Automatic coupling system, automatic coupling method and tractor

Abstract

An automatic hitching system and an automatic hitching method for enabling automatic hitching between a hitching mechanism of a towing vehicle and a towed piece of a trailer, the automatic hitching system comprising a first detection module and a second detection module for detecting a position and an orientation of the towed piece relative to a lifting portion and a hitching portion of the hitching mechanism, the lifting portion for lifting the towed piece, the hitching portion for hitching the towed piece, a processing module, and a control module for controlling a movement of the towing vehicle in actual relative position to align the lifting portion and the hitching portion with the towed piece in a vertical direction, respectively, wherein the control module controls the lifting portion to lift the towed piece when the lifting portion is aligned with the towed piece in the vertical direction, and wherein the control module controls the hitching portion to hitch the towed piece when the hitching portion is aligned with the towed piece in the vertical direction. A towing vehicle comprising the above system is also provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automatic driving, and more particularly to an automatic hitching system, an automatic hitching method and a tractor. BACKGROUND

[0002] In the wharf and airport warehouse, the trailer is usually used to transport goods in batches. One type of trailer includes a tractor (locomotive) and a trailer, the load of the trailer is borne by itself, and is connected with the tractor by a hitching part to be driven by the tractor.

[0003] In actual operation, after the goods have been loaded onto the trailer or the goods have been transported to the destination, the tractor and the trailer need to be hitched and separated (also known as "up and down hitching").

[0004] Currently, such up and down hitching operation is usually completed manually by people, but manual up and down hitching has not a small defect: first, manual up and down hitching requires the cooperation of the driver and the assisting personnel, and the assisting personnel and the driver need to communicate frequently and cooperate tacitly to complete the task, which not only affects the success rate of hitching, but also is a great burden on the cost of using people. Second, in the process of alignment, the misalignment between the trailer and the tractor may cause damage to the equipment or injury to the personnel, so very high safety measures are required in the process of alignment. When alignment fails or accidents occur, how to respond quickly and take emergency measures is also an important challenge.

[0005] In order to avoid the above defects, the automatic hitching technology between the tractor and the trailer is currently being studied, that is, the automatic up and down hitching (including automatic hitching and separation of the trailer) between the tractor and the trailer, which is an important research direction of the fully automatic driving of the tractor.

[0006] It is currently desirable to develop an automatic hitching system or device to meet the above requirements as much as possible. SUMMARY

[0007] In order to solve the problem of heavy manual hitching burden of the trailer and the need for very high safety measures, the present application proposes an automatic hitching system to replace manual hitching, which greatly avoids the burden and the possibility of injury of the operator.

[0008] Specifically, the automatic hitching system is used to realize automatic hitching between a towing vehicle and a trailer, the towing vehicle comprising a towing mechanism, the trailer comprising a towed part, the automatic hitching system comprising a first detection module for detecting the position and pose of the towed part and a lifting part of the towing mechanism, the lifting part being configured to lift the towed part; a second detection module for detecting the position and pose of the towed part and a hitching part of the towing mechanism, the hitching part being configured to hitch with the towed part; a processing module; and a control module for controlling the movement of the towing vehicle and the towing mechanism, wherein the automatic hitching system operates in a first stage and a second stage, wherein the processing module obtains a first actual relative position of the towed part and the lifting part based on the position and pose of the towed part and the lifting part detected by the first detection module and transmits the first actual relative position to the control module, wherein the processing module obtains a second actual relative position of the towed part and the hitching part based on the position and pose of the towed part and the hitching part detected by the second detection module and transmits the second actual relative position to the control module, wherein the control module controls the movement of the towing vehicle to align the lifting part and the hitching part with the towed part in a vertical direction respectively based on the first actual relative position and the second actual relative position respectively, and wherein in the first stage, when the lifting part is aligned with the towed part in the vertical direction, the control module controls the lifting part to lift the towed part, and in the second stage, when the hitching part is aligned with the towed part in the vertical direction, the control module controls the hitching part to hitch with the towed part.

[0009] In embodiments of the present application, the position and pose of the towed part and the lifting part are derived from the first image.

[0010] Preferably, the first image is obtained by a first camera, the first camera being mounted at the top of the towing vehicle such that the field of view of the first camera contains the towed part and the lifting part.

[0011] The first detection module converts the first image taken by the first camera into an aerial view based on perspective transformation to obtain the positional relationship between the towed part and the towing mechanism in a horizontal plane, which is more convenient for subsequent calculation of the actual physical position between the towed part and the towing mechanism.

