METHOD AND DEVICE FOR DETERMINING A TRAILER ANGLE

Abstract

A method for determining a trailer angle includes obtaining a bird's-eye view image of a commercial vehicle using image data from at least one camera mounted on the commercial vehicle and depicting at least one side of the trailer. The commercial vehicle includes a tractor and a trailer. The method further includes determining which of a plurality of lines in the bird's-eye view image intersect a predefined pivot region including and surrounding a pivot point between the trailer and the tractor, determining a trailer angle of the trailer relative to the tractor based on at least a first portion of the plurality of lines intersecting the predefined pivot region, and panning a video feed of images from the at least one camera based on the trailer angle.Determining the trailer angle includes excluding a second portion of the plurality of lines that do not intersect the predefined turning range from determining the trailer angle or weighting the first portion of the plurality of lines more heavily than the second portion of the plurality of lines when determining the trailer angle. A system for a commercial vehicle is also disclosed.

Description

TECHNICAL FIELD

[0001] This disclosure relates to a method and apparatus for determining a trailer angle between a trailer and a tractor. BACKGROUND

[0002] Vehicle camera systems, either as a replacement for mirrors or to supplement mirror views, are used in commercial vehicles to enhance a driver's ability to view the surrounding environment. Camera monitor systems (CMS) utilize one or more cameras to provide a driver with an enhanced field of view. In some examples, mirror replacement systems cover a larger field of view than a conventional mirror or include views that cannot be fully obtained via a conventional mirror.

[0003] In a commercial vehicle environment, as a tractor-trailer truck consisting of a tractor unit and a trailer turns or corners, the trailer centerline deviates at an increasingly large angle relative to the tractor unit centerline. Therefore, it is desirable for the driver to have a field of view on the side of the vehicle corresponding to the turning direction with a larger "outward" angle relative to the tractor unit centerline to ensure the rearmost trailer wheels remain visible on the screen. This functionality in a CMS display has been described as "auto-panning"—the digital camera system simulates the axial rotation (a panning motion) of a conventional camera to translate the displayed view from side to side.However, detecting an angle of the trailer relative to the tractor during such an image pan presents technical challenges because the physical environment is not well suited for a mechanical or limit switch sensor.

[0004] When a vehicle is traveling in a forward direction, a geometric, "kinematic" model can be used to estimate the trailer angle based on the relevant dimensions of the vehicle combination (axle base, speed, steering angle), provided they are known. However, when the vehicle is reversing, the kinematics become unreliable, as errors in such algorithms accumulate over time (e.g., a vehicle combination with the tractor reversing straight ahead may, over time, yield a full range of trailer angles based on very small differences in the initial trailer angle). SUMMARY

[0005] A method for determining a trailer angle according to an exemplary embodiment of the present disclosure includes obtaining a bird's-eye view image of a commercial vehicle using image data from at least one camera mounted on the commercial vehicle and depicting at least one side of the trailer. The commercial vehicle includes a tractor and a trailer. The method includes determining which of a plurality of lines in the bird's-eye view image intersect a predefined pivot region that includes and surrounds a pivot point between the trailer and the tractor, determining a trailer angle of the trailer relative to the tractor based on at least a first portion of the plurality of lines intersecting the predefined pivot region, and panning a video feed of images from the at least one camera based on the trailer angle.Determining the trailer angle includes excluding a second portion of the plurality of lines that do not intersect the predefined turning range from determining the trailer angle or giving greater weight to the first portion of the plurality of lines than to the second portion of the plurality of lines in determining the trailer angle.

[0006] In another embodiment of the above embodiment, determining the trailer angle comprises performing the step of excluding the second portion of the plurality of lines that do not intersect the predefined turning range from the determination of the trailer angle.

[0007] In a further embodiment of any of the preceding embodiments, determining the trailer angle comprises performing the step of weighting the first portion of the plurality of lines more heavily than the second portion of the plurality of lines in determining the trailer angle.

[0008] In a further embodiment of any of the preceding embodiments, the at least one camera comprises a first camera and a second camera on opposite sides of the commercial vehicle, both of which are rear-facing cameras, and obtaining the bird's eye view image of the commercial vehicle from the image data from the at least one camera comprises obtaining an image set containing a respective image from each of the first camera and the second camera, and applying reverse perspective mapping to convert image data from the respective images of the image set into the bird's eye view image.

