Method and system for calculating the angle between a vehicle and a trailer

Abstract

The yaw angle of a trailer connected to a towing vehicle is calculated based on image information provided by a camera on the vehicle. [Solution] The present invention relates to a method for determining the yaw angle (YA) of a trailer (2) relative to the longitudinal axis (LAV) of a towing vehicle (1) by setting a perpendicular bisector (B1) based on features contained in multiple captured images.

Description

Technical Field

[0001] The present invention generally relates to the field of vehicle auxiliary systems. More specifically, the present invention relates to a method for determining the yaw angle of a trailer connected to a towing vehicle based on image information provided by a camera of the vehicle. calculation

Background Art

[0002] Determining the angle of a trailer relative to a towing vehicle based on image information provided by a camera of the vehicle calculation is Multiple methods known.

[0003] In particular , painting when the image is of low quality, While the calculations are less complex, it is robust. it provides do not angle information It is publicly known . In addition, publicly known In the method of , Tou determining the yaw angle, the exact position of the ball (a type of towing device) I understand must be known.

Summary of the Invention

[0004] The object of embodiments of the present invention is to Tou provide a method for determining the yaw angle of a trailer with high Robust accuracy and high reliability without knowing the position of the ball. This problem is solved by the features of the independent claims. Preferred embodiments are set forth in the dependent claims. Unless otherwise specified, embodiments of the present invention can be freely combined with each other calculation . can

[0005] 1 perspective According to one aspect, the present invention provides a method for determining the yaw angle of a trailer relative to the longitudinal axis of a towing vehicle. This method includes the following steps Those who . Regarding This method includes the following steps Prepare .

[0006] First, at least the first 1 and the second​​​2 The image is captured by a camera Use . The first 1 and the second 2 images are captured such that the orientation of the trailer with respect to the vehicle is different on at least 2 the images

[0007] After the images are captured, at least the first 1 feature of the trailer is determined. The first 1 feature must be visible within the first 1 and the second 2 images. The first 1 feature is, for example, a prominent first 1 location feature in the first 1 feature It's okay to have it. .

[0008] the After the feature is determined, a relationship between the camera and the first 1 feature on the second decision captured second 1 feature is set, and the relationship rays of light is projected onto a horizontal plane, thereby obtaining a first rays of light projection feature 1 (the first position of the projected feature 1 position ). Similarly, a relationship between the camera and the first 2 feature on the second decision captured second 1 feature is set, and the relationship rays of light is projected onto the horizontal plane, thereby obtaining a second rays of light projection feature 2 . The projection of the features is respectively position rotated in the vertical direction , rays by You can go, in other words degrees Inclined rays while keeping the azimuth angle unchanged and horizontally rays of light shifted by May I forward it? degrees

[0009] Based on the projection of the first 1 feature, a difference between the location of the first 1 projection feature position and the location of the second place projection feature 2 is determined, and a third position difference between the location of the first 1A perpendicular bisector is set. More specifically, the 1 perpendicular bisector is the 1 ray intersection point with the horizontal plane (e.g., z = 1) and the 2 rays intersection point with the horizontal plane, and can be a vertical line passing through the center of the line connecting them. The place is determined by converting the 1 ray image coordinates of the calibration feature in the camera image into Use optical 1 coordinates. The 1 is determined by converting the rays of light image coordinates of the 2 rays feature in the camera image into calibration optical Use coordinates. The above-mentioned 2 perpendicular bisector can also be set in the horizontal plane. 1 After setting the above-mentioned rays of light perpendicular bisector, the 1 intersection point between the perpendicular bisector and the reference axis Yo is obtained. The reference axis

[0010] can be the central longitudinal axis of the vehicle on which both the camera and the 1 ball are located. Therefore, this intersection point represents the rotation center of the 1 feature. line Finally, the yaw angle of the trailer is 1 obtained based on decision angle estimation. The above-mentioned line angle estimation is based on the angle between the Tou projection feature in the horizontal plane from the line to the 1 intersection point and the

[0011] and the 1 projection feature in the horizontal plane from the calculation to the 1 intersection point. 1 The projection feature position from the 1 to the Continuous intersection point and the 1 line and the 2 projection feature position from the 1 to the Continuously intersection point. 2 wires is used as a reference.

[0012] The above method is one or more A perpendicular bisector is defined, and the yaw angle is determined based on one or more of the aforementioned perpendicular bisectors. calculation In order to release, Two or more Images Use, and one or more trailer of Features use Therefore, even if there is a lot of noise or the image quality is low, the trailer of Feature detection is very accurate and robust It is advantageous because it can be implemented quickly. In addition Yaw angle decision to Because we use the perpendicular bisector. , Tou The exact position of the ball does not need to be known.

[0013] 1 According to the embodiment, 1 or the 2 In the image, the yaw angle of the trailer relative to the vehicle is zero. Therefore, the image is a "zero-pause image". "vinegar In other words, the vertical axis of the vehicle line and the vertical axis of the trailer line As a reference for precise alignment 、 messenger For use Cut. however Other yaw angle values ​​are also used as a reference. Available The other yaw angles mentioned above are I don't know In this case, the system measures the change in the trailer angle, rather than the absolute trailer angle. You may calculate it. .