[0012] Specifically, obtaining the first actual relative position of the towed part and the lifting part further comprises cutting and converting the aerial view into a sub-view; detecting the image of the towed part and the image of the lifting part in the sub-view; calculating the pixel distance from the image of the towed part to the image of the lifting part in the sub-view; calculating the center axis of the image of the towed part in the sub-view; and obtaining the first actual relative position based on the pixel distance and the center axis.

[0013] In embodiments of the present application, the position and pose of the towed part and the hitching part are derived from the second image.

[0014] Preferably, the second image is acquired by a second camera, the second camera being mounted at a side of the towing vehicle such that the field of view of the second camera contains the trailer and the hitching portion.

[0015] In a specific embodiment, the trailer has a trailer head at its front end, the trailer head having a trailer hole, and the hitching portion of the hitching mechanism comprises a hitch pin, wherein the control module is capable of controlling the hitch pin to rise or to lower, and wherein the control module controls the hitching portion to hitch the trailer comprises controlling the hitch pin by the control module to lower through the trailer hole.

[0016] In another specific embodiment, the lifting portion of the hitching mechanism comprises a hitch saddle, wherein the control module is capable of controlling the hitch saddle to rise or to lower, and wherein the control module controls the lifting portion to lift the trailer comprises controlling the hitch saddle by the control module to rise to contact and lift the trailer.

[0017] In a further embodiment, after the second phase, the automatic hitching system operates in a third phase, wherein in the third phase the second detection module detects the position of the hitch pin and the hitch saddle.

[0018] Advantageously, the hitch pin can have a hitch pin positioning nut, wherein the second detection module detects the position of the hitch pin by detecting the position of the hitch pin positioning nut. The hitch saddle can also have a hitch saddle positioning nut, wherein the second detection module detects the position of the hitch saddle by detecting the position of the hitch saddle positioning nut.

[0019] The present application also provides an automatic hitching method for hitching a hitching mechanism of a towing vehicle to a trailer, the method comprising the steps of: detecting the position and attitude of the trailer and a lifting portion of the hitching mechanism; acquiring a first actual relative position of the trailer and the lifting portion based on the position and attitude of the trailer and the lifting portion; controlling the towing vehicle to move based on the first actual relative position to align the lifting portion and the trailer in a vertical direction; controlling the lifting portion to lift the trailer; detecting the position and attitude of the trailer and a hitching portion of the hitching mechanism; acquiring a second actual relative position of the trailer and the hitching portion based on the position and attitude of the trailer and the hitching portion; controlling the towing vehicle to move based on the second actual relative position to align the hitching portion and the trailer in the vertical direction; and controlling the hitching portion to hitch the trailer.

[0020] In another aspect, the present application also provides a vehicle-mounted unit comprising a memory, a processor and a computer program stored on the memory, the processor executing the computer program to implement the steps of the automatic hitching method as described above

[0021] In another aspect, the present application also provides a computer program product storing a program which, when executed by a computer, causes the computer to perform the steps of the automatic hitching method as described above.

[0022] In another aspect, the present application also provides a non-transitory computer- readable storage medium storing a program which, when executed by a computer, causes the computer to perform the steps of the automatic hitching method as described above.

[0023] The present application also provides a towing vehicle for hitching with a trailer, comprising a towing mechanism and an automatic hitching system as described above.

[0024] Additional features and advantages of the described automatic hitching system will be set forth in the detailed description which follows, and will be readily apparent to those skilled in the art from that description, including the detailed description which follows, and the appended drawings, which describe embodiments of the application, including the best mode contemplated for carrying out the application. BRIEF DESCRIPTION OF DRAWINGS

[0025] Reference will now be made to the drawings in which the various aspects and embodiments of the application will be discussed with reference to the following illustrative figures. The drawings are intended to be illustrative, and not limiting of the present application.

[0026] Figure 1 A general flowchart of a method for automatic hitching with an automatic hitching system according to the present application is shown;

[0027] Figure 2 A schematic view of a towing vehicle utilizing an automatic hitching system according to an embodiment of the present application is shown;

[0028] Figure 3 A schematic view of a towing mechanism of a towing vehicle of Figure 2 and a hitching relationship with an exemplary trailer is shown;

[0029] Figure 4 A schematic line drawing of a first image according to an embodiment of the present application is shown;

[0030] Figure 5 A sub-view of an aerial view converted from the first image is shown;

[0031] Figure 6 A schematic view of an obtained pixel sub-view based on the image of Figure 5 is shown;

[0032] Figure 7 A flowchart of an exemplary method for obtaining a first actual relative position of a towed element and a towing mechanism with a processing module is shown;

[0033] Figure 8 a schematic wireframe diagram showing a second image according to an embodiment of the present invention; and

[0034] Figure 9 A hooking flow chart using an automatic hooking system according to an embodiment of the present invention is shown.

