Intelligent automatic docking device and method based on monocular vision

Abstract

The invention discloses an intelligent automatic docking device and method based on monocular vision, the device comprises a docking identifier, a high-definition camera, an image processing unit and a docking control unit, the docking identifier is arranged on a plane where a docking port of to-be-replenished equipment is located, and is used as a visual ranging and direction finding marker; the high-definition camera is arranged on one side of a replenishment port of replenishment equipment and is used for acquiring an image picture of a plane where a docking port of to-be-replenished equipment is located and sending the image picture to the image processing unit; the image processing unit is arranged on the supply equipment and is used for processing and analyzing the images, calculating the distance between the high-definition camera and the plane where the docking port is located and determining the normal included angle between the optical axis of the high-definition camera and the plane where the docking port is located; and the docking control unit is used for generating a docking control strategy according to the spatial position relationship between the replenishment equipment and the equipment to be replenished and controlling the action of the replenishment equipment during docking.

Description

Technical Field

[0001] The present invention relates to the technical field of docking, and in particular to an intelligent automatic docking device and method based on monocular vision. Background Art

[0002] The premise of rapid replenishment is to quickly and accurately complete the establishment of the replenishment channel. The establishment of the replenishment channel requires the position and angle of the replenishment equipment and the equipment to be replenished to be aligned. In the automatic replenishment operation of traditional replenishment equipment, the alignment of the replenishment equipment and the equipment to be replenished is mostly achieved by manual command or video assistance. The automation level is low, the speed is slow, and the technical requirements for the driver are high. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent automatic docking device and method based on monocular vision, so as to solve the problems of low replenishment efficiency and the like existing in the prior art.

[0004] In order to achieve the above tasks, the present invention adopts the following technical solutions:

[0005] An intelligent automatic docking device based on monocular vision, comprising: a docking mark, a high-definition camera, an image processing unit and a docking control unit, wherein:

[0006] The docking mark is arranged on the plane where the docking port of the equipment to be supplied is located, and is used as a visual distance and direction finding marker; the high-definition camera is arranged on one side of the supply port of the supply equipment, and is used to collect the image of the plane where the docking port of the equipment to be supplied is located and send it to the image processing unit; the image processing unit is arranged on the supply equipment, and is used to process and analyze the image, calculate the distance between the high-definition camera and the plane where the docking port is located, and determine the angle between the optical axis of the high-definition camera and the normal of the plane where the docking port is located; the docking control unit is used to generate a docking control strategy and action control of the supply equipment during docking according to the spatial position relationship between the supply equipment and the equipment to be supplied.

[0007] Furthermore, the docking mark is arranged on the plane where the docking port of the equipment to be supplied is located, specifically outside the docking port; the docking mark is centered on the docking port, and at least one circle is set outside the docking port, and each circle of docking marks is a square mark composed of a pair of horizontal marks and a pair of vertical marks.

[0008] Furthermore, when multiple circles of docking marks are provided, the horizontal mark and the vertical mark of the inner circle docking mark are half of the adjacent outer circle docking mark.

[0009] Furthermore, the high-definition camera is arranged near the supply port at the tail of the equipment to be supplied, and its field of view should be able to fully capture the docking mark and the docking port.

[0010] An intelligent automatic docking method based on monocular vision, comprising:

[0011] Step 1: Before the docking operation, the supply equipment is located at one side of the rear of the equipment to be supplied, the tails of the supply equipment and the equipment to be supplied are at a preset distance, and the longitudinal axes of the supply equipment and the equipment to be supplied do not overlap; the position of the equipment to be supplied is adjusted so that the docking port of the equipment to be supplied appears in the field of view of the high-definition camera;

[0012] Step 2: Arrange the docking mark around the docking port of the equipment to be supplied; the docking mark is centered on the docking port, and is composed of a pair of horizontal marks and a pair of vertical marks to form a square mark, and the docking mark is arranged in multiple circles;

[0013] Step 3, using a high-definition camera to collect the image of the plane where the replenishment equipment interface is located, performing distortion correction and image optimization processing on the image, and then transmitting it to the image processing unit;

[0014] Step 4: The image processing unit identifies multiple circles of docking marks in the image, selects one circle of docking marks as the identification mark, and calculates the spatial position relationship between the supply equipment and the equipment to be supplied by obtaining the coordinate position information of the identification mark, and obtains the distance and angle between the supply equipment and the equipment to be supplied;

[0015] Step 5: The image processing unit sends the calculated distance and angle between the supply equipment and the equipment to be supplied to the docking control unit; the docking control unit formulates a docking control strategy according to the distance and angle and performs action control during docking.