[0014] 1 According to one embodiment, the method is as follows: Furthermore, it is equipped with . The 1 In addition to its features, 1 and 2 The visible trailer in the image 2 Features decision The steps to do the above. 2 The distinguishing feature is the first on the trailer. 1 It is located in a different position from the characteristic. 2 Features include, for example, the trailer's... 2 locations The most prominent in 2 Features That's good . In addition, the camera and 1 The determined number on the image 2 Between features rays of light However, it is projected onto the horizontal plane, and as a result the 3 Projection features position The step of obtaining the camera and the 2 The determined number on the image 2 Between features rays of light However, it is projected onto the horizontal plane, and as a result the 4 Projection features position It can also be obtained. Sara to, the 3 Projection features position of place and the 4 Projection features position of place Between, the first 2 Perpendicular bisector Set Step. The aforementioned 2 Based on the perpendicular bisector, 2 Perpendicular bisector and the aforementioned reference axis line The first 2 The intersection is decision It will be done. The 2 Based on the perpendicular bisector, 2 Angle estimation is set, 3 Projection features position From 2 Extend to the intersection to be The 1 line And in the horizontal plane, 4 Projection features position From 2 Extend to the intersection to be The 2 wires Based on the angle between the two, 2 A step in which angle estimation is performed. Finally, the yaw angle is as described above. 1 and 2 Based on angle estimation calculation The steps to take.

[0015] 2 or more trailer ofBy using the characteristics and multiple perpendicular bisectors, the perpendicular bisector and the reference axis can be identified. line Between Turning Reduce noise and inconsistencies when determining angles. Japanese Cut.

[0016] 1 According to the embodiment, the third 1 and 2 In addition to the features, at least one of the trailers Sara Characteristics but Yaw angle calculation to Used 3 or more By using the features, the yaw angle can be determined. Robust Quality and reliability will be further improved.

[0017] 1 According to the embodiment, the yaw angle is determined by setting the median value based on at least two angle estimates. calculation This results in a very stable yaw angle. decision but acquisition It will become possible.

[0018] According to other embodiments, the yaw angle is determined by taking the average of at least two angle estimates, or by using a statistical approach applicable to the angle estimates. calculation It will be done.

[0019] 1 According to the embodiment, the method is Sara The following angle frame Decide Steps Prepare The aforementioned angle frame has an upper and lower limit that encloses the aforementioned yaw angle. It is good to be prepared. In addition, the set of features is decision The features within the feature set are used as the basis for estimating angles located within the angle frame. decision Features of Preferably, the set contains only the features included in the feature set for the future yaw angle. calculation It is used. it is , in other words Then , previous yaw angle decision The information is, decision It is quite close to the yaw angle ( In other words Two trailers that derive angle estimation (within the angle frame) or more Features decision death, decision Significant deviation from the yaw angle ( In other words Features that lead to angle estimation (outside the angle frame) To make it non-trackable It is used. This reduces the complexity of the angle estimation calculation and accuracy significantly lower Decrease Cut.

[0020] 1 According to the embodiment, the camera calibration information is 1. Features and The 2 Features At least one of the two The location is changed from the image's local domain to the vehicle's local domain. Used to convert to For example, camera calibration information Use By knowing the camera's position, the location of specific features in the image can be determined by the camera being built into the vehicle. Location or Depending on the location where it is installed, or By correlating them, location Turning information into Exchange Cut.

[0021] 1 According to the embodiment, the reference axis line The cameras and vehicles Tou The ball is on the vertical axis of the towing vehicle. line If the vehicle is positioned in a vertically oriented plane including the longitudinal axis of the towing vehicle, line This is the result. it is , in other words Then The yaw angle is set based on angle estimation, but the angle estimation is based on the longitudinal axis of the towing vehicle. line The angle between the resulting perpendicular bisector is used as the reference point.

[0022] According to other embodiments, camera and Tou ball At least one of the two The lateral direction relative to the longitudinal axis of the towing vehicle It has a discrepancy (offset). In this case, reference axis line The camera and Tou The space between you and the ball ContinuousIt is a straight line. By doing this, the camera and Tou Lateral offset between the ball Compensable .

[0023] 1 According to the embodiment, the camera is a rear-view camera of the vehicle. By using a rear-view camera as a base, images of the trailer can be captured with minimal technical effort.

[0024] 1 According to the embodiment, 1 Features place is, 1 and 2 Only in images decision Rather than being done, at least the 3 In images as well decision It will be done. 1 The first characteristic 3 Within the image place is, 1 and 2 The number in the image 1 Features place It is different from the first. 1 The perpendicular bisector is the 1 Image and 2 The number in the image 1 Between features Can make a decision . Sara The perpendicular bisector is the 1 Image and 3 The number in the image 1 Between features Can make a decision Next, Sara The intersection point is, 1 The perpendicular bisector and the aforementioned Sara Based on the intersection with the perpendicular bisector Can make a decision .