[0035] Reference numerals

[0036] 1 tractor

[0037] 2 trailers

[0038] 11First Camera

[0039] 12 Second Camera

[0040] 20 trailer parts

[0041] 20' trailer pixel image

[0042] 21 trailer head

[0043] 22 trailer holes

[0044] 100 traction mechanism

[0045] 110 traction on the base

[0046] 120 traction lower base

[0047] 130 Traction Saddle

[0048] 130' traction saddle pixel image

[0049] 131 Traction saddle positioning nut

[0050] 140 traction bolt

[0051] 141 Drawbar bolt retaining nut. DETAILED DESCRIPTION

[0052] The present invention will be further described in detail below with reference to the accompanying drawings and examples, but they are not intended to limit the present invention in any way.

[0053] As used herein, the terms "upper", "lower", "left", "right", "front", "rear", "vertical", "horizontal" and their derivatives refer to the orientation of components in actual use, for example, the "rear" or "tail" of a vehicle refers to the rear of the vehicle in actual normal driving, and "horizontal plane" refers to a plane parallel to the ground (assuming the ground is flat), and so on. However, it should be understood that unless expressly specified to the contrary, the components may adopt various alternative orientations.

[0054] As used herein, the term "pose" refers to the appearance of an object as detected by a detector. The pose as detected by different detectors can be different. For example, a cylinder in a top view case can appear as a circle, while the same cylinder in a side view case can appear as a rectangle.

[0055] As used herein, the term "center" refers to the center of gravity of the pose of an object as presented in an image, unless explicitly stated otherwise.

[0056] As used herein, the term "distance" refers to the distance between the centers of two objects.

[0057] As used herein, the term "aligned" refers to the centers of two objects being approximately on a straight line in a certain direction, without necessarily being aligned at every point.

[0058] As used herein, the term "actual relative position" refers to the distance, heading angle, etc. of an object in reality.

[0059] As used herein, the term "median axis" refers to a straight line that passes through the vertices of a figure and equally divides the area of the figure.

[0060] Figures 1-9An automatic coupling system is generally shown for enabling automatic coupling between a towing vehicle and a trailer, the towing vehicle comprising a towing mechanism, the trailer comprising a towed part, the automatic coupling system comprising a first detection module for detecting a position and an attitude of the towed part and a lifting part of the towing mechanism, the lifting part being configured to lift the towed part; a second detection module for detecting a position and an attitude of the towed part and a coupling part of the towing mechanism, the coupling part being configured to couple with the towed part; a processing module for obtaining an actual relative position between the towed part and the lifting part based on the position and the attitude of the towed part and the lifting part detected by the first detection module or the second detection module; and a control module for controlling a movement of the towing mechanism of the towing vehicle, wherein the automatic coupling system operates in a first phase and a second phase, wherein the processing module obtains a first actual relative position of the towed part and the lifting part based on the position and the attitude of the towed part and the lifting part detected by the first detection module and transmits the first actual relative position to the control module, wherein the processing module obtains a second actual relative position of the towed part and the coupling part based on the position and the attitude of the towed part and the coupling part detected by the second detection module and transmits the second actual relative position to the control module, wherein the control module controls the movement of the towing vehicle to align the lifting part and the coupling part with the towed part in a vertical direction, respectively, based on the first actual relative position and the second actual relative position, respectively, wherein in the first phase, the control module controls the lifting part to lift the towed part when the lifting part is aligned with the towed part in the vertical direction, and in the second phase, the control module controls the coupling part to couple with the towed part when the coupling part is aligned with the towed part in the vertical direction.

[0061] Referring to Figure 1 a general flowchart of a method of automatic coupling using the automatic coupling system according to the present application is shown. Specifically, the method comprises detecting a position and an attitude of a towed part and a lifting part of a towing mechanism; obtaining a first actual relative position of the towed part and the lifting part based on the position and the attitude of the towed part and the lifting part; controlling a movement of the towing vehicle to align the lifting part and the coupling part with the towed part in a vertical direction based on the first actual relative position; controlling the lifting part to lift the towed part; detecting a position and an attitude of the towed part and a coupling part of the towing mechanism; obtaining a second actual relative position of the towed part and the coupling part based on the position and the attitude of the towed part and the coupling part; controlling a movement of the towing vehicle to align the lifting part and the coupling part with the towed part in a vertical direction based on the second actual relative position; and controlling the coupling part to couple with the towed part.

