Method for calibrating the positioning and / or alignment of an optical detection system mounted on an agricultural vehicle and corresponding arrangement

Abstract

A method and apparatus for calibrating the positioning and / or alignment of an optical detection system (62) attached to an agricultural vehicle (10) comprise the following steps or means for carrying them out automatically: successive positioning of an unmanned robot (66) at several predetermined positions (AE, JL) relative to the vehicle; Detection of the robot (66) positioned in this manner and / or of a marker (86) attached to the robot (66) by means of the detection system (62) which generates an image signal; Processing the image signal of the detection system (62), comprising recognizing the position of the robot (66) and / or the marker (86) in the image; and calculating the positioning and / or orientation of the detection system (62) relative to a reference position and / or a reference direction of the vehicle (10) based on the respective positions (AE, JL) of the robot (66) and the associated position of the robot (66) and / or the marker (86) detected in the image.

Description

[0001] The invention relates to a method for calibrating the positioning and / or alignment of an optical detection system mounted on an agricultural vehicle and a corresponding arrangement. State of the art

[0002] In agriculture, optical detection systems are used to record specific parameters of a field and, based on these, to control a vehicle and / or attached implements. Examples include steering systems that optically detect the field in front of the vehicle and, based on this, generate steering signals to steer the vehicle and / or implement, for example, along crop rows, a lane, a tramline, a swath, a crop edge or a processing edge. In addition, such optical detection systems can be used to record properties of plants in a field and, based on these, to control specific working parameters of an implement. For example, to specify the position of the boom of a field sprayer or a reel of a harvesting header based on the crop height, or to control the forward speed of the vehicle based on the crop density.

[0003] These acquisition systems use cameras (mono or stereo versions) or scanning laser sensors (lidar) and image processing systems to derive the required field parameters from the image signal. In most cases, not only qualitative but also quantitative evaluations are performed to generate a control signal for the vehicle and / or implement. An example would be a lateral offset between the camera and a row of plants, on which a steering signal is to be based. When evaluating the images, it is therefore important to recognize the relationship between the position of a sensor pixel and the spatial coordinates represented by the pixel.

[0004] To determine this relationship or to establish a desired relationship, the optical detection systems must be calibrated before use. A distinction is made between determining intrinsic parameters and extrinsic parameters. The intrinsic parameters represent the relationship between the position of a pixel and the corresponding two-dimensional angle of incidence, while the extrinsic parameters define the arrangement of the detection system in space or relative to the vehicle (see C. Fries, Model-based vehicle detection of a driver assistance system for autonomous following in a convoy, dissertation, Munich 2019).

[0005] The intrinsic parameters are usually provided by the camera manufacturer or can be determined by detecting suitable markers that show certain patterns, such as checkerboards or so-called ArUco patterns, and which are to be positioned in specific positions relative to the camera (Y. Ren et al., Camera Calibration with Pose Guidance, https: / / arxiv.orq / abs / 2102.10202; for the calibration of the intrinsic parameters of a drone camera by taking images of a marker from different positions, see M. Pérez et al., Digital Camera Calibration Using Images Taken From an Unmanned Aerial Vehicle, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-1 / C22, 2011, ISPRS Zurich 2011 Workshop, 14-16 September 2011, Zurich, Switzerland, pages 167 to 171).

[0006] In the simplest case, a mechanical alignment of the camera relative to the vehicle, e.g. in the case of a laterally adjustable hoe, can be achieved in such a way that the camera image is displayed on a user interface and the operator adjusts the camera relative to its attachment to the vehicle in such a way that a marker contained in the image is shown at a desired location in the image and the camera is then fixed (M. Heinold et al., Tractor-Integrated Active Device Control for High-Speed ​​Hoes, in A. Ruckelshausen et al. (eds.): Digital Marketplaces and Platforms, Lecture Notes in Informatics (LNI), Gesellschaft für Informatik, Bonn 2018, pages 95 to 98). This ensures that the camera is mounted in a suitable position so that an image processing-based control system can then automatically control the hoe in such a way that the marker remains at the desired location in the image.This does not involve measuring the extrinsic parameters of the camera, but simply aligning the camera to a suitable position.