[0025] The aforementioned 1 Based on the perpendicular bisector, 1 Angle estimation calculation The said 1 Angle estimation is, 1 ray Of the intersections of the plane and the horizontal plane, the 2 rays This is the rotation angle around the aforementioned further intersection point with respect to the intersection point of the horizontal plane. 1Angle estimation is, 1 Image and 2 Changes in the trailer's yaw angle between the image and the trailer corresponding . it is , in other words Then The turning point of the trailer is 1 Reference axis of perpendicular bisector line It is not determined by the intersection with, 3 or more It is determined by intersecting at least two perpendicular bisectors obtained by tracking the features on the image.

[0026] Sara Naru perspective According to this, the yaw angle of the trailer relative to the longitudinal axis of the towing vehicle is determined. ru shi The system is disclosed. The system displays images of the trailer. Take a picture Camera and, Filmed Image processing Where Reason entity The system is equipped with the following steps. Possible It is composed of. - The trailer's orientation relative to the vehicle is different in at least two images, camera Use at least the trailer 1 image and 2 Steps to take an image, - No. 1 and 2 At least one of the visible trailers in the image 1 Features Decide Step, - Camera and the 1 On the image decision The first 1 Between features rays of light Project it onto a horizontal plane, and thereby, 1 Projection features position Obtain hand , camera and 2 On the image decision The first 1 Between features rays of light The horizontal plane is projected onto the horizontal plane, thereby the 2 Projection features position Steps to obtain - No. 1Projection features position Location and number 2 Projection features position Between the places 1 Steps to set the perpendicular bisector, - No. 1 Perpendicular bisector and reference axis line The first 1 Intersection Decide Step, - The yaw angle, 1 Based on angle estimation calculation This is a step that, 1 Angle estimation is, In the aforementioned horizontal plane, The 1 Projection features position From 1 To the intersection Continuous The 1 line And, 2 Projection features position From 1 To the intersection Continuously The first 2 wires angle between This is about , the yaw angle, 1 A step that calculates based on angle estimation.

[0027] The above-described embodiment of this method optional The features of the system in this disclosure are of It can also be applied as a characteristic.

[0028] Sara In another embodiment, a vehicle is disclosed that includes a system according to any of the embodiments described above.

[0029] The term "vehicle" as used in this specification "teeth , cars, trucks, buses, trains, or This refers to other powered vehicles. good .

[0030] As used herein, the term "yaw angle" refers to the pivot angle (turning angle) between the longitudinal axis of the vehicle and the longitudinal axis of the trailer. good .

[0031] As used herein, the term "median" refers to the median of a data sample. or This refers to the value that divides the probability distribution from the lower half to the upper half. good .

[0032] Terms used in this specification basic target " or The term "abbreviated" refers to a difference of + / - 10%, preferably + / - 5%, from the exact value. That is , that change, function and at least one of the traffic laws The difference is insignificant. At least one of the two It means that.

[0033] Various aspects of the present invention, including specific features and advantages. perspective This can be easily understood from the detailed explanation and attached drawings below. Probably. [Brief explanation of the drawing]

[0034] [Figure 1] Figure 1 shows an example of a top view of a vehicle towing a trailer. [Figure 2] Figure 2 schematically illustrates angle estimation based on a single feature captured by multiple camera images at different turning angles between the trailer and the towing vehicle. [Figure 3] Figure 3 schematically shows multiple angle estimates based on first and second features captured by multiple camera images at different turning angles between the trailer and the towing vehicle. [Figure 4] Figure 4 schematically shows the geometric determination of the rotation point based on the first and second perpendicular bisectors obtained from the single trailer feature present in three different images. [Figure 5] Figure 5 shows a schematic block diagram illustrating the steps for determining the yaw angle of the trailer relative to the longitudinal axis of the towing vehicle. [Modes for carrying out the invention]

[0035] Next, the present invention will be described more fully by reference to the accompanying drawings showing exemplary embodiments. While embodiments in the drawings may relate to preferred embodiments, all elements and features described in relation to such embodiments may, to the extent appropriate, be used in combination with other embodiments and features described herein, particularly in relation to the embodiments described in more detail earlier. In any case, the present invention should not be construed as being limited to the embodiments described herein. To the extent appropriate, similar reference numerals will be used throughout the following description to indicate similar elements, parts, items, or functions.

[0036] This specification The features of the present invention disclosed in the claims and examples, or the features of the present invention disclosed in the figures are , Each, or In both of these combinations, the present invention can be implemented in various forms. It can be a material. .

[0037] Figure 1 shows a top view of vehicle 1 towing trailer 2. vinegar Vehicle 1 has a vertical axis that passes through the center of Vehicle 1. line LAV Prepare Similarly, Trailer 2 has a longitudinal axis that runs through the center of Trailer 2. line LAT Prepare Trailer 2 is Tou Ball 4 Prepare It is connected to vehicle 1 by a trailer hitch.

[0038] In specific driving conditions, the vertical axis of Vehicle 1 line LAV and Trailer 2 vertical axis line LAT means that even if they are not parallel and aligned, or They do not need to overlap with each other, but the axis line This limits the yaw angle YA. (Rewording) Then The yaw angle YA is the vertical axis of the vehicle 1. line Vertical axis of Trailer 2 relative to LAV line LAT offset angle Definition The yaw angle YA is the longitudinal axis of vehicle 1. line Similar to LAV, the vertical axis of Trailer 2 line In a horizontal plane including LAT May it be measured? . Knowledge regarding yaw angle YA is particularly relevant to, for example, trailers. Support system It is advantageous in this regard.