[0009] In a further embodiment of any of the preceding embodiments, the method comprises adjusting a first image in the image set from the first camera to account for distortion of the first camera and adjusting a second image in the image set from the second camera to account for distortion of the second camera, and the bird's eye view image is obtained from the adjusted first image and the adjusted second image.

[0010] In a further embodiment of any of the preceding embodiments, determining the trailer angle of the trailer comprises determining a plurality of angles between a centerline of the tractor and at least the first portion of each of the plurality of lines and determining the trailer angle based on the plurality of angles.

[0011] In a further embodiment of any of the preceding embodiments, the rotation range in the bird's eye view image is within a perimeter of the trailer.

[0012] In a further embodiment of any of the preceding embodiments, the rotation area has an area which, in the bird's eye view image, is less than 20% of an area of the trailer.

[0013] In a further embodiment of any of the preceding embodiments, the method further comprises repeating the step of obtaining and the steps of determining for a plurality of the image sets to determine a sequence of a plurality of trailer angles, determining a confidence value for each of the plurality of trailer angles, and setting a frequency with which the step of applying, the step of obtaining, and the steps of determining are performed based on the confidence value.

[0014] In another embodiment of any of the preceding embodiments, the method further comprises repeating the step of obtaining and the steps of determining for a plurality of image sets to determine a plurality of trailer angles, determining an aggregated trailer angle based on the plurality of trailer angles, and performing the panning based on the aggregated trailer angle.

[0015] A system for a commercial vehicle according to an exemplary embodiment of the present disclosure includes at least one camera mounted on a commercial vehicle. The commercial vehicle includes a tractor and a trailer, and processing circuitry operatively connected to a memory. The processing circuitry is configured to obtain a bird's-eye view image of the commercial vehicle based on image data from the at least one camera, determine which of a plurality of lines in the bird's-eye view image intersect a predefined pivot region including and surrounding a pivot point between the trailer and the tractor, determine a trailer angle of the trailer relative to the tractor based on at least a first portion of the plurality of lines intersecting the predefined pivot region, and pan a video feed of images from the at least one camera based on the trailer angle.To determine the trailer angle, the processing circuitry is configured to exclude a second portion of the plurality of lines that do not intersect the predefined turning range from the determination of the trailer angle or to give greater weight to the first portion of the plurality of lines than to the second portion of the plurality of lines when determining the trailer angle.

[0016] In a further embodiment of the above embodiment, for determining the trailer angle, the processing circuitry is configured to exclude the second part of the plurality of lines that do not intersect the predefined turning range from the determination of the trailer angle.

[0017] In a further embodiment of any of the preceding embodiments, for determining the trailer angle, the processing circuitry is configured to give greater weight to the first part of the plurality of lines when determining the trailer angle than to the second part of the plurality of lines.

[0018] In a further embodiment of any of the preceding embodiments, the at least one camera comprises a first camera and a second camera on opposite sides of the commercial vehicle, both of which are rear-facing cameras. To obtain the bird's-eye view of the commercial vehicle from image data from the at least one camera, the processing circuitry is configured to obtain an image set including a respective image from each of the first camera and the second camera, and to apply reverse perspective mapping to convert image data from the respective images of the image set into a bird's-eye view image.

[0019] In a further embodiment of any of the preceding embodiments, the processing circuitry is configured to adjust a first image in the image set from the first camera to account for distortion of the first camera and to adjust a second image in the image set from the second camera to account for distortion of the second camera. The bird's-eye view image is obtained from the adjusted first image and the adjusted second image.

[0020] In a further embodiment of any of the preceding embodiments, to determine the trailer angle of the trailer with respect to the tractor based on at least a first portion of the plurality of lines intersecting the predefined turning range, the processing circuitry is configured to determine a plurality of angles between a centerline of the tractor and at least the first portion of the plurality of lines, respectively, and to determine the trailer angle based on the plurality of angles.

[0021] In a further embodiment of any of the preceding embodiments, the rotation range in the bird's eye view image is within a perimeter of the trailer.

[0022] In a further embodiment of any of the preceding embodiments, the rotation area has an area which, in the bird's eye view image, is less than 20% of an area of the trailer.

[0023] In a further embodiment of any of the preceding embodiments, the processing circuitry is configured to determine a plurality of the trailer angles, determine a confidence value for each of the plurality of trailer angles, and adjust a frequency with which the trailer angles are determined based on the confidence value.