[0007] One possible, but error-prone, approach to calibrating the camera's extrinsic parameters is for an operator to enter certain data regarding the camera's position and orientation into a controller (see, for example, the VSN operating manual from Raven, printing note 016-2020-001DE Rev. C 4 / 2021 E36973).

[0008] Another approach to calibrating the extrinsic parameters involves positioning the vehicle on a flat surface and placing a number of markers at defined positions on the ground in front of the vehicle. By detecting the markers in the image, the camera's inclination relative to the forward direction and its lateral offset relative to the vehicle's longitudinal center plane can be determined (EP 4 085 740 A1).

[0009] Y. Zhang et al., "Camera Calibration for Long-Distance Photogrammetry Using Unmanned Aerial Vehicles," Hindawi Journal of Sensors Volume 2022, Article ID 8573315, https: / / doi.orq / 10.1155 / 2022 / 8573315, describe a calibration of a stereo camera for remote sensing. A drone is equipped with a navigation system and, based on this, is moved to specific positions in space. The drone is captured in the image, and the corresponding pixels are used to determine the differences between the extrinsic parameters of the left and right cameras. This ultimately determines the position of a detected object based on the different images from both cameras. Thus, the position and orientation of the stereo camera relative to a reference system, such as a vehicle, are not determined; instead, the relative orientation of the two cameras is determined based on the differences in both images at specific three-dimensional positions of a detected object. Task

[0010] Calibrating the position and orientation of optical detection systems on agricultural vehicles is relatively time-consuming and must be repeated whenever the camera's position or orientation changes. This may be necessary, for example, after changing between different-sized tires, as this changes the distance between the ground and the camera, or after changing or adjusting the camera.

[0011] The object underlying the invention is to provide a method for calibrating an optical detection system of an agricultural vehicle and a corresponding arrangement which does not have the said disadvantages or has them to a reduced extent. Solution

[0012] This object is achieved according to the invention by the teaching of patent claims 1 and 8, wherein the further patent claims list features which further develop the solution in an advantageous manner.

[0013] A method for calibrating the positioning and / or alignment of an optical detection system mounted on an agricultural vehicle comprises the following steps or means for automatically performing them: successively positioning an unmanned robot at several predetermined positions relative to the vehicle; detecting the robot thus positioned and / or a marker attached to the robot by means of the detection system which generates an image signal; Processing the image signal of the detection system, comprising detecting the position of the robot and / or the marker in the image; and calculating the positioning and / or orientation of the detection system relative to a reference position and / or a reference direction of the vehicle based on the respective positions of the robot and the associated position of the robot and / or the marker detected in the image.

[0014] In other words, it is proposed to move one (or more) robots to one or more known positions relative to the vehicle, to detect them or a marker attached to them in the image of an optical detection system, which does not necessarily have to be sensitive only in the visible wavelength range, but can also be sensitive in other wavelength ranges such as UV, infrared, microwaves, and radar waves and can be passive (e.g., as a black-and-white, color, or multispectral camera) or active (e.g., as a lidar sensor), and to calibrate the optical detection system with regard to position and / or orientation based on this. This eliminates the previous procedure of positioning markers at specific locations and allows the calibration to be carried out, for example, in a field.

[0015] The vehicle and the robot can each be equipped with a navigation unit for detecting the spatial position, which makes it possible to move the robot to the desired position relative to the vehicle.

[0016] The robot can be an aircraft, which allows for suitable positioning in three-dimensional space. However, the use of a ground vehicle as a robot would also be conceivable. If necessary, this could include a height-adjustable marker, which also allows for positioning in a desired space. The ground vehicle makes it possible to assume certain positions that are difficult or impossible to reach by an aircraft, particularly under the vehicle for calibrating an optical detection system located there.

[0017] Preferably, the robot is located at the predetermined positions in a plane parallel to the vehicle's support plane. This facilitates the calculation of the position and / or orientation of the optical detection system relative to the vehicle. However, it would also be conceivable to position the robot at different heights relative to the support plane and to consider these heights in the evaluation. This can be particularly useful if the robot is designed as a land vehicle and is therefore bound to the local topography.