[0039] Yaw angle YA Regarding the decision, At least part of the Leira 2 multiple Images are captured by camera 3. Camera 3 captures images of the area around the vehicle when it is reversing, for example. For filming, Even if it's a rearview camera on a vehicle good .photograph multiple Image 1 is a towing vehicle Both In the known angular arrangement of trailer 2 relative to 1 Regarding it is fine. The image above shows the yaw angle YA. calculation In terms of standards May you use it? In the known angular arrangement of the trailer 2 relative to the towing vehicle 1, the yaw angle YA is 0° or , or any other arbitrary angle value.

[0040] Figure 2 shows trailer 2 with different yaw angles YA relative to towing vehicle 1. to hold Trailer 2 at different points in time 1 A schematic diagram showing the angular relationship of feature F1 is provided.

[0041] Camera 3 shows the trailer 2 at different points in time, with different angular positions relative to vehicle 1. Two or more You may take pictures. For example, a series The image photograph It may also be used.

[0042] Book In the example, 2 The image may show the orientation of the trailer relative to the vehicle at a yaw angle YA=0°.

[0043] Passage of time spanning The angular motion of Trailer 2 The reason is The specific features detected on the trailer were, 1 and 2 Appearing in different locations within the image It will In Figure 2, the 1 The features are shown as squares.

[0044] The 1 Features (represented by a solid line connecting the feature and camera 3) rays of light The description in the upper section (associated with R) is, 1 Identified in the image, 1 Features (represented by a dashed line connecting the feature and camera 3) rays of light The lower section (associated with R) shows the first at different points in time. 2 It is identified in the image. 1 Features place To correlate this with the location of vehicle 1, specifically with a fixed point on vehicle 1, the camera 3 calibration You may use the information. Specifically, 1 Connecting features with camera 3 rays of light R Regarding the decision Camera 3 calibration Information use , in image coordinates the preceding 1 Features place of rays of light It can also be converted to this. (Alternative translation) Then , camera position and feature position connection wear but , features on the image place And, camera 3 calibration The position of the vehicle's fixed points is correlated with the information.

[0045] Features on the trailer are used for feature detection and matching. Gua Lugorism Use it to pinpoint location and match For example, Harris Corner Detector, Scale-Invariant Feature Transform (SIFT) algorithm, Speeded Up Robust Features (SURF) algorithm, Binary Robust Invariant Scalable Keypoints (BRISK) algorithm, Binary Robust Independent Elementary Features (BRIEF), Oriented FAST and Rotated BRIEF (ORB) algorithms, or other suitable feature detection and matching algorithms. Gua Using Lugorism It will be usable .

[0046] Feature detection and matching Gua Lugorism is on the trailer or Images not on the trailer of Detect features They will Trailer of Features, non-trailer of To distinguish them by their characteristics, many Use different methods It should be possible For example, when moving forward in a straight line, the trailer of The key feature is to look for features that remain in the same position even after time has passed, and non-trailer of Distinguish by characteristics can Alternatively, background of Features motion( motion )of, Modeling over time using known vehicle motions death Yes, this is fine. This can be extracted from CAN data related to speed and steering. deaf That is, the epipolar constraint of the fundamental matrix Does not apply Features: Trailer of It is considered a characteristic. It is possible .

[0047] Features of each image of After identification, 1 and 2 Image number 1 The characteristic is that it is projected onto a common horizontal plane. For more details, see Camera 3 and 1 The determined number on the image 1 Between features rays of light However, it is projected onto the horizontal plane, and as a result, 1 Projection features position PFP1a is obtained. In addition, camera 3 and 2 The determined number on the image 1 Between features rays of light However, it is similarly projected onto the horizontal plane, and as a result, 2 Projection features position PFP1b is obtained. Here, the projection is performed in the vertical direction, thereby rays of light It is worth noting that only the elevation angle changes, while the azimuth angle remains unchanged.

[0048] The 1 and 2 Projection features position PFP1a, PFP1b decision After that, the aforementioned 1 and 2 Projection features position Based on PFP1a and PFP1b, 1 The perpendicular bisector B1 is set. As shown in Figure 2, 1 The perpendicular bisector B1 is the 1 and 2 Projection features position It is perpendicular to the connecting line between PFP1a and PFP1b. In addition, 1 The perpendicular bisector B1 passes through the center of the aforementioned connecting line. 1 The perpendicular bisector B1 is the reference axis, which is the vertical axis of the vehicle LAV in this embodiment. line It intersects with the aforementioned. 1 Perpendicular bisector B1 and reference axis line The intersection point gives the point of rotation around which the trailer rotates. More specifically, the intersection point is Tou Provide the position of ball 4.