[0062] Referring to Figures 2-3 which schematically shows a specific structure of a part of a towing vehicle 1 and an exemplary trailer 2 using the automatic coupling system according to an embodiment of the present application, respectively. As Figure 2As shown, the towing vehicle 1 comprises a first camera 11, a second camera 12 and a towing mechanism 100. The first camera 11, also referred to as a roof camera, takes a position and posture of the towed part 20 (see Figure 4 ) of the trailer 2 at a distance in a substantially downward angle. The second camera 12, also referred to as a side camera, takes a position and posture of the towed part 20 (see Figure 4 ) of the trailer 2 close to the towing mechanism 100 in a substantially side angle.

[0063] As shown in Figure 3 , the towing mechanism 100 comprises a towing upper base 110, a towing lower base 120, a towing saddle 130 and a towing pin 140. The towing upper base 110 and the towing lower base 120 are fixedly installed at the tail of the towing vehicle, the two bases are arranged substantially in parallel and are spaced apart in the vertical direction to define a receiving space for receiving a part of the towed part 20 (see Figure 4 ) of the trailer, in which space the towing upper base 110 and the towing lower base 120 restrict the movement of the towed part 20 in the vertical direction. The towing saddle 130 can be threadedly connected with the towing upper base 110 and the towing lower base 120 by, for example, a lead screw, and can be driven to ascend or descend by, for example, a stepping motor. The towing saddle 130 is used to lift the towed part 20 to the height between the towing upper base 110 and the towing lower base 120, so as to receive a part of the towed part 20 between the towing upper base 110 and the towing lower base 120. The towing pin 140 is installed at the tail of the towing vehicle, above and through the towing upper base 110, which can be driven to perform the action of ascending or descending by, for example, a hydraulic mechanism, for which purpose the towing upper base 110 and the towing lower base 120 can both have through holes aligned in the vertical direction for the towing pin 140 to descend through the towing upper base 110 and the towing lower base 120. Preferably, the towing saddle 130 has an opening aligned with the through hole of the towing lower base 120 in the vertical direction, so as to facilitate the descent of the towing pin 140 through the opening of the towing saddle 130 without hindering the movement of the towing pin 140.

[0064] Referring to Figure 3 and additionally referring to Figure 4An exemplary trailer 2 can include a hitch 20, which can have a hitch head 21 at the front end of the hitch 20, the hitch head 21 having a hitch hole 22, which can be aligned with the tow pin when the hitch head 21 of the hitch 20 is received between the upper tow base 110 and the lower tow base 120 to enable the tow pin to sequentially pass through the through hole of the upper tow base, the hitch hole 22 and the through hole of the lower tow base 120, thereby achieving the hitching of the towing vehicle to the trailer 2. The hitch 20 is in a generally triangular shape and is pivotably connected to the trailer 2 so that the hitch 20 can be pivoted upward or downward. It should be understood that the hitch 20 shown in the figure is only an example and is not limiting.

[0065] The stages experienced by the automatic hitching system and the specific operation manner in these stages will be described in detail below. The automatic hitching system operates in the first stage, the second stage and the optional third stage, and completes the automatic hitching between the towing vehicle and the trailer after experiencing these stages.

[0066] First stage

[0067] In the first stage, the automatic hitching system first detects the position and posture of the hitch and the lifting portion of the towing mechanism by using the first detection module. The position and posture of the hitch 20 and the lifting portion of the towing mechanism 100 are derived from the first image as shown in Figure 4 In this context, the lifting portion of the towing mechanism 100 is the towing saddle 130 (see Figure 3 ).

[0068] Preferably, as mentioned above, the first image is acquired by the first camera, which is installed at the top of the towing vehicle so that the angle of view of the first camera contains the hitch 20 and the lifting portion of the towing mechanism 100. In combination with Figure 2 and 4 It is shown that the first camera 11 is positioned at the middle of the rear edge of the roof of the towing vehicle 1, but it can also deviate from the middle of the rear edge. In another embodiment, the first camera 11 can also not be installed at the rear edge of the roof of the towing vehicle 1, as long as the shooting angle of the first camera 11 can meet the requirement of including the towing mechanism 100 and the hitch in the captured image. It should be understood that although the first camera is used to acquire the image in the embodiment of the present application, other ways can also be used to acquire the specific position and posture of the hitch and the lifting portion, such as using laser radar to scan.

[0069] In practice, it is difficult for the first camera to take a top shot (i.e. the shooting direction is perpendicular to the ground) due to the position and the angle of view. Therefore, in the preferred embodiment, the first detection module first converts the image taken by the first camera into a bird’s eye view (BEV view) based on perspective transformation to obtain the positional relationship between the trailer 20 and the hitch saddle 130 on the horizontal plane, which is more convenient for subsequent calculation of the actual physical position between the trailer 20 and the hitch saddle 130. The conversion of the image into a bird’s eye view can be performed by a machine learning software library such as OpenCV.