[0018] The robot can determine the vehicle's orientation (in one, two, or three directions) by approaching different positions on the vehicle and detecting the corresponding position of its navigation unit, thus eliminating the need for angle measurements by the vehicle's navigation unit. Alternatively, or in addition, the vehicle's navigation unit can detect its orientation.

[0019] The targeted positions, at least some of them, can be referenced to a reference system of the vehicle. The robot is therefore moved to positions that are a defined distance from the vehicle. Alternatively or additionally, the targeted positions, at least some of them, can also be referenced to a reference system of the detection system. The robot is therefore moved to a position where it appears at a predetermined location in the image of the detection system, for example, on its optical axis.

[0020] The positioning and / or orientation of the detection system calculated in the manner described can be used to generate a steering signal for the vehicle and / or a vehicle attachment based on the detection system signal and / or to control an actuator of the vehicle and / or attachment. Reference is made to the examples in the first paragraph of the discussion on the prior art. Example

[0021] The drawings illustrate an embodiment of the invention, described in more detail below. They show: Fig. 1 a side view of an agricultural vehicle with an optical detection system for guiding the vehicle across a field, Fig. 2 a top view of the vehicle working in the field, Fig. 3 a side view of the vehicle during calibration of the optical detection system with a robot at different positions, Fig. 4 the image captured by the acquisition system in the situation according to Fig. 3, Fig. 5 a top view of the vehicle when measuring the roll and yaw angle, Fig. 6 two images captured by the acquisition system in the situation after Fig. 5, and Fig. 7 a flowchart for detecting the positioning and alignment of the optical detection system.

[0022] The Fig. Figure 1 shows a side view of a self-propelled agricultural vehicle 10 in the form of an agricultural tractor and an implement 12 in the form of a field sprayer mounted on a three-point linkage 14 of the vehicle 10. The vehicle 10 is built on a supporting frame 16, which is supported on steerable front wheels 18 and drivable rear wheels 20 and carries a cabin 22 in which an operator's workplace 24 is located. The operator's workplace 24 includes a steering wheel 26, a seat 28, pedals (not shown), and an input and output unit 30. In the following, directional references such as front and rear or sideways refer to the forward direction of the vehicle 10, which in the Fig. 1 runs to the left.

[0023] The vehicle 10 comprises an arrangement for automatically guiding the vehicle 10 across a field, which arrangement comprises an optical detection system 62 with a camera 32 designed as a mono or stereo camera (and / or a scanning, active sensor, e.g., a lidar system), a further camera 64, an image processing system 36 connected to the image sensor of the camera 32 for signal transmission, an image processing system 36 connected to the image sensor of the camera 64 for signal transmission, a steering signal calculation unit 46, and a steering control 34. The camera 32 is attached to the roof of the cabin 22 and looks forward and downward, thus detecting the front area of ​​the vehicle 10 and the field 58 in front of it (cf. Fig. 2 to 4) and the plants 60 standing on it. The further camera 64 is attached below the frame 16 and also looks forward and downward.

[0024] The image processing systems 36, the steering signal calculation unit 46 and the steering control 34 are interconnected via a bus system 50, which can operate, for example, according to the protocol according to ISO 11783 or a successor thereto.

[0025] The steering signal calculation unit 46 is spatially integrated into an input and output unit 30, which is provided with a user interface in the form of a display unit 54 and an input device 56. The input and output unit 30 can fulfill the function of a virtual display according to ISO 11783.

[0026] The steering control 34 is connected to a valve 38 for controlling a steering actuator 40 and to the steering angle sensor 44 for detecting the current steering angle of the front wheels 18.