[0049] The aforementioned 1 Based on the perpendicular bisector B1, 1 Angle estimation α1 calculation It will be done. 1 Angle estimation α1 teeth , the 1 Projection features position PFP1a and the 1 Intersection with perpendicular bisector B1 and reference axis line The one connecting to 1 line L1 and the 2 Projection features position PFP1b and the 1 Perpendicular bisector B1 and reference axis line The intersection point connecting the two points 2 wires Angle between L2 This is about This intersection is, Tou It may also indicate the position of ball 4. More specifically, the position on the horizontal plane. 1 The angle estimate α1 is the projection onto the horizontal plane. 1The number in the image 1 Features place and projected onto the horizontal plane 2 The number in the image 1 Features place Between ( Tou (This is the position of ball 4) 1 On the horizontal plane of the trailer 2 centered on intersection IP1 Turning It characterizes the angle.

[0050] The 1 The angle estimate α1 represents the yaw angle YA of trailer 2 centered on the actual rotation point.

[0051] Figure 3 shows the yaw angle YA. In the settings The first of the trailer 2 photographed at different points in time (when trailer 2 has a different yaw angle YA relative to towing vehicle 1). 1 and 2 Features F1, F2 use Figure 2 shows an embodiment similar to that shown in Figure 2. Multiple different features may be identifiable in the image captured by camera 3. As shown in Figure 3, these features are identified at different angular positions with respect to a fixed point of vehicle 1. 1 Features: square, 2 The characteristic feature is that it is drawn as a triangle. The aforementioned fixed point is camera 3 place , or Tou Ball 4 place in It's okay to have it. .

[0052] In Figure 3, (Indicated by PFP1a and PFP2a, and associated with solid rays connecting camera 3 and the features) The 1 and 2 The upper row of features is 、 The 1 image inside Identified, (Indicated by PFP1b and PFP2b, associated with dashed rays connecting features with camera 3) The 1 and 2 Features: The lower row of F1 and F2 teeth, The first at different points in time 2 image inside It is identified.

[0053] Yo - Corner YA decisionThis is carried out in the same manner as the embodiment in Figure 2. The main difference is the two angle estimates α1 and α2. is set The trailer's yaw angle is determined based on the two angle estimates α1 and α2. but It will be developed. For more details, see Chapter 1 The perpendicular bisector B1 is set, 1 Obtaining the angle estimate α1 but It will be implemented as described above.

[0054] In addition , the 2 The angle estimation α2 is, 3 Projection features position PFP2a and the 4 Projection features position Set PFP2b, 2 To obtain intersection IP2, 2 Set the perpendicular bisector B2, and 3 Projection features position PFP2a and the 4 Projection features position PFP2b 2 It is obtained by connecting to intersection IP2. 3 Projection features position PFP2a is the 1 In the image, 2 The features are projected onto the aforementioned horizontal plane, and 4 Projection features position PFP2b is the 2 In the image, 2 The features are projected onto the aforementioned horizontal plane. 2 Intersection IP2 is, 2 Perpendicular bisector B2 is the reference axis line This is the point where it intersects with the vertical axis of the vehicle LAV in this embodiment. line It may also be a point where it intersects with the other. 2 The angle estimation α2 is, 3 Projection features position The line connecting PFP2a and intersection IP2 1 line And, 4 Projection features position The line connecting PFP2b and intersection IP2 2 wires It is the angle formed by the two.

[0055] Book In this embodiment, not only camera 3 but also Tou Ball 4 is also on the aforementioned vertical axis of vehicle 1. line Because it is on LAV, the reference axis line This is the vertical axis of towing vehicle 1. line LAV is another embodiment So Camera 3 or Tou Ball 4 is on the vertical axis of Vehicle 1. line Lateral direction relative to LAV There is a discrepancy. case, or Vehicle 1 vertical axis line Camera 3 and for LAV Tou If the lateral displacement of ball 4 is different, the reference axis line This is camera 3 and Tou It may be formed by a straight line connecting ball 4.

[0056] Under ideal conditions, 1 Angle estimation α1 and 2 Angle estimation α2 is 、 Equally (α1=α2), and The yaw angle YA should be represented. However, noise and mismatch Therefore, 1 and the 2 The angle estimates α1 and α2 may differ.

[0057] to Three or more features of Leira 2 are visible across multiple images. Decision and It is noteworthy that it is traceable. In addition, preferably, the result of the yaw angle estimation Enhance Therefore, at different points in time More than two An image is taken. This determines the yaw angle. decision Two ways to improve quality more Angle estimation α1 and α2 are set. Regular Cut.

[0058] The third with different values 1 and 2 A statistical scale may be used to determine the yaw angle YA based on the angle estimates α1 and α2. 1 According to the embodiment, two or more The yaw angle YA is calculated using the median of the angle estimates α1 and α2. It is OK to make a decision. According to other embodiments, statistical methods are used, or more The yaw angle YA may be determined based on angle estimates α1 and α2. Statistical methods could include, for example, the RANSAC algorithm (Random Sample Consensus) or the least squares algorithm.