[0070] After taking the overhead image of the trailer and the lifting part of the hitch mechanism by the first camera and converting the overhead image into a bird’s eye view based on perspective transformation, the present application gives a specific method to obtain the actual physical position between the trailer and the hitch mechanism. First, the bird’s eye view is cut and converted into a sub-view as shown in Figure 5 The first detection module detects the image of the trailer and the image of the lifting part in the sub-view using the algorithm model of target detection. In the above steps, cutting and converting into a sub-view includes cutting a square sub-view with the center of the bird’s eye view as the reference point, which helps the subsequent image segmentation and target detection of the algorithm model. The sub-view can be, for example, a 640*640 pixel or other size sub-view, depending on the algorithm model used. Regarding the detection step, the Yolo series algorithm model (e.g. Yolo v5) can be used to train and detect the feature parts of the trailer and the lifting part, such as the triangular frame and the trailer head (the trailer head has a red-painted ring part) of the trailer and the hitch saddle (black), thereby outputting the horizontal cross-sectional pixel images of both. Figure 6 The horizontal cross-sectional pixel images of the trailer and the hitch saddle detected using the Yolo v5 algorithm model are shown. The roughly triangular shape above is the trailer pixel image 20’, and the black rectangle below can be regarded as the hitch saddle pixel image 130’ (since it is a bird’s eye view, the horizontal cross-section of the hitch saddle covers other parts).

[0071] It should be understood that the first detection module can not be limited to detecting the feature parts of the trailer and the lifting part. For example, in the case where the hitch mechanism is the same, the position of the hitch saddle is fixed, so only the feature parts of the trailer can be detected to determine the pixel image of the trailer. In addition, since the position of the first camera and the size of the sub-view are fixed, the algorithm model can omit the scaling step to further optimize the code. Additionally, the colors of the feature parts of the trailer and the lifting part, such as the trailer head and the hitch saddle, are not limited to the colors described.

[0072] Continuing to refer to Figure 6, the processing module will obtain the first actual relative position of the trailer and the hitch saddle based on the position and pose between the trailer and the hitch saddle detected by the first detection module. The processing module can use an algorithm model such as Yolo v5 as described above, or also use a computer program programmed in other languages to calculate the distance between the trailer image 20' and the hitch saddle image 130'. Specifically, after the first detection module transmits the data of the trailer pixel image 20' and the hitch saddle pixel image 130' to the processing module, the processing module will calculate the pixel distance from the trailer to the lifting part, for example, the pixel distance from the center of the trailer hole to the center of the hitch saddle. Further, the processing module calculates the center axis of the feature part of the trailer. Specifically, in the case of a substantially triangular trailer, the lower vertex of the trailer pixel image 20' is taken as the vertex of the center axis, and then an initial center axis is drawn. The areas of the triangles on both sides of the center axis are calculated and the center axis is adjusted until the areas of the triangles on both sides of the center axis are the same. The obtained center axis vector represents the heading angle of the trailer, which can indicate the skew angle between the towing vehicle and the trailer. Figure 6 The lower vertex of the trailer pixel image 20' is taken as the vertex of the center axis, and then an initial center axis is drawn. The areas of the triangles on both sides of the center axis are calculated and the center axis is adjusted until the areas of the triangles on both sides of the center axis are the same. The obtained center axis vector represents the heading angle of the trailer, which can indicate the skew angle between the towing vehicle and the trailer.

[0073] Then, the processing module obtains the first actual relative position based on the pixel distance and the center axis, where the first actual relative position mainly includes the actual physical distance from the center of the trailer hole to the center of the hitch saddle on the horizontal plane and the heading angle of the trailer on the horizontal plane. In actual operation, since the position of the first camera is fixed and the trailer and the hitch saddle do not move when the first detection module detects, the pixel distance can be directly converted to the actual distance by the calibration coefficient; and since the processing module processes based on the bird's eye view, the heading angle can be directly used to indicate the skew angle between the towing vehicle and the trailer.

[0074] After that, the processing module transmits the first actual relative position (actual distance and skew angle) to the control module. The control module controls the towing vehicle to move, such as steering and reversing, to align the trailer with the lifting part (for example, the center of the trailer hole and the center of the hitch saddle) in the vertical direction, and complete the first stage.

[0075] In actual operation, the alignment of the center of the trailer hole and the center of the hitch saddle can be judged in the following way: since the position of the first camera and the size of the sub-view are fixed, a target coordinate can be set in the pixel image in Figure 6 as the coordinate of the center of the hitch saddle. When the coordinate of the center of the trailer hole is equal to the target coordinate, it can be considered that the center of the trailer hole coincides with the center of the hitch saddle, at which time it is considered that the alignment is completed.