[0016] Furthermore, the step of calculating the spatial position relationship between the replenishment equipment and the equipment to be replenished by acquiring the coordinate position information of the identification mark includes:

[0017] For the left vertical mark in the identification mark, its length is L1, and its upper and lower ends are at the vertical coordinates Y1 and Y2 in the image screen. Then the height angle A corresponding to the left vertical mark is:

[0018] A=(Y2-Y1)β / H ppi

[0019] Where β is the vertical field of view of the HD camera, H ppi It is the vertical resolution of the HD camera image;

[0020] The distance S between the HD camera and the vertical sign on the left is:

[0021] S=L1 / tan(A)

[0022] Based on the distance S between the high-definition camera and the vertical mark on the left, determine the angle K between the plane where the docking interface is located and the optical axis of the high-definition camera.

[0023] Further, the determining of the angle K between the plane where the docking interface is located and the optical axis of the high-definition camera based on the distance S between the high-definition camera and the left vertical mark includes:

[0024] When the two vertical marks in the identification mark are located on the left and right sides of the center of the image:

[0025] Calculate the height angle A corresponding to the two vertical marks on the left and right sides of the identification mark n and A f , and the distance S between the HD camera and the two vertical markers on the left and right sides n and S f ; It is known that the length of the horizontal mark in the identification mark is L2, then the angle K between the plane where the interface is located and the camera optical axis is:

[0026] K=sin -1 ((S n ×sin(A n )+S f ×sin(A f )) / L2)

[0027] When the two vertical marks in the identification mark are located on the same side of the center of the image screen:

[0028] Note that the horizontal coordinates of the two vertical marks on the left and right sides of the image are X n and X f , the calculation formula of the azimuth corresponding to the vertical mark is:

[0029]

[0030] Where i = n or f; according to the horizontal coordinate X n and X f The corresponding azimuth angle B can be obtained n and B f ;

[0031] Calculate the distance S between the HD camera and the vertical mark on the left n Corresponding angle C:

[0032] C=sin -1 ((sin(|B f |-|B n |)×S n / L2))

[0033] Finally, the angle K between the plane where the docking interface is located and the optical axis of the high-definition camera is obtained as:

[0034] K=180-|B f |-C

[0035] The distance S between the HD camera and the vertical mark on the left, and the angle K between the plane where the docking port is located and the optical axis of the HD camera are used as the distance and angle between the supply equipment and the equipment to be supplied.

[0036] Furthermore, the preset distance is 10m to 30m.

[0037] Furthermore, the image processing unit identifies the multi-circle docking mark in the image through binarization processing, edge extraction and feature judgment.

[0038] Furthermore, during the docking process, the movement speed of the supply equipment is stabilized within the preset speed range. The docking control strategy is divided into two:

[0039] Strategy 1: When the angle K between the supply equipment and the equipment to be supplied is within the preset angle range, the docking control unit controls the supply equipment to approach the equipment to be supplied at a preset speed range, and the direction of the supply equipment is slightly adjusted until the angle K is 0 or close to 0, indicating that the supply equipment and the equipment to be supplied are aligned. At this time, the supply equipment is controlled to move straight backward until the supply port and the docking port are docked;

[0040] Strategy 2: When the angle K between the supply equipment and the equipment to be supplied is outside the preset angle range, control the movement speed and turning direction of the supply equipment to remain stable until the angle K is within the preset angle range and is basically aligned, and then dock according to Strategy 1.

[0041] Furthermore, in Strategy 2, the supply equipment is controlled to move along a specific arc trajectory. When the control strategy is completed, the two goals of accurate positioning and basic angle alignment can be achieved; the radius R of the arc is:

[0042] R=0.5×(S 2 ×cos 2 (K)+4-4×S×cos(K)+S 2 ×sin 2 (K)) / S / sin(K).

[0043] Furthermore, the supply equipment moves along a specific arc trajectory, and the turning radius needs to be controlled; assuming that the supply equipment is rear-wheel driven, the front wheels are responsible for steering, and the front and rear wheelbase is X, then the supply equipment controls the steering angle θ:

[0044] Θ=2*sin -1 (0.5*X / R)

[0045] There is a deviation between the initial longitudinal axis of the supply equipment and the specific arc trajectory, and the deviation angle is also -Θ. Therefore, at the beginning of the movement, the longitudinal axis of the supply equipment should be quickly rotated through -Θ to make the longitudinal axis of the vehicle tangent to the target trajectory, and then the supply equipment should be quickly controlled to control the steering angle to Θ and maintain it; approach the equipment to be supplied along the specific arc trajectory, and when the angle K is within the preset angle range, switch to control strategy 1.