[0059] photograph All visible features on the image ga yo - Corner YA calculation to equally Suitable That's not the case. . Reduces the complexity of calculations and Robust For this reason, it is very close to the actual yaw angle. Turning Features with angles are selected, and furthermore, the yaw angle YA decision Used for [specific purpose]. Features of Selection To choose In future images, the actual yaw angle will be shown. A predetermined surrounding Within the frame Turning Only the features that give angles α1 and α2 are tracked. May you leave a trace? For example, the frame is defined by an upper limit and a lower limit. Defined and well The upper and lower limits are set by the actual yaw angle. surround angle frame Definition For example, the frame is 2° from 10°, particularly preferably 3° from 5° You can straddle it. The last two or more In the yaw angle determination step, within the frame Turning All the features that give an angle are, To be photographed image inside Further tracking is being pursued.

[0060] In multiple images, a specific trailer of When tracking features, the sample of the features is distributed on the circular segment due to the movement of Trailer 2. Place Cut. before The center of the circular segment is, Tou Ball 4 place This represents a specific trailer. of By tracking features across multiple images, Tou Ball 4 place Lead Go out Cut.

[0061] To reduce noise, Tou Ball 4 place of decision and multiple trailers tracked over a certain period of time across multiple images of Consider the characteristics You can be considerate. Each trailer has its own unique characteristics, of It may also correspond to circular segments with estimations. multiple center of By applying statistical methods to estimation, Tou The actual Ball 4 place to Open The statistical method may be, for example, the RANSAC algorithm or the least squares algorithm.

[0062] Figure 4 shows the reference axis in the upper section. line Determine the rotation point geometrically without using [a specific method / tool]. ru, sara This shows an embodiment.

[0063] times The turning point is characterized by being at a different angular position relative to the towing vehicle 1. 3 Based on the image, at least 2 Book It is set by unfolding the perpendicular bisectors B1 and B2. In Figure 4, 1 Projection features position As mentioned above, PFP1a is projected onto a common horizontal plane. 1 Features contained in the image Regarding Similarly, 2 Projection features position PFP1b is the 2 Features contained in the image This is about , the 3 Projection features position PFP1c is the first 3 Features contained in the image Regarding ru.

[0064] The 1 The perpendicular bisector B1 is the 1 and 2 Projection features position Connecting cable between PFP1a and PFP1b Regarding It is. The 2 The perpendicular bisector B2 is the 1 and the 3 Projection feature position Connects the connection lines of PFP1a and PFP1c finger It is. The 1 and the 2 The intersection of the perpendicular bisectors B1 and B2 indicates the rotation point of the trailer 2 , that is, indicating the position of toe ball 4. Intersection point IP Definition It is.

[0065] As shown in the embodiments of FIGS. 2 and 3, the intersection point IP is the decision of the usable .

[0066] FIG. 5 shows a block diagram showing the steps of a method for obtaining the yaw angle YA of the trailer 2 with respect to the longitudinal axis line LAV of the towing vehicle 1.

[0067] As the 1 first step, at least the 1 and the 2 images of the trailer are taken using a camera (S10).

[0068] [[ID=4,7]]After image taking, the 1 and the <00,00923>visible of, features of the trailer at least one are decision identified (S11).

[0069] After the decision features, by feature projection, the 1 projection feature position and the 2 projection feature position are set (S12).

[0070] After of feature 1 projection, the

[0071] As the 1 perpendicular bisector of is set, after the1 The perpendicular bisector and 、 the reference axis Line or the intersection of the perpendicular bisector 1 is developed (S14).

[0072] Finally, the yaw angle is 1 estimated based on the angle estimation calculation (S15).

[0073] It should be noted that the above description and drawings merely illustrate the principles of the proposed invention. Those skilled in the art will be able to implement various configurations that embody the principles of the present invention even if not explicitly described or in this specification.

Explanation of Reference Numerals

[0074] 1 Vehicle 2 Trailer 3 Camera 4 Tou Ball α1 First 1 angle estimation α2 Second 2 angle estimation B1 First 1 perpendicular bisector B2 Second 2 perpendicular bisector PFP1a First 1 projection feature of the feature in the image 1 position PFP1b First 2 projection feature of the feature in the image 1 position PFP1c First 3 projection feature of the feature in the image 1 position PFP2a First 1 projection feature of the feature in the image 2 position PFP2b First 2 projection feature of the feature in the image 2 position IP Intersection point​​​​​ IP1 1 intersection IP2 No. 2 intersection LAT Trailer's Vertical Axis line LAV vehicle vertical axis line R rays of light YA Yaw angle