[0024] In a further embodiment of any of the preceding embodiments, the processing circuitry is configured to determine a plurality of the trailer angles, determine an aggregated trailer angle based on the plurality of trailer angles, and perform the pivoting based on the aggregated trailer angle.

[0025] The embodiments, examples, and alternatives of the preceding paragraphs, the claims, or the following description and drawings, including all of their various aspects or respective individual features, may be considered independently of one another or in any combination. Features described in connection with one embodiment are applicable to all embodiments unless those features are incompatible. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The disclosure will be better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings, in which: Fig. Figure 1A is a schematic front view of a truck with a camera mirror system (CMS) used to provide at least Class II and Class IV views. Fig. 1B is a schematic view of the truck of Fig. 1A bird's eye view with a CMS providing Class II, Class IV, Class V and Class VI views. Fig. 1C is a schematic view of the truck of Fig. 1A from a bird's eye view with a trailer angle of zero. Fig. 1D is a schematic view of the truck of Fig. 1A from a bird's eye view with a trailer angle different from zero. Fig. 2 is a schematic perspective top view of a vehicle cabin having displays and interior cameras. Fig. 3A schematically shows an exemplary image of a commercial vehicle from a bird's eye view. Fig. 3B an enlargement of the image from a bird’s eye view of Fig. 3A. Fig. 4 a flowchart of an exemplary method for determining a trailer angle. Fig. 5 a flowchart of an exemplary implementation of a step of the flowchart of Fig. 4. DETAILED DESCRIPTION

[0027] In the Fig. 1A, Fig. 1B, Fig. 1C and Fig. 1D, schematic views of a commercial vehicle 10 are shown. The vehicle 10 includes a cab or "tractor" 12 for towing a trailer 14, with the trailer 14 pivoting relative to the tractor 12 when cornering. While the vehicle 10 is illustrated in this disclosure as a truck with a single trailer, it is understood that other commercial vehicle configurations may be used (e.g., different types or quantities of trailers, articulated buses, etc.).

[0028] Two camera arms 16a, 16b each have a base, which is attached, for example, to the driver's cab 12. A pivot arm is mounted on the base and can move relative to it in an articulated manner. At least one rearward-facing camera 20a, 20b (generally camera 20) is arranged in each of the camera arms 16a, 16b. The external cameras 20a, 20b each provide an external field of view (FOV). EX1 , FOV EX2 each containing a Class II and / or Class IV view ( Fig. 1B), which are legally required views in the commercial trucking industry. The Class II view on a given side of the vehicle 10 is a subset of the Class IV view on the same side of the vehicle 10. If desired, multiple cameras in each camera arm 16a, 16b may also be used to provide these views. Class II (narrow) and Class IV (wide) views are defined, for example, in European legislation R46, and similar visibility requirements exist in the United States and other countries when driving trucks. Any reference to a "class" view is not intended to be limiting, but to serve as an example of the type of view provided by a particular camera to a display. Each arm 16A-16B may also provide a housing that houses electronics, e.g.a control unit, which is configured to provide various features of the CMS 15. The camera arms 16A-B may be mounted either at a roof mounting location above the cab door (as shown) or, for example, at a door-mounted bracket or position.

[0029] If videos of Class V and Class VI views are also desired, a camera housing 16C and a camera 20C may be arranged at or near the front of the vehicle 10 to provide these views ( Fig. 1B).

[0030] A trailer rear camera 20D can be provided, which has a field of view FOV EX3The trailer rear camera 20D can, for example, be mounted on a top edge / centerline of the trailer, at the level of a bumper / bed of the trailer, or at an upper corner of the rear of the trailer. Alternatively, or in addition to the trailer rear camera, a "fifth wheel camera" 20E can be provided, which is mounted on a rear side of the tractor unit and has a field of view FOV EX4 The fifth wheel camera 20E can, for example, be mounted anywhere between the lateral plane of the fifth wheel mounting and the top edge / roof edge of the tractor unit.

[0031] As in Fig. 1C-1D, a "fifth wheel pin" 11 is a pivot point between the tractor 12 and the trailer 14 that allows the trailer 14 to rotate relative to the tractor 12 during cornering. The tractor 12 has a central longitudinal axis L1, and the trailer 14 has a central longitudinal axis L2. As shown in Fig. 1B, when the tractor 12 is not cornering, the axes L1, L2 are parallel and coaxial, respectively, and there is no angle between the axes L1, L2. As shown in Fig. 1C, when the tractor 12 corners, an angle θ T between the axes L1, L2. The angle between the axes L1, L2, which is Fig. 1C is approximately 20°, is referred to herein as the “trailer angle”.

