well as their advantages, will become apparent from the above description.

[0098] In particular, vehicle alignment with respect to the ADAS calibration system is performed by a camera array integrally mounted on the ADAS support, and the cameras are configured to identify alphanumeric codes on the license plate.

[0099] Therefore, this solution is more compact and easier for human operators to use. In fact, the camera array may consist of a single camera or a pair of stereo cameras mounted on an ADAS support.

[0100] Once the vehicle model is identified from the detected alphanumeric code on the front license plate, the alignment process is simple and quick, as the human operator simply drives the vehicle according to the position commands from the ADAS system.

[0101] This solution is even more convenient at quick service centers such as windshield replacement centers.

[0102] Therefore, optical targets or clamps adjacent to the wheels are no longer needed. The vehicle moves according to precise position commands calculated by the ADAS system, but the support structure no longer moves.

[0103] Furthermore, both accuracy and reliability will improve. [Prior art documents] [Patent Documents]

[0104] [Patent Document 1] EP3588001 [Patent Document 2] US2019 / 0392610

Claims

1. A method for aligning a vehicle (100) to an ADAS calibration target (4), comprising: The vehicle (100) has a front license plate (101) integrally attached to the vehicle (100), the front license plate (101) having a substantially rectangular surface bearing an alphanumeric code (102) uniquely associated with the vehicle (100), and the method comprises: providing a support (2) for supporting a camera array (5) and said calibration target (4); positioning the vehicle (100) within a service area (200); positioning the support (2) near the service area (200) so that the calibration target (4) and the camera array (5) are in front of the vehicle (100); moving the vehicle (100) relative to the support (2) until the front license plate (101) of the vehicle (101) is detected by the camera array (5); acquiring images by said camera array (5) and identifying said corresponding alphanumeric code (102) therein in response to detecting said front license plate (101); querying a first database to retrieve a model of said vehicle (100) uniquely associated with said alphanumeric code (102); querying a second database to retrieve predetermined locations associated with the model of the vehicle (100) to be calibrated; and providing instructions to a driver to drive the vehicle (100) to reach the predetermined location.

2. The method of claim 1 , wherein the camera array (5) is mounted below the calibration target (4).

3. The method according to claim 1 or 2, wherein the camera array (5) comprises a single camera (15).

4. The method according to claim 1 or 2, wherein the camera arrangement (5) comprises a pair of stereo cameras (5a, 5b).

5. 3. The method according to claim 1 or 2, wherein the detection of the front license plate (101) is performed by color analysis or contour analysis.

6. An ADAS calibration system (1) for calibrating at least one ADAS sensor of a vehicle (100) having a front license plate (101) integrally attached thereto, comprising: The front license plate (101) has a substantially rectangular surface having an alphanumeric code (102) uniquely associated with the vehicle (100), and the ADAS calibration system (1) comprises: A movable support (2), a calibration device (3) comprising a target (4) attached to the support (2); a camera array (5) attached to the support (2) and configured to capture images of the front license plate (101) of the vehicle (100); A control unit (6), receiving images acquired from said camera array (5) and recognizing a front license plate (101) therein; responsive to recognizing a front license plate (101) in the captured image, identifying a corresponding alphanumeric code (102) in the captured image; querying a first database to retrieve a model of said vehicle (100) uniquely associated with said alphanumeric code (102); querying a second database to retrieve predetermined locations associated with the model of the vehicle (100) to be calibrated; and a control unit (6) configured to give a driver a command to drive the vehicle (100) so that the vehicle (100) reaches the predetermined position.

7. 7. The ADAS calibration system (1) of claim 6, wherein the target (4) comprises a monitor configured to project an image or video.

8. 8. The ADAS calibration system (1) of claim 7, wherein the camera array (5) is mounted below the monitor (4).

9. 9. The ADAS calibration system (1) of claim 8, wherein the camera array (5) comprises a single camera (15).

10. 10. The ADAS calibration system (1) according to claim 9, wherein the camera (15) is positioned so that its center is located at the center of the lower edge of the monitor.

11. 9. The ADAS calibration system (1) according to claim 8, wherein the camera array (5) comprises a pair of stereo cameras (5a, 5b).

12. 12. The ADAS calibration system (1) according to claim 11, wherein the pair of stereo cameras (5a, 5b) are arranged symmetrically with respect to a vertical axis of symmetry of the monitor (4).