[0027] In the Fig. Figure 2 shows a plan view of the vehicle 10 working in a field 58 containing rows of crops 60. The steering control 34 is controlled by the steering signal calculation unit 46 based on the signals from the optical detection system 62 (cameras 32 and 64) such that the front wheels 18 and the rear wheels 20 travel between the rows of crops 60 without driving over the crops 60. The optical detection system 62 can detect the plants 60 (cf. for example US 6 445 983 B1, EP 1 266 553 A2, WO 2008 / 150418 A1, WO 2018 / 067473 A1, Wan, X., Liu, G. (2008). Automatic Navigation System with Multiple Sensors. In: Li, D. (eds) Computer and Computing Technologies in Agriculture, Volume II. CCTA 2007. The International Federation for Information Processing, vol 259. Springer, Boston, Ma.), and / or detect driving lanes or tracks between the plants 60, cf. DE 10 2016 209 437 A1.The implement 12 could also be any other implement 12, e.g., a hoe used to remove unwanted weeds between the rows of plants 60. In another embodiment, the implement 12 could be a soil cultivation implement, such as a plough or harrow, and the optical detection system 62 could detect the boundary between the uncultivated and cultivated field 58. In yet another embodiment, the vehicle 10 could be a harvester with a harvesting header, or it could move a harvesting implement, e.g., a mower, across the field, and the optical detection system 62 could detect a crop edge of the unharvested crop.

[0028] Furthermore, the vehicle 10 is equipped with a navigation unit 48 for receiving signals from a satellite-based navigation system (GNSS, such as GPS, Galileo, etc.), which is mounted on the roof of the cabin 22. The navigation unit 48 can be equipped with an inertial navigation system to improve the accuracy of the position signals, particularly under unfavorable reception conditions. Furthermore, it can be connected to a receiver for receiving correction signals from a corresponding system (DGPS, RTK) to improve the precision of the navigation system signals, and / or to odometric sensors that detect the steering direction and movement of the front wheels 18, i.e., a steering angle sensor 44 and a speed sensor for detecting the rotation of the front wheels. The odometrically detected signals are typically merged with those of the inertial navigation system and the satellite-based, possibly corrected signals, e.g.through a Kalman filter. The navigation unit 48 outputs a signal at regular intervals regarding its three-dimensional position and optionally its speed and / or orientation (pitch, yaw, and roll angles) relative to a reference direction, which may coincide with the longitudinal axis of the vehicle 10.

[0029] The optical detection system 62 with the two cameras 32 and 64 and the image processing systems 36 thus supplies the steering signal calculation unit 46 with a signal that contains information at least regarding a lateral offset of the vehicle 10 relative to a target path and regarding a deviation between the orientation of the vehicle and the target path. Reference is made to the literature mentioned in the previous paragraph. The signals from the navigation unit 48 can be merged with the signals from the image processing systems 36 by the steering signal calculation unit 46 in a manner known per se to improve steering accuracy.

[0030] It is evident that the optical detection system 62 can only output a meaningful signal to the steering signal calculation unit 46 if a relationship is known between the position of a feature contained in the image signal from the cameras 32 and 64, such as a plant 60, and its position (distance) and direction (angle) relative to a reference point and a reference direction. Therefore, the image processing systems 36 each have, on the one hand, a relationship between the position of a pixel of the sensor of the cameras 32, 64 and the associated angle of incidence of the light (so-called intrinsic calibration, which can be supplied by the manufacturer of the camera 32, 64 or provided by measurements and stored as a data set in the image processing system 36), and, on the other hand, data regarding the positioning and orientation of the cameras 32, 64 on the vehicle 10 (so-called extrinsic calibration).Based on both calibrations and the image signal from the cameras 32, 64, the image processing systems 36 generate the signal for the steering signal calculation unit 46.