[0046] Furthermore, the preset speed range is 0.1 m / s to 0.3 m / s, and the preset angle range is within ±3°.

[0047] Compared with the prior art, the present invention has the following technical features:

[0048] 1. The present invention adopts a docking mark, a single high-definition camera and an image processing unit to achieve distance measurement and direction measurement, which has a cost advantage over solutions using multiple cameras or a pan-tilt head.

[0049] 2. The docking control unit of the present invention can generate action control strategies and control actions according to the spatial position relationship between the supply equipment and the equipment to be supplied, which can be used for both wheeled equipment and tracked equipment.

[0050] 3. The present invention is applicable to manned equipment to assist manual docking, and is also applicable to unmanned equipment to automatically complete docking. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is composed of the intelligent automatic docking device system of the present invention;

[0052] Figure 2 It is the docking mark of the present invention;

[0053] Figure 3 It is the image picture matched with the high-definition camera of the present invention;

[0054] Figure 4 This is a schematic diagram of angle alignment imaging;

[0055] Figure 5 This is a schematic diagram of the imaging when the angle is not aligned;

[0056] Figure 6 is the target trajectory of control strategy 1;

[0057] Figure 7 is the target trajectory of control strategy 2;

[0058] Figure 8 This is the target trajectory geometry of control strategy 2 of the present invention. DETAILED DESCRIPTION

[0059] For the equipment supply problem, the key is to determine the spatial position relationship between the supplying equipment and the equipment being supplied, mainly including distance and angle; then generate the docking action control strategy based on the spatial position relationship between the supplying equipment and the equipment being supplied, and then measure the docking deviation in real time during the docking process and generate control actions.

[0060] The present invention can be used in the process of ground equipment replenishing ammunition. The supply equipment and the equipment to be supplied are located on the same horizontal ground, the supply port is located at the tail of the equipment to be supplied, and the supply docking requires that the supply port position fits well and the angle is aligned.

[0061] Referring to the accompanying drawings, the present invention provides an intelligent automatic docking device based on monocular vision, comprising: a docking mark, a high-definition camera, an image processing unit and a docking control unit, wherein:

[0062] The docking mark is arranged on the plane where the docking port of the equipment to be supplied is located, and is used as a visual distance and direction finding marker; the high-definition camera is arranged on one side of the supply port of the supply equipment, and is used to collect the image of the plane where the docking port of the equipment to be supplied is located and send it to the image processing unit; the image processing unit is arranged on the supply equipment, and is used to process and analyze the image, calculate the distance between the high-definition camera and the plane where the docking port is located, and determine the angle between the optical axis of the high-definition camera and the normal of the plane where the docking port is located; the docking control unit is used to generate a docking control strategy and action control of the supply equipment during docking according to the spatial position relationship between the supply equipment and the equipment to be supplied.

[0063] See also Figure 2 The docking mark is arranged on the plane where the docking port of the equipment to be supplied is located, specifically arranged outside the docking port; the docking mark is centered on the docking port, and multiple circles are set outside the docking port, each circle of docking marks is a square mark composed of a pair of horizontal marks and a pair of vertical marks; the horizontal mark and vertical mark of the inner circle docking mark are half of the adjacent outer circle docking mark; the color, brightness or shape of the horizontal mark and the vertical mark have a significant contrast with the docking port, which is easy to identify and extract; Figure 2 In the example, the horizontal and vertical logos are set with alternating red and yellow colors.

[0064] The high-definition camera is arranged near the supply port at the tail of the equipment to be supplied, and its field of view should be able to completely capture the docking mark and the docking port; the high-definition camera includes an optical lens, an image sensor, an image processor, a power supply, etc.; the resolution of the high-definition camera is not less than 1920×1080, and the frame frequency is not less than 25fps.

[0065] The process of docking using the docking device of the present invention is as follows:

[0066] Step 1: Before the docking operation, the supply equipment is located on one side behind the equipment to be supplied. The tails of the supply equipment and the equipment to be supplied are 10m to 30m apart. The longitudinal axes of the supply equipment and the equipment to be supplied do not overlap (that is, the initial angles are not aligned). Adjust the position of the equipment to be supplied so that the docking port of the equipment to be supplied appears in the field of view of the high-definition camera.

[0067] Step 2: Arrange the docking marks around the docking port of the equipment to be resupplied. The docking marks are centered on the docking port and are arranged in three circles from the inside to the outside, corresponding to observations at near, medium and far distances. The length of the horizontal and vertical marks in each circle from the inside to the outside is reduced by 1 / 2 to accommodate observations at different distances.

[0068] Step 3: Use a high-definition camera to collect the image of the plane where the replenishment equipment interface is located, perform distortion correction and