[0076] Figure 7The step of obtaining the first actual relative position of the towed part with respect to the towing part using the automatic hitching system is summarized as follows.

[0077] Second stage

[0078] After the first stage is completed, the towing part and the lifting part (e.g. the center of the towing hole and the center of the towing saddle) are aligned in the vertical direction, and then the control module can control the towing saddle to rise to contact and lift the towing part into Figure 8 the state shown.

[0079] As shown in Figure 8 , since the towing saddle 130 moves in the vertical direction and the towing part 20 rotates, the distance between the towing head 21 of the towing part 20 and the towing pin 140 in the horizontal plane changes after the towing part 20 is lifted, so it is necessary to re-detect the position and attitude of the two to determine the actual distance.

[0080] In the second stage, the automatic hitching system first detects the position and attitude of the towing part and the hitching part of the towing mechanism using the second detection module. The position and attitude of the towing part 20 and the hitching part of the towing mechanism 100 are derived from the second image as shown in Figure 8 . In this context, the hitching part of the towing mechanism 100 is the towing pin 140 (see Figure 3 ).

[0081] Similarly to the first detection module, the second detection module detects the position and attitude of the towing part 20 and the towing pin 140 of the towing mechanism 100 based on the image captured by the second camera. In combination with Figure 2 and 8 , the second camera 12 is positioned on the side of the rear of the towing vehicle 1, and its shooting angle can meet the requirement of including the towing pin 140 of the towing mechanism 100 and the towing part 20 in the captured image. The reason for using two cameras with different shooting angles is that the first camera 11 at the rear edge of the roof can better capture the profile of the towing part 20 to facilitate subsequent relative position calculation, but the installation position of the first camera 11 makes it difficult to clearly capture the towing pin of the towing mechanism 100, so it is difficult to know the positional relationship between the towing part and the towing pin and thus the dropping timing of the towing pin, so the second camera 12 is added to help obtain the positional relationship between the towing part and the towing pin to facilitate automatic hitching. It should also be understood that although a second camera is used to obtain the second image in the embodiment of the present application, other ways can also be used to obtain the specific position and attitude of the towing part and the hitching part, such as using a laser radar to scan.

[0082] Unlike the first detection module, since the second camera is relatively close to the towed part 20 (specifically the towed head 21) and the towing mechanism 100, the angle of view does not need to be adjusted, and the position and posture of the towed head 21 and the towing pin 140 in the image can be directly cut and recognized, which causes very small errors. Similarly, although the second camera is used to obtain the image in the embodiment of the present application, it should be understood that other ways can be used to obtain the specific posture of the towing mechanism and the towed part, such as using a laser radar to scan.

[0083] Similarly, after detecting the image of the towed head 21 and the towing pin 140, the processing module will calculate the pixel distance from the towed part to the lifting part, such as the horizontal pixel distance from the center of the towed hole 22 to the center of the towing pin 140. Since the position of the second camera is fixed and the towed part 20 and the towing pin 140 do not move when the second detection module detects, the horizontal pixel distance can be directly converted to the second actual relative position of the towed head 21 to the towing pin 140 by a calibration coefficient. Here, the second actual relative position mainly includes the actual horizontal distance.

[0084] Then, the processing module transmits the second actual relative position to the control module. The control module controls the towing vehicle to further reverse based on the second actual relative position, so that the towed hole 22 is aligned with the towing pin 140 in the vertical direction, and then controls the towing pin 140 to descend through the towed hole 22, completing the second stage, i.e. completing the automatic hitching.

[0085] Third stage

[0086] Continuing to refer to Figure 8 In order to ensure the completion of automatic hitching, there is also a subsequent safety check stage, i.e. the third stage. In the preferred embodiment, the towing pin 140 can have a towing pin positioning nut 141, wherein the second detection module detects the position of the towing pin 140, i.e. whether it is in the raised position or the lowered position, by detecting the position of the center of the towing pin positioning nut 141. For example, a picture height threshold in the image can be defined, and the processing module can determine whether the towing pin 140 is in the raised position or the lowered position by determining whether the center of the towing pin positioning nut 141 is higher or lower than the picture height threshold in the vertical direction. After the second stage is completed, i.e. after the towing pin 140 descends through the towed hole 22, if it is detected that the towing pin 140 is in the raised position, the processing module can provide this information to the control module to inform the towing vehicle and the trailer that the hitching has failed (for example, the towing pin 140 is not through the towed hole 22 but is stuck by the towed head 21), in order to improve the safety of the automatic hitching system.