[0031] The positioning and orientation of the cameras 32, 64 are detected in this case by a robot 66. The robot 66 is designed here as an aircraft (drone), but could also be a land vehicle and equipped with wheels or crawler tracks. The robot 66 comprises a supporting frame 68 to which a number of propellers 70 are attached, driven by electric motors 72. Also attached to the supporting frame 68 are a power source (battery) 78, a controller 80, a transmitting and receiving unit 76, and a navigation unit 74, which corresponds in structure and function to the navigation unit 48 of the vehicle 10. The transmitting and receiving unit 76 is connected to a similar transmitting and receiving unit 82 of the vehicle 10, which in turn is connected to the steering signal calculation unit 46 via the bus system 50 or a separate line.The robot 66 could also be connected to the vehicle 10 via a cable, cf. DE 10 2014 201 203 A1. In addition to the calibration of the optical detection system 62 presented here, the robot 66 could also be used for other tasks, e.g., for exploring the field in the sense of locating locations with plants 60 that are to be sprayed with spray liquid by the implement 12 for pest control or fertilization, using a camera 84. When not in use, the robot 66 can dock with its landing gear 88 at a suitable location, e.g., on the roof of the cabin 22 of the vehicle 10 (see DE 10 2015 206 844 A1). To improve its recognizability, the robot 66 can be provided with a marking 86 on which any pattern, e.g. a checkerboard pattern or an ArUco pattern, is applied, and / or at least partially with a clearly recognizable color that distinguishes it from a usual field environment (e.g.a neon color such as orange, yellow or bright green) or with a pattern.

[0032] The Fig. 3 to 7 show a possible procedure for detecting the positioning and alignment of the cameras 32, 64 of the optical detection system 62. This is done in such a way that the operator makes a corresponding input into the input device 56 that a new or initial calibration of the positioning and alignment of the cameras 32, 64 of the optical detection system 62 is desired. A processor of the input and output unit 30 or the steering control unit 46 and the controller 80 of the robot 66 then proceed according to the flow chart of the Fig. 7. In this case, the vehicle 10 is stationary or is brought to a stop, although it would also be conceivable to let it travel at a constant speed.

[0033] After starting in step 100, the robot 66 flies to a first, predetermined position (step 102) and a first image is taken by means of the cameras 32, 64, in which the robot 66 and, if present, the marker 86 are present. This step is repeated at other positions until all positions have been flown over (step 104) and then, in step 106, the images are evaluated and, based on the images, the positioning and orientation of the cameras 32, 64 of the optical detection system 62 are determined and the process ends in step 108. Step 106 can be performed by the image processing systems 36, to which corresponding information is sent from the processor of the input and output unit 30 or the steering control unit 46 that steps 102 or 106 are currently being carried out.In another embodiment, the image processing can be performed directly by the processor of the input and output unit 30 or the steering control unit 46, in particular when the image processing systems 36 are integrated into the input and output unit 30.

[0034] Both Fig. 3 and Fig. 4, in steps 102, positions A to E in front of vehicle 10 are approached in any order, and the robot is detected there by camera 32, while position F, together with position E, serves to detect the pitch angle (i.e., the angle of inclination relative to the longitudinal axis of vehicle 10) of camera 32. To calibrate camera 64, robot 66 can assume the positions shown below vehicle 10. For this purpose, instead of a flight-capable robot 66, another robot designed as a land vehicle could also be used.

[0035] For the described procedure, the controller 80 of the robot 66 receives data via the transmitting and receiving units 76, 82 regarding the current, three-dimensional position of the navigation device 48 of the vehicle 10 and regarding the orientation of the navigation device 48 or the vehicle 10 in space (roll, pitch, and yaw angles, generally measured relative to the longitudinal center axis of the vehicle 10) and, based thereon, calculates suitable positions and orientations for the robot 66 for the steps 102. In particular, the robot 66 can be positioned in the (contact) plane on which the wheels 18, 20 of the vehicle 10 are on the ground. Depending on the ground profile, the robot 66 can land or hover above the ground.If, during the approach to the positions, it becomes apparent that the ground profile in front of the vehicle 10 rises relative to the vehicle 10 at at least one of the approach points, so that the robot 66 cannot be positioned in the said plane on which the wheels 18, 20 of the vehicle 10 rest on the ground, it can be positioned above the ground at all approach points of steps 102, so that steps 102 are repeated so that all measurements of the position of the robot 66 are taken in a common plane. The three-dimensional positions of the robot 66 are reported back to the processor of the input and output unit 30 or the steering control unit 46 via the transmitting and receiving units 76, 82.Calibrating the positioning and alignment of the cameras 32, 64 of the optical detection system 62 in a plane above the plane on which the wheels 18, 20 of the vehicle 10 rest on the ground may also be useful if the robot 66, in the middle position (E), would damage one or more of the plants 60. The other positions A to D can, if necessary, be selected so that they are located between adjacent rows of plants 60.