[0032] With reference to Fig. 2 and further referring to the Fig. 1A-B has a camera mirror system (CMS) 15 driver and passenger side camera arms 16A, 16B, which are mounted on the outside of the vehicle cab 12 ( Fig. 1A). If desired, the camera arms 16A, 16B may also have conventional mirrors integrated therewith, although the CMS 15 may serve to replace mirrors entirely. In additional examples, each side may have multiple camera arms, with each arm housing one or more cameras and / or mirrors. As shown in Fig. 2, the CMS 15 includes a CMS controller 13 that includes processing circuitry that supports the operation of the CMS 15 and is operatively connected to a memory (which may include any one or a combination of volatile memory elements (e.g., random access memory (RAM, such as DRAM, SRAM, SDRAM, VRAM, etc.)) and / or non-volatile memory elements (e.g., ROM, hard disk, tape, CD-ROM, etc.)). The processing circuitry may include one or more microprocessors, microcontrollers, application-specific integrated circuits (ASICs), or the like.

[0033] First and second electronic displays 18A-B, which may be video displays, are disposed on the driver and passenger sides, respectively, within the vehicle cab 12 at or near the A-pillars 19A, 19B to display Class II and Class IV views on the respective side of the vehicle 10, providing rearward-looking side views along the vehicle 10 captured by the exterior cameras 20A, 20B.

[0034] As explained above, if video recording of Class V and Class VI views is also desired, the camera housing 16C and the camera 20C may be arranged at or near the front of the vehicle 10 to provide these views ( Fig. 1B). A third display 18C, located in the cab 12 near the top center of the windshield, may be used to show the driver the Class V and Class VI views directed toward the front of the vehicle 10.

[0035] If video recording of Class VIII views is desired, camera housings may be arranged on the sides and rear of the vehicle 10 to provide fields of view including some or all of the Class VIII zones of the vehicle 10. In such examples, the third display 18C may include one or more image fields on which the Class VIII views are displayed. Alternatively, additional displays may be added proximate the first, second, and third displays 18A, 18B, 18C, creating a display dedicated to providing a Class VIII view. The displays 18A, 18B, 18C face a driver area 34 within the cab 32 in which a vehicle operator sits in a driver seat 36.

[0036] In Fig. 3A schematically illustrates an exemplary bird's-eye view image 40 of the commercial vehicle 10 during a right turn maneuver based on image data from cameras 20A, 20B, and 20D. The bird's-eye view image 40 includes a first portion 42A corresponding to image data from camera 20A, a second portion 42B corresponding to image data from camera 20B, and a third portion 42C corresponding to image data from camera 20D. The bird's-eye view image 40 includes an assumed trailer area 44A in which the commercial vehicle 10 is expected to be located if the commercial vehicle 10 is not turning. However, in the image 40, the commercial vehicle is turning similarly to Fig. 1C, wherein the centerline / central longitudinal axis L1 of the tractor 12 rotates about the kingpin 11 and is angled with respect to the centerline / central longitudinal axis L2 of the trailer 14. Consequently, a right side of the trailer is shown in image area 44B.

[0037] U.S. Department of Transportation ("DOT") reflective tape sections 48A-B are shown in image area 44B of trailer 14. The DOT tape has alternating colors (red and white) and is typically applied horizontally along the bottom of a trailer.

[0038] In Fig. 3B is a schematic enlarged portion of the bird's eye view image 40 of Fig. 3A. As shown, the trailer image area 44B includes a plurality of lines 50A-G that have been extended to see if they extend through a pivot area 52 that includes and surrounds a pivot point (knight pin 11) between the tractor 12 and the trailer 14. Lines 50A-B (corresponding to edges of the DOT tape portion 48A) and lines 50C-D (corresponding to edges of the DOT tape portion 48B) all pass through the pivot area 52. Lines 50E-G, which correspond to other edges of the DOT tape, do not pass through the pivot area. In addition to reflective DOT tape, so-called "rubbing struts" that run along a portion of a length of the trailer and protect the trailer 14 from rubbing against outside objects or other items on the side of a trailer can provide useful lines that intersect the pivot area 52.It should be understood that the use of DOT tape is a non-limiting example of a line that may be provided on a trailer and pass through the turning area 52, and that other lines that do not conform to DOT tape and pass through the turning area 52 may be used as a basis for determining a trailer angle.