[0087] Furthermore, preferably, the hitch saddle 130 also needs to be lowered to the original position for the next hitching in the third stage, for this purpose, the hitch saddle 130 can have a hitch saddle positioning nut 131, wherein the second detection module detects the position of the hitch saddle 130 by detecting the position of the center of the hitch saddle positioning nut 131. The position of the center of the hitch saddle positioning nut 131 can also be determined by the method of defining the picture height threshold. After the second stage is completed, i.e. after the draw pin 140 is lowered through the hitch hole 22, if it is detected that the draw pin 140 is in the raised position, the processing module can provide this information to the control module to prevent the situation where the hitch saddle control signal is issued but not successfully executed, thereby improving the safety of the system.

[0088] Figure 9 The hitching flowchart using the automatic hitching system according to an embodiment of the present application is summarized.

[0089] Specifically, in the first stage, the first detection module captures the overall pose of the towed part and the hitch saddle from a top-down perspective by means of the first camera, i.e. the top camera, corrects the picture based on the camera internal and external parameters, and converts the camera picture to an aerial view based on perspective transformation, and then uses the Yolo series algorithm model to perform image segmentation on the adjusted picture to detect the features of the towed part and the hitch saddle. After calibration (pixel distance corresponds to actual physical distance), the relative physical position information of the towed part image in the picture to the target point of the raised part at the bottom of the picture can be obtained, i.e. the position of the hitch hole relative to the hitch saddle is known. Subsequently, the processing module calculates the pixel distance from the hitch hole to the hitch saddle, and the actual physical distance can be obtained by conversion through the calibration coefficient. Then, the central axis of the towed part (triangular shape) is calculated based on geometric method to obtain the heading angle representing the towed part. Next, the processing module transmits the physical distance and heading angle data to the control module. Finally, the control module controls the movement of the towing vehicle to align the hitch saddle and the hitch hole in the vertical direction and to lift the towed part.

[0090] In the second stage, the image of the second camera, i.e. the side camera, is segmented and detected to detect the position of the hitch hole and calculate the pixel distance between the center of the hitch hole and the center of the draw pin in the picture, and then the actual physical distance from the center of the hitch hole to the center of the draw pin can also be obtained based on the calibration coefficient. Finally, the control module controls the movement of the towing vehicle to align the draw pin and the hitch hole in the vertical direction, and controls the draw pin to be lowered.

[0091] In addition, there is a third stage, detecting the position of the center of the drawbar positioning nut, if the height of the center of the drawbar positioning nut is lower than the picture height threshold, it means that the drawbar is fallen, otherwise it is not. The drawbar saddle positioning nut can also be used to detect whether the position of the drawbar saddle has risen or fallen. This information is provided to the vehicle control system to prevent the situation that the control command is not successfully executed, and to improve the safety of the system.

[0092] The automatic hitching system of the present application has the following technical advantages:

[0093] 1. Completely without human intervention: no need to rely on the experience and skills of the operator, completely realized by software.

[0094] 2. High positioning accuracy: the recognition position error of the towed part is kept within ±2 cm, and the heading angle error is kept within ±2 degrees, which can realize accurate alignment.

[0095] 3. Low hardware cost: only two cameras are needed, and no further modification of the towing vehicle or trailer or other site facilities is needed.

[0096] 4. High adaptability to trailers, no matter the type of trailer, after model training, the position and posture of the towed part can be accurately detected.

[0097] Although the structure and operation method of the present application have been described above in combination with the preferred embodiments, it should be recognized by those skilled in the art that the above examples are only used for illustration, and cannot be regarded as a limitation of the present application. Therefore, modifications and variations can be made to the present application, and these modifications and variations will fall within the scope defined by the appended claims.

Claims

1. An automatic coupling system for automatically coupling a tractor to a trailer, wherein the tractor includes a traction mechanism, and the trailer includes a towing member. The automatic coupling system comprises: a first detection module, the first detection module being configured to detect the position and posture of the towing member and a lifting portion of the traction mechanism, the lifting portion being configured to lift the towing member; a second detection module, the second detection module being configured to detect a position and a posture of a hooking portion between the towing member and the traction mechanism, the hooking portion being configured to be hooked to the towing member; Processing module; as well as a control module for controlling the movement of the tractor and the traction mechanism, The automatic hooking system operates in the first and second stages. wherein the processing module obtains a first actual relative position of the towing member and the lifting part based on the position and posture of the towing member and the lifting part of the traction mechanism detected by the first detection module, and transmits the first actual relative position to the control module; The processing module obtains a second actual relative position of the towing member and the hitch portion based on the position and posture of the towing member and the hitch portion of the traction mechanism detected by the second detection module, and transmits the second actual relative position to the control module. wherein the control module controls the movement of the tractor based on the first actual relative position and the second actual relative position so that the lifting portion and the hitch portion are aligned with the towing member in a vertical direction, wherein, in the first stage, when the lifting portion is aligned with the towing member in the vertical direction, the control module controls the lifting portion to lift the towing member, and Wherein, in the second stage, when the hooking portion and the towing member are aligned in the vertical direction, the control module controls the hooking portion to hook the towing member.