[0036] In step 106, the images acquired in steps 102, together with the associated three-dimensional positions of the robot 66, are evaluated by the processor of the input and output unit 30 or the steering control unit 46. In this case, the robot 66 itself can be recognized in the image or, if present, the marking 86 can be detected. The pitch angle of the camera 32 can be estimated based on the positions E (in which the robot 66 is located on the optical axis of the camera 32, i.e., is guided there based on the image signal from the camera 32) and the position F. Using trigonometric calculations known per se, the position of the camera 32 relative to any reference point of the vehicle 10 can also be calculated, as can the yaw and roll angles of the camera 32.The reference point of the vehicle 10 can, in particular, coincide in the lateral direction with the longitudinal center plane of the vehicle 10 and in the vertical direction with the contact plane of the vehicle 10 on the ground, and in the forward direction with the center of an axle, e.g., the axle of the rear wheels 20 of the vehicle 10. These calculations are simplified by the fact that points A to E lie in a common plane that is parallel to or coincides with the contact plane of the vehicle 10. Ultimately, the processor of the input and output unit 30 or the steering control unit 46 has information about the position and orientation of the camera 32 (and its image scale) relative to the vehicle, based on which it can assign the pixels of the image sensor of the camera 32 to a two- or three-dimensional position in space.

[0037] To detect the yaw and / or roll angle of the camera 32, the robot 66 can also land on the roof of the cabin 22 of the vehicle 10. The roll angle (lateral inclination relative to the horizontal contact plane of the vehicle) of the vehicle 10 can be detected by two or three of the points G, H, I of the Fig. 5 (if the roof is flat, two points are sufficient; if it is curved, points G and I are useful) and the three-dimensional position of the robot 66 with its navigation unit 74. The yaw angle of the vehicle 10 (angle around the vertical axis) can be detected by approaching points H and N and detecting the position of the robot 66. In order to detect the yaw angle of the camera 32, i.e. the deviation of its image axis from the longitudinal axis of the vehicle 10, the robot 66 flies to position K in front of the vehicle 10. Point K lies on the imaginary extension line HN, which must be perpendicular to line GHI. The yaw angle of the camera 32 is calculated by determining the position of the robot 66 or the marker 86 in the image. If the robot is centrally located in the image of the camera 32 ( Fig. 6, upper part, position K'), the yaw angle of the camera 32 is 0°. If the robot 66 or the marker 86 is not positioned centrally in the image, but offset by a number of pixels, as in the lower part of the Fig. 6, one can calculate the number of pixels between the points K (lower part of the Fig. 6, with the position of the robot 66 captured in the image) and K' (in the center of the image) are converted into a lateral offset Δ using trigonometric calculations. The pitch angle of the vehicle 10 can also be detected in an analogous manner by flying to one of the points J to L and determining the position of the robot 66 in the image of the camera 32 and its navigation unit 74. With this procedure, the navigation unit 48 of the vehicle 10 is not required.

[0038] The procedure described can also be carried out for camera 64, as shown in the Fig. 3 shown.

[0039] As already noted above, the point E can be controlled using the signal from the camera 32 in such a way that it lies at a predetermined position in the image of the camera 32, namely exactly in its center, ie on the optical axis of the camera 32. This facilitates the determination of the orientation (pitch angle and yaw angle) of the camera 32 with respect to the vehicle 10. The "predetermined position" in the sense of the claims can therefore relate, at least at point E, to the reference system of the camera 32. The points A to D can also be referenced to the reference system of the camera 32, iethe robot 66 can be positioned such that it appears at specific locations in the camera image, or they are referenced to point E (which, as described here, can be located based on the reference system of the camera 32 or at a predetermined distance in front of the vehicle 10 on its longitudinal center plane) and offset from it by specific distances in the forward and / or transverse direction. As described above, however, it is possible to bring the robot 66 into a predetermined spatial position relative to the vehicle 10 using its navigation unit 74, to detect it in the image of the camera 32, 64, and to derive therefrom the positioning and / or orientation of the detection system 62 relative to a reference position and / or a reference direction of the vehicle 10. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] EP 4 085 740 A1