Claims

1. A method for determining the yaw angle (YA) of a trailer (2) with respect to the longitudinal axis (LAV) of a towing vehicle (1) without knowing the position of the tow ball of the towing vehicle (1) in advance, wherein the method is - Step (S10) of taking at least the first and second images of the trailer (2) at different points in time when the angular position of the trailer (2) relative to the towing vehicle (1) is different, using a camera (3) provided on the towing vehicle (1), such that the orientation of the trailer (2) relative to the towing vehicle (1) is different in at least the first and second images, - A step (S11) of determining at least a first feature of the trailer (2) that is visible in the first and second images, - Step (S12) of projecting the light rays between the camera (3) and the first feature determined on the first image onto a horizontal plane to obtain a first projected feature position (PFP1a), and projecting the light rays between the camera (3) and the first feature determined on the second image onto the horizontal plane to obtain a second projected feature position (PFP1b), - Step (S13) of setting a first perpendicular bisector (B1) between the location of the first projection feature position (PFP1a) and the location of the second projection feature position (PFP1b), - The first step (S14) is to determine the first intersection point (IP1) between the first perpendicular bisector (B1) and the reference axis, - A step (S15) in which the yaw angle (YA) is calculated based on a first angle estimation (α1), wherein the first angle estimation (α1) is the angle between a first line continuous in the horizontal plane from the first projection feature position (PFP1a) to the first intersection (IP1) and a second line continuous from the second projection feature position (PFP1b) to the first intersection (IP1), and a step (S15) in which the yaw angle (YA) is calculated based on a first angle estimation (α1) A method for determining the yaw angle (YA) of a trailer (2) with respect to the longitudinal axis (LAV) of a towing vehicle (1).

2. The method according to claim 1, wherein in the first or second image, the yaw angle (YA) of the trailer (2) relative to the towing vehicle (1) is zero or any known yaw angle (YA) available as a reference angle.

3. - A step of determining a second feature of the trailer (2) that is visible in the first and second images, wherein the second feature is located at a different position on the trailer (2) from the first feature, - The steps of projecting the light rays between the camera (3) and the second feature determined on the first image onto the horizontal plane to obtain a third projected feature position (PFP2a), and projecting the light rays between the camera (3) and the second feature determined on the second image onto the horizontal plane to obtain a fourth projected feature position (PFP2b), - A step of setting a second perpendicular bisector (B2) between the location of the third projection feature position (PFP2a) and the location of the fourth projection feature position (PFP2b), The steps include determining the second intersection point (IP2) of the second perpendicular bisector (B2) and the reference axis, - A step of calculating a second angle estimate (α2), wherein the second angle estimate (α2) is the angle between a first line continuous in the horizontal plane from the third projection feature position (PFP2a) to the second intersection (IP2) and a second line continuous from the fourth projection feature position (PFP2b) to the second intersection (IP2), - A step of calculating the yaw angle (YA) based on the first and second angle estimates (α1, α2) and The method according to claim 1 or 2, further comprising:

4. The method according to claim 3, wherein, in addition to the first and second features described above, at least one further feature of the trailer (2) is used in the calculation of the yaw angle (YA).

5. The method according to claim 3 or 4, wherein the yaw angle (YA) is calculated by setting a median based on the at least two angle estimates.

6. The method according to any one of claims 3 to 5, wherein the yaw angle (YA) is calculated by setting the mean of the at least two angle estimates, or by using a statistical approach applied to the angle estimates.

7. The step of determining the angle frame is further included. The angle frame includes an upper and lower limit that encloses the yaw angle (YA), and the angle frame determines a set of features that lead to angle estimation within the angle frame. The method according to any one of claims 1 to 6, wherein the set of features determined above is used to calculate the future yaw angle (YA).

8. The method according to any one of claims 3 to 6, or the method according to claim 7, which incorporates any one of claims 3 to 6, wherein camera calibration information is used to convert the position of at least one of the first feature and the second feature from the local domain of the image to the local domain of the towing vehicle (1).

9. The method according to any one of claims 1 to 8, wherein if the camera (3) and the tow ball (4) of the towing vehicle (1) are arranged in a vertically oriented plane having the vertical axis (LAV) of the towing vehicle (1), the reference axis is the vertical axis (LAV) of the towing vehicle (1).

10. The method according to any one of claims 1 to 9, wherein if at least one of the camera (3) and the tow ball (4) of the towing vehicle (1) is misaligned laterally with respect to the vertical axis (LAV) of the towing vehicle (1), the reference axis is a straight line continuous between the camera (3) and the tow ball (4).

11. The method according to any one of claims 1 to 10, wherein the camera (3) is a rear-view camera of the towing vehicle (1).

12. A system for determining the yaw angle (YA) of a trailer (2) with respect to the longitudinal axis (LAV) of a towing vehicle (1) without knowing the position of the tow ball of the towing vehicle (1) in advance, wherein the system comprises a camera (3) installed on the towing vehicle (1) for capturing images of the trailer (2), and a processing entity for processing the captured images, and the system - Step (S10) of using a camera (3) provided on the towing vehicle (1) to capture at least first and second images of the trailer (2) at different points in time when the angular position of the trailer (2) relative to the towing vehicle (1) is different, such that the orientation of the trailer (2) relative to the towing vehicle (1) is different in at least the first and second images, - A step (S11) of determining at least a first feature of the trailer (2) that is visible in the first and second images, - Step (S12) of projecting the light rays between the camera (3) and the first feature determined on the first image onto a horizontal plane to obtain a first projected feature position (PFP1a), and projecting the light rays between the camera (3) and the first feature determined on the second image onto the horizontal plane to obtain a second projected feature position (PFP1b), - Step (S13) of setting a first perpendicular bisector (B1) between the location of the first projection feature position (PFP1a) and the location of the second projection feature position (PFP1b), - A step (S14) to determine the first intersection point (IP1) between the first perpendicular bisector (B1) and the reference axis, - A step (S15) in which the yaw angle (YA) is calculated based on a first angle estimation (α1), wherein the first angle estimation (α1) is the angle between a first line continuous from the first projection feature position (PFP1a) to the first intersection (IP1) and a second line continuous from the second projection feature position (PFP1b) to the first intersection (IP1) in the horizontal plane, and A system is further configured to determine the yaw angle (YA) of the trailer (2) with respect to the longitudinal axis (LAV) of the towing vehicle (1).

13. A vehicle comprising the system described in claim 12.

14. A method for determining the yaw angle (YA) of a trailer (2) with respect to the longitudinal axis (LAV) of a towing vehicle (1) without knowing the position of the tow ball of the towing vehicle (1) in advance, wherein the method is - Step (S10') of using a camera (3) provided on the towing vehicle (1) to capture at least the first, second, and third images of the trailer (2) at different points in time when the angular position of the trailer (2) relative to the towing vehicle (1) is different, such that the orientation of the trailer (2) relative to the towing vehicle (1) is different in at least the first, second, and third images, - A step (S11') of determining at least a first feature of the trailer (2) that is visible in the first, second and third images, - The steps (S12') are: - Projecting the light rays between the camera (3) and the first feature determined on the first image onto a horizontal plane to obtain a first projected feature position (PFP1a); projecting the light rays between the camera (3) and the first feature determined on the second image onto the horizontal plane to obtain a second projected feature position (PFP1b); and projecting the light rays between the camera (3) and the first feature determined on the third image onto the horizontal plane to obtain a third projected feature position (PFP1c). - A step (S13') of setting a first perpendicular bisector (B1) between the location of the first projection feature position (PFP1a) and the location of the second projection feature position (PFP1b), and setting a further perpendicular bisector between the location of the first projection feature position (PFP1a) and the location of the third projection feature position (PFP1c), - A step (S14') to determine the first intersection point (IP1) of the first perpendicular bisector (B1) and the further perpendicular bisector, - A step (S15) in which the yaw angle (YA) is calculated based on a first angle estimation (α1), wherein the first angle estimation (α1) is the angle between a first line continuous in the horizontal plane from the first projection feature position (PFP1a) to the first intersection (IP1) and a second line continuous from the second projection feature position (PFP1b) to the first intersection (IP1), and a step (S15) in which the yaw angle (YA) is calculated based on a first angle estimation (α1) A method for determining the yaw angle (YA) of a trailer (2) with respect to the longitudinal axis (LAV) of a towing vehicle (1).

15. The method according to claim 14, wherein in the first or second image, the yaw angle (YA) of the trailer (2) relative to the towing vehicle (1) is zero or any known yaw angle (YA) available as a reference angle.

16. Further comprising the step of determining the angle frame, The angle frame includes an upper and lower limit that encloses the yaw angle (YA), and the angle frame determines a set of features that lead to angle estimation within the angle frame. The method according to claim 14 or 15, wherein the set of characteristics determined above is used to calculate the future yaw angle (YA).

17. The method according to any one of claims 14 to 16, wherein the camera (3) is a rear-view camera of the towing vehicle (1).

18. A system for determining the yaw angle (YA) of a trailer (2) with respect to the longitudinal axis (LAV) of a towing vehicle (1) without knowing in advance the position of the tow ball of the towing vehicle (1), wherein the system comprises a camera (3) provided on the towing vehicle (1) for taking an image of the trailer (2), and a processing entity for processing the captured image, wherein the system - A step (S10') of using a camera (3) provided on the towing vehicle (1) to capture at least first, second, and third images of the trailer (2) at different points in time when the angular position of the trailer (2) relative to the towing vehicle (1) is different, such that the orientation of the trailer (2) relative to the towing vehicle (1) is different in at least the first, second, and third images. - A step (S11') of determining at least a first feature of the trailer (2) that is visible in the first, second and third images, - The steps (S12') are: - Projecting the light rays between the camera (3) and the first feature determined on the first image onto a horizontal plane to obtain a first projected feature position (PFP1a); projecting the light rays between the camera (3) and the first feature determined on the second image onto the horizontal plane to obtain a second projected feature position (PFP1b); and projecting the light rays between the camera (3) and the first feature determined on the third image onto the horizontal plane to obtain a third projected feature position (PFP1c). - A step (S13') of setting a first perpendicular bisector (B1) between the location of the first projection feature position (PFP1a) and the location of the second projection feature position (PFP1b), and setting a further perpendicular bisector between the location of the first projection feature position (PFP1a) and the location of the third projection feature position (PFP1c), - A step (S14') of determining the first intersection point (IP1) of the first perpendicular bisector (B1) and the further perpendicular bisector, - A step (S15) in which the yaw angle (YA) is calculated based on a first angle estimation (α1), wherein the first angle estimation (α1) is the angle between a first line continuous from the first projection feature position (PFP1a) to the first intersection (IP1) and a second line continuous from the second projection feature position (PFP1b) to the first intersection (IP1) in the horizontal plane, and A system is further configured to determine the yaw angle (YA) of the trailer (2) with respect to the longitudinal axis (LAV) of the towing vehicle (1).

19. A vehicle comprising the system described in Claim 18.

Images