2. The automatic hooking system according to claim 1, wherein: The position and posture of the towing member and the lifting part are derived from a first image, wherein the first image is acquired by a first camera mounted on top of the tractor such that the viewing angle of the first camera includes the towing member and the lifting part.

3. The automatic hooking system according to claim 2, characterized in that: The first detection module converts the first image captured by the first camera into a bird's-eye view based on perspective transformation.

4. The automatic hooking system according to claim 3, characterized in that: The first detection module is further configured to: Cutting and converting the bird's-eye view into sub-views; and detecting the images of the towing member and the lifting part in the subview, and Wherein, the processing module is further configured to: Calculating a pixel distance between the image of the trailer and the image of the lifting portion in the subview; Calculating a central axis of the image of the trailer in the subview; and The first actual relative position is obtained based on the pixel distance and the central axis.

5. The automatic hooking system according to claim 1, wherein: The position and posture of the towing member and the hitch portion are derived from a second image, wherein the second image is acquired by a second camera mounted on a side of the tractor such that a viewing angle of the second camera includes the towing member and the hitch portion.

6. The automatic hooking system according to claim 1, wherein: The towing member has a towing head at its front end, the towing head has a towing hole, and the hooking part of the towing mechanism includes a towing bolt, The control module can control the traction bolt to rise or fall, and The control module controlling the hooking portion to hook the towing member includes controlling the towing bolt to descend through the towing hole through the control module.

7. The automatic hooking system according to claim 6, characterized in that: The lifting part of the traction mechanism includes a traction saddle, The control module can control the traction saddle to rise or fall, and The control module controlling the lifting portion to lift the towing component includes the control module controlling the towing saddle to rise to contact and lift the towing component.

8. The automatic hooking system according to claim 7, wherein: The second detection module is further configured to detect positions of the towing hitch and the towing saddle after the towing hitch is lowered through the towing hole.

9. The automatic hooking system according to claim 8, characterized in that: The traction bolt has a traction bolt positioning nut, wherein the second detection module detects the position of the traction bolt by detecting the position of the traction bolt positioning nut, and / or The traction saddle has a traction saddle positioning nut, and the second detection module detects the position of the traction saddle by detecting the position of the traction saddle positioning nut.

10. An automatic coupling method for coupling a traction mechanism of a tractor to a towing member, the method comprising the following steps: detecting the position and posture of the towing member and the lifting portion of the traction mechanism; Acquire a first actual relative position of the towing member and the lifting part based on the position and posture of the towing member and the lifting part; controlling the movement of the tractor based on the first actual relative position to align the lifting portion with the hitch in a vertical direction; controlling the lifting portion to lift the towing member; detecting the position and posture of the hooking portion between the towing member and the traction mechanism; Acquiring a second actual relative position of the towing member and the hitch portion based on the position and posture of the towing member and the hitch portion; controlling the tractor to move based on the second actual relative position so that the hitch portion is aligned with the tow member in a vertical direction; as well as The hooking portion is controlled to hook the towing member.

11. The automatic attachment method according to claim 10, wherein: The step of detecting the position and posture of the towing member and the lifting portion of the traction mechanism includes: Acquire a first image including the towing member and the lifting portion; converting the first image into a bird's-eye view based on a perspective transformation; Cutting and converting the bird's-eye view into sub-views; and Images of the towing part and the lifting part are detected in the sub-view.

12. The automatic attachment method according to claim 11, wherein: The step of obtaining a first actual relative position between the towing member and the lifting portion includes: Calculating a pixel distance between the image of the trailer and the image of the lifting portion in the subview; Calculating a central axis of the image of the trailer in the subview; and The first actual relative position is obtained based on the pixel distance and the central axis.

13. The automatic attachment method according to claim 10, wherein: After the step of controlling the hitch portion to hitch the towing member, the automatic hitch method further includes detecting positions of the lifting portion and the hitch portion.

14. A vehicle-mounted unit comprising a memory, a processor, and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the automatic mounting method according to claim 10.

15. A computer program product storing a program, which, when executed by a computer, causes the computer to execute the steps of the automatic mounting method according to claim 10. 16 . A non-transitory computer-readable storage medium storing a program executed by a computer, the program causing the computer to execute the steps of the automatic mounting method according to claim 10 .

17. A tractor for coupling with a trailer, the tractor comprising a traction mechanism and the automatic coupling system according to claims 1 to 9.

Images