[0039] In the Fig. 3A-B each include a trailer region 44A, in one or more embodiments the bird's eye view image may narrow or omit this region, e.g., such that the pivot region is a line segment extending along the axis L2.

[0040] In one or more embodiments (e.g. as in Fig. 3B), the turning area 52 lies within a perimeter of the trailer in the bird's eye view image 40. In one or more embodiments, the turning area 52 has an area that is less than 20% of an area of the trailer in the bird's eye view image.

[0041] Fig. 4 is a flowchart of an exemplary method 100 for determining a trailer angle. Referring to Fig. 4 and further with reference to the Fig. 3A-B, a bird's eye view image 40 of a trailer of the commercial vehicle 10 is obtained (step 102) using image data from at least one camera 20 mounted on the commercial vehicle 10 that images at least one side of the trailer 14 (e.g., using image data corresponding to portions 42A-B or 42A-C). Using conventional image processing techniques, a plurality of lines in the bird's eye view image are detected (step 104).

[0042] It is determined which of the plurality of lines 50A-G in the bird's-eye view image 40 cross the predefined rotation area 52 (step 106). In the present example, the lines 50A-G include a first portion of the lines (50A-D) that cross the rotation area 52 and a second portion of the lines (50E-G) that do not cross the rotation area 52.

[0043] Based on at least the first portion of the plurality of lines that cross the predefined turning area (i.e., lines 50A-D), a trailer angle of the trailer 14 relative to the tractor 12 is determined (step 108), and a video feed of images from the at least one camera from step 102 is panned (step 110) based on the camera angle (e.g., to maintain an area surrounding one or more sides of the trailer 14 in view for a vehicle occupant on the displays 18A and / or 18B).

[0044] The determining of step 108 includes excluding the second portion of the plurality of lines (50E-G) that do not intersect the predefined turning range from the determination of the trailer angle or weighting the first portion of the plurality of lines more heavily than the second portion of the plurality of lines in the determination of the trailer angle. Both options serve as a type of filtering, giving less weight to lines outside the turning range.

[0045] Fig. 5 is a flowchart of an exemplary implementation of a step of the flowchart of Fig. 4. An image set is obtained (step 120) containing a respective non-bird's-eye view image from a first camera (e.g., camera 20A) and a second camera (e.g., camera 20B) on opposite sides of the commercial vehicle 10, both of which are rear-facing cameras. "Rear-facing" here means that a camera is directed toward the rear of a vehicle.

[0046] The images are pre-processed (step 122) to account for intrinsic camera parameters (e.g., device-specific distortions, including the positioning of the camera lens relative to the video sensor) and / or extrinsic camera parameters (e.g., the physical positioning of the camera on the vehicle, aligned to a common coordinate system under real-world conditions, which may vary somewhat from vehicle to vehicle). In one or more embodiments, this includes adjusting a first image in the image set from the first camera to account for distortions of the first camera and adjusting a second image in the image set from the second camera to account for distortions of the second camera, and the bird's-eye view image of step 102 is obtained from the adjusted first image and / or the adjusted second image.

[0047] To convert image data from the respective images in the image set into a bird's-eye view, a "reverse perspective mapping" is applied (step 124). The OpenCV function "warpPerspective" can be used as part of this conversion. The "warpPerspective" function selects four reference points on a camera image and specifies their pixel coordinates and the corresponding world coordinates in the tractor's coordinate system as input parameters for a perspective mapping transformation.

[0048] By mapping the reverse perspective of step 124, the non-bird's-eye view from the images in the image set of step 120 are converted and mapped to a bird's-eye view of the commercial vehicle 10.

[0049] Again with reference to Fig.4, step 106 (determining which of the plurality of lines in the bird's-eye view image intersect the region of rotation) may include one or more techniques such as Hough transforms, ellipse filters / detectors, Ransack, and / or other CV (computer vision) techniques used to detect linear and near-linear features. The same techniques may be used to determine boundaries of the trailer 14 in the images from the camera(s) 20.

[0050] The pivot point at which the trailer 14 is attached to the tractor 12 (i.e., the "fifth wheel pin") remains at the same coordinates in a correctly transformed bird's eye view image regardless of the trailer angle and therefore provides a useful anchor reference point for determining which lines within the bird's eye view image correspond to the trailer angle, while filtering out other lines and other image data that may otherwise confound conventional trailer angle detection techniques.

[0051] As explained above, the determining of step 108 includes excluding the second portion of the plurality of lines (50E-G) that do not intersect the predefined turning range from the determination of the trailer angle or weighting the first portion of the plurality of lines more heavily than the second portion of the plurality of lines in the determination of the trailer angle. Both options utilize "radial filtering."

[0052] With the "Weighting" option, radial filtering results in greater importance / weight being given to linear features that intersect the image's rotational region than to those that do not. The radial filtering approach to selecting the most relevant lines within the transformed image can enable a high degree of computational efficiency relative to other line detection and filtering techniques.

[0053] In one or more embodiments, the method 100 includes determining a plurality of angles between a centerline of the tractor 12 (e.g., axle L1) and each of the lines 50A-D intersecting the turning area, and determining the trailer angle based on the plurality of angles (e.g., an average of the plurality of angles).

[0054] In one or more embodiments, method 100 includes repeating steps 102, 104, 106, and 108 to determine a plurality of trailer angles, determining an aggregated trailer angle based on the plurality of trailer angles, and providing a notification and / or pivoting from step 110 based on the aggregated trailer angle. The aggregate may be an average or a weighted average (e.g., of the most favorable weighted lines).

[0055] In one or more embodiments, the method includes repeating steps 102, 104, 106, and 108 for multiple image sets to determine a sequence of multiple trailer angles, determining a confidence value for each of the multiple trailer angles, and setting a frequency with which to repeat the method 100 based on the confidence value. Thus, time series filter calculations (e.g., a Kalman filter) or other calculations may be used to combine a series of trailer angle estimates over time to reduce noise and improve accuracy and usability in the overall implementation of the camera-mirror system.

[0056] Such calibration calculations (which may be continuous) may involve comparing trailer angle estimates derived by method 100 with trailer angles derived from other known techniques (e.g., trailer angle estimates using kinematic models) to improve overall accuracy and reliability. The calibration calculation may involve introducing a correction value into the bird's-eye view-derived trailer angle estimate such that the corrected value results in a "zero" angle at times when the tractor-trailer combination is moving forward at high speed (i.e., > 80 km / h).

[0057] The method 100 provides a new and innovative approach for determining the trailer angle from a "bird's-eye view" image (which facilitates the incorporation of information from multiple cameras into the algorithm's logic). The radial filtering feature extraction technique explained above (e.g., giving less weight to lines outside the rotation range) results in a groundbreaking level of performance in terms of the angular range and conditions under which the trailer angle can be calculated accurately and computationally efficiently.

[0058] While exemplary embodiments have been disclosed, one of ordinary skill in the art will recognize that certain modifications would fall within the scope of the claims. Therefore, the following claims should be studied to determine their true scope and content.

Claims

[1] Method for determining a trailer angle, comprising: Obtaining a bird's-eye view image of a commercial vehicle using image data from at least one camera mounted on the commercial vehicle, representing at least one side of the trailer, the commercial vehicle comprising a tractor and a trailer, Determine which of several lines in the bird's eye view image intersect a predefined pivot area containing and surrounding a pivot point between the trailer and the tractor, Determining a trailer angle of the trailer with respect to the tractor based on at least a first portion of the plurality of lines intersecting the predefined turning range, and Panning a video feed of images from the at least one camera based on the trailer angle, wherein determining the trailer angle comprises: Excluding a second part of the plurality of lines that do not intersect the predefined turning area from the determination of the trailer angle or Weighting the first part of the multiple lines more heavily than the second part of the multiple lines when determining the trailer angle. [2] The method of claim 1, wherein determining the trailer angle comprises performing the step of excluding the second portion of the plurality of lines that do not intersect the predefined turning range from determining the trailer angle. [3] The method of claim 1 or 2, wherein determining the trailer angle comprises weighting the first portion of the plurality of lines more heavily than the second portion of the plurality of lines in determining the trailer angle. [4] Method according to one of the preceding claims, in which: the at least one camera comprises a first camera and a second camera on opposite sides of the commercial vehicle, both of which are rear-facing cameras, and Obtaining the bird's-eye view image of the commercial vehicle from the image data from the at least one camera comprises: Obtaining an image set containing a respective image from each of the first camera and the second camera, and Applying reverse perspective mapping to convert image data from the respective images of the image set into the bird's eye view image. [5] A method according to claim 4, comprising: Adjusting a first image in the image set from the first camera to account for distortion of the first camera and Adjusting a second image in the image set from the second camera to account for distortion of the second camera, where the bird's eye view image is obtained from the adjusted first image and the adjusted second image. [6] A method according to any one of the preceding claims, wherein determining the trailer angle of the trailer comprises: Determining a plurality of angles between a center line of the tractor and at least the first part of the plurality of lines and Determine the trailer angle using the multiple angles. [7] A method according to any one of the preceding claims, wherein the rotational area in the bird's eye view image lies within a perimeter of the trailer. [8] A method according to any one of the preceding claims, wherein the turning area has an area which, in the bird's eye view image, is less than 20% of an area of the trailer. [9] A method according to any one of the preceding claims, comprising: Repeating the step of obtaining and the steps of determining for several of the image sets to determine a sequence of several trailer angles, Determining a confidence value for each of the multiple trailer angles and Setting a frequency with which the applying step, the obtaining step, and the determining steps are performed based on the confidence value. [10] A method according to any one of the preceding claims, comprising: Repeating the obtaining step and the determining steps for multiple image sets to determine multiple trailer angles, Determining an aggregated trailer angle based on the multiple trailer angles and Perform slewing based on the aggregated trailer angle. [11] System for a commercial vehicle, comprising: at least one camera mounted on a commercial vehicle, the commercial vehicle comprising a tractor and a trailer, and processing circuitry operatively connected to a memory and arranged to: based on image data from at least one camera, receives a bird's eye view image of the commercial vehicle, determines which of several lines in the bird's-eye view image intersect a predefined pivot area containing and surrounding a pivot point between the trailer and the tractor, determines a trailer angle of the trailer relative to the tractor based on at least a first portion of the plurality of lines intersecting the predefined turning range, and based on the trailer angle, a video feed of images from at least one camera is panned, wherein, for determining the trailer angle, the processing circuitry is arranged to: excludes a second part of the plurality of lines that do not intersect the predefined turning area from the determination of the trailer angle or When determining the trailer angle, the first part of the multiple lines is given greater weight than the second part of the multiple lines. [12] The system of claim 11, wherein, to determine the trailer angle, the processing circuitry is configured to exclude the second portion of the plurality of lines that do not intersect the predefined turning range from the determination of the trailer angle. [13] The system of claim 11, wherein, to determine the trailer angle, the processing circuitry is configured to give greater weight to the first portion of the plurality of lines than to the second portion of the plurality of lines when determining the trailer angle. [14] A system according to claim 11, wherein: the at least one camera comprises a first camera and a second camera on opposite sides of the commercial vehicle, both of which are rear-facing cameras, and to obtain the image of the commercial vehicle from a bird's eye view based on image data from the at least one camera, the processing circuit arrangement is arranged such that it: receives an image set containing a respective image from each of the first camera and the second camera, and uses reverse perspective mapping to convert image data from the respective images of the image set into a bird's eye view image. [15] The system of claim 14, wherein the processing circuitry is arranged to: adjusts a first image in the image set from the first camera to account for distortion of the first camera and adjusts a second image in the image set from the second camera to account for distortion of the second camera, where the bird's eye view image is obtained from the adjusted first image and the adjusted second image. [16] The system of claim 11, wherein, to determine the trailer angle of the trailer relative to the tractor based on at least a first portion of the plurality of lines intersecting the predefined turning range, the processing circuitry is arranged to: several angles are determined between a center line of the tractor and at least the first part of the plurality of lines, the trailer angle is determined based on the multiple angles. [17] The system of claim 11, wherein the rotational range in the bird's eye view image is within a perimeter of the trailer. [18] The system of claim 11, wherein the rotation area has an area that is less than 20% of an area of the trailer in the bird's eye view image. [19] The system of claim 11, wherein the processing circuitry is arranged to: several of the trailer angles are determined, determines a confidence value for each of the multiple trailer angles and sets a frequency with which the trailer angles are determined based on the confidence value. [20] A system according to claim 11, comprising: Determine several of the trailer angles, Determining an aggregated trailer angle based on the multiple trailer angles and Perform slewing based on the aggregated trailer angle.

Images