[0008] US 6 445 983 B1

[0027] EP 1 266 553 A2

[0027] WO 2008 / 150418 A1

[0027] WO 2018 / 067473 A1

[0027] DE 10 2016 209 437 A1

[0027] DE 10 2014 201 203 A1

[0031] DE 10 2015 206 844 A1

[0031] Cited non-patent literature

[0000] M. Pérez et al, Digital Camera Calibration Using Images Taken From an Unmanned Aerial Vehicle, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-1 / C22, 2011, ISPRS Zurich 2011 Workshop, September 14-16, 2011, Zurich, Switzerland, pages 167 to 171

[0005] M. Heinold et al., Tractor-Integrated Active Device Control for High-Speed ​​Hoes, in A. Ruckelshausen et al. (eds.): Digital Marketplaces and Platforms, Lecture Notes in Informatics (LNI), German Informatics Society, Bonn 2018, pages 95 to 98

[0006] Y. Zhang et al., Camera Calibration for Long-Distance Photogrammetry Using Unmanned Aerial Vehicles, Hindawi Journal of Sensors Volume 2022, Article ID 8573315, https: / / doi.orq / 10.1155 / 2022 / 8573315

[0009] ISO 11783

[0025] Wan, X., Liu, G. (2008). Automatic Navigation System with Multiple Sensors. In: Li, D. (eds) Computer and Computing Technologies in Agriculture, Volume II. CCTA 2007

[0027]

Claims

[1] Method for calibrating the positioning and / or alignment of an optical detection system (62) mounted on an agricultural vehicle (10), comprising the following steps: successively positioning an unmanned robot (66) at a plurality of predetermined positions (AE, JL) relative to the vehicle (10); detecting the robot (66) positioned in this way and / or a marker (86) attached to the robot (66) by means of the detection system (62) which generates an image signal; Processing the image signal of the detection system (62), comprising detecting the position of the robot (66) and / or the marker (86) in the image; and calculating the positioning and / or orientation of the detection system (62) relative to a reference position and / or a reference direction of the vehicle (10) based on the respective positions (AE, JL) of the robot (66) and the associated position of the robot (66) and / or the marking (86) detected in the image. [2] Method according to claim 1, wherein the vehicle (10) and the robot (66) are each equipped with a navigation unit (48, 74) for detecting the spatial position. [3] Method according to one of the preceding claims, wherein the robot (66) is an aircraft. [4] Method according to claim 3, wherein the robot (66) is located at the predetermined positions (AE, JL) in a plane which is parallel to the contact plane of the vehicle (10). [5] Method according to claim 4, wherein the robot (66) detects the orientation of the vehicle (10) by flying to different positions on the vehicle and detecting the associated position of its navigation unit (74) or the navigation unit (48) of the vehicle (10) detects its orientation. [6] Method according to one of the preceding claims, wherein the predetermined positions (AE, JL) relate to a reference system of the vehicle (10) and / or a reference system of the detection system (62). [7] Method according to one of the preceding claims, wherein the calculated positioning and / or orientation of the detection system (62) is used to generate a steering signal based on the signal of the detection system (32) for the vehicle (10) and / or an attachment (12) of the vehicle (10) and / or to control an actuator of the vehicle (10) and / or attachment (12). [8] Device for calibrating the positioning and / or alignment of an optical detection system (62) mounted on an agricultural vehicle (10), comprising means for automatically performing the following steps: successively positioning an unmanned robot (66) at a plurality of predetermined positions (AE, JL) relative to the vehicle; detecting the robot (66) positioned in this way and / or a marker (86) attached to the robot (66) by means of the detection system (62) which generates an image signal; Processing the image signal of the detection system (62), comprising detecting the position of the robot (66) and / or the marker (86) in the image; and calculating the positioning and / or orientation of the detection system (62) relative to a reference position and / or a reference direction of the vehicle (10) based on the respective positions (AE, JL) of the robot (66) and the associated position of the robot (66) and / or the marking (86) detected in the image.

Images