# US 12,223,675 B2 · Method for ascertaining an articulation angle, processing unit, and vehicle > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,223,675 B2 - Document ID: US12223675B2 - Status: Granted - Kind: B2 - Application number: 18/698,020 - Filed: 2022-09-23 - Published: 2024-10-03 - Granted: 2025-02-11 - Earliest claimed: 2021-10-15 - Assignees: ZF CV Systems Global GmbH - Applicants: Zf Cv Systems Global GmbH - Inventors: Tobias Klinger - Web page: https://patentgrep.com/patent/US12223675B2 ## CPC - G06T 7/73: using feature-based methods - G06T 3/06: Topological mapping of higher dimensional structures onto lower dimensional surfaces - G06T 3/60: Rotation of whole images or parts thereof - G06T 7/12: Edge-based segmentation - G06T 7/207: for motion estimation over a hierarchy of resolutions (multi-resolution motion estimation or hierarchical motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/53) - G06T 7/55: from multiple images - G06T 2207/10028: Range image; Depth image; 3D point clouds - G06T 2207/30252: Vehicle exterior; Vicinity of vehicle - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract A method for determining an articulation angle between a towing vehicle and a trailer, wherein the trailer can be imaged in single images with single-image pixels using a camera system. The method includes determining a point cloud from at least one single image and forming at least one planar reference surface from adjacent object points of the point cloud that lie in a same plane. The method further includes reading in at least one single image, in which the trailer coupled to the towing vehicle is imaged, and determining at least one partial surface and fitting the at least one previously formed planar reference surface to the at least one determined partial surface by applying a geometric transformation. The method further includes determining the articulation angle between the towing vehicle and the trailer as a function of a rotational component of the applied geometric transformation. ## Claims ### Claim 1 A method for determining an articulation angle between a towing vehicle and a trailer coupled to the towing vehicle, wherein the trailer can be imaged in single images with single-image pixels using a camera system, the method comprising: determining a point cloud from at least one single image in which at least the trailer is represented, wherein the point cloud comprises object points assigned to the trailer and wherein the at least one single image is captured when the trailer is not coupled to the towing vehicle; forming at least one planar reference surface corresponding to a contiguous surface of the trailer from adjacent object points of the point cloud that lie in a same plane, wherein the respective planar reference surface is in a fixed spatial orientation and/or a fixed spatial position relative to the trailer; reading in at least one single image, in which the trailer coupled to the towing vehicle is imaged, and determining at least one partial surface in this at least one single image read in; fitting the at least one planar reference surface to the at least one determined partial surface of the read-in single image by applying a geometric transformation; and determining the articulation angle between the towing vehicle and the trailer as a function of a rotational component of the applied geometric transformation, wherein the rotational component of the applied geometric transformation characterizes a rotation of the planar reference surface about a pivot axis of the trailer. ### Claim 2 (depends on 1) The method as claimed in claim 1, wherein the point cloud determined from the at least one single image and/or the at least one planar reference surface that is formed is stored. ### Claim 3 (depends on 1) The method as claimed in claim 1, wherein the point cloud is determined by a Structure from Motion algorithm from at least two single images, and wherein the trailer in the at least two single images is captured from at least two different viewpoints. ### Claim 4 (depends on 2) The method as claimed in claim 2, wherein the point cloud is formed from a plurality of object points, which are assigned at least to the trailer, and wherein for this purpose depth information for the respective object points is obtained from at least two single images by triangulation. ### Claim 5 (depends on 1) The method as claimed in claim 1, wherein the at least one planar reference surface is defined from adjacent object points which lie in the same plane, such that the planar reference surface can be assigned to a region delimited or delimitable by color and/or structure comprising adjacent single-image pixels in the at least one single image. ### Claim 6 (depends on 5) The method as claimed in claim 5, wherein the planar reference surface is formed from the adjacent object points of the point cloud which lie in the same plane and/or to which single-image pixels with the same or at least similar color values and/or intensity values in the at least one single image are assigned. ### Claim 7 (depends on 1) The method as claimed in claim 1, wherein the at least one planar reference surface completely covers a trailer front or only partially occupies the trailer front. ### Claim 8 (depends on 1) The method as claimed in claim 1, wherein adjacent single-image pixels with the same or at least similar color values and/or intensity values are assigned to the same partial surface of the read-in single image. ### Claim 9 (depends on 1) The method as claimed in claim 1, further comprising: forming an intermediate surface from the at least one previously formed planar reference surface by means of a coordinate transformation, wherein the planar reference surface exists in a three-dimensional coordinate system and the intermediate surface in a two-dimensional coordinate system; and fitting the respectively formed intermediate surface to the respective partial surface of the read-in single image by the geometric transformation. ### Claim 10 (depends on 9) The method as claimed in claim 9, wherein the respectively formed intermediate surface is translated and/or rotated and/or sheared and/or scaled by the geometric transformation. ### Claim 11 (depends on 9) The method as claimed in claim 9, wherein pixel-by-pixel color values and/or intensity values in the respective partial surface and in the formed intermediate surface are compared for fitting the at least one previously formed planar reference surface to the at least one determined partial surface. ### Claim 12 A processing unit for a vehicle including a towing vehicle and a trailer the processing unit being configured to: determine a point cloud from at least one single image captured by a camera system, in which at least the trailer is imaged, wherein the point cloud comprises object points assigned to the trailer and wherein the at least one single image is captured when the trailer is not coupled to the towing vehicle; form at least one planar reference surface corresponding to a contiguous surface of the trailer from the adjacent object points of the point cloud that lie in a same plane, wherein the respective planar reference surface is in a fixed spatial orientation and/or a fixed spatial position relative to the trailer; read in at least one single image, in which the trailer coupled to the towing vehicle is imaged, and to determine at least one partial surface in this at least one single image read in; fit at least one previously formed planar reference surface to the at least one determined partial surface by applying a geometric transformation; and determine an articulation angle between the towing vehicle and the trailer as a function of a rotational component of the applied geometric transformation, wherein the rotational component of the applied geometric transformation characterizes a rotation of the planar reference surface about a pivot axis of the trailer. ### Claim 13 (depends on 12) A vehicle comprising: a towing vehicle; a trailer; the processing unit as claimed in claim 12; and a camera system, the viewing range of which is or can be aligned to the trailer in a rear space of the towing vehicle. ### Claim 14 (depends on 1) The method as claimed in claim 1, wherein to determine the point cloud, the at least one single image is captured during a coupling process. ### Claim 15 (depends on 2) The method as claimed in claim 2, wherein the point cloud determined from the at least one single image and/or the at least one planar reference surface that is formed is stored with the spatial orientation and/or the spatial position of the respective planar reference surface relative to the trailer. ### Claim 16 (depends on 3) The method as claimed in claim 3, wherein the trailer in the at least two single images is captured from at least two different viewpoints by a monocular camera of the camera system. ### Claim 17 (depends on 10) The method as claimed in claim 10, wherein the respectively formed intermediate surface is translated and/or rotated and/or sheared and/or scaled by the geometric transformation in a series of iteration steps. ### Claim 18 (depends on 13) The vehicle as claimed in claim 13, wherein the camera system includes a monocular camera. ### Claim 19 A method for determining an articulation angle between a towing vehicle and a trailer coupled to the towing vehicle, wherein the trailer can be imaged in single images with single-image pixels using a camera system, the method comprising: determining a point cloud from at least one single image in which at least the trailer is represented, wherein the point cloud comprises object points assigned to the trailer; forming at least one planar reference surface from adjacent object points of the point cloud that lie in a same plane, wherein the respective planar reference surface is in a fixed spatial orientation and/or a fixed spatial position relative to the trailer; reading in at least one single image, in which the trailer coupled to the towing vehicle is imaged, and determining at least one partial surface in this at least one single image read in; fitting the at least one planar reference surface to the at least one determined partial surface of the read-in single image by applying a geometric transformation; and determining the articulation angle between the towing vehicle and the trailer as a function of a rotational component of the applied geometric transformation, wherein the rotational component of the applied geometric transformation characterizes a rotation of the planar reference surface about a pivot axis of the trailer, wherein the at least one planar reference surface is defined from adjacent object points which lie in the same plane, such that the planar reference surface can be assigned to a region delimited or delimitable by color and/or structure comprising adjacent single-image pixels in the at least one single image, and wherein the planar reference surface is formed from the adjacent object points of the point cloud which lie in the same plane and/or to which single-image pixels with the same or at least similar color values and/or intensity values in the at least one single image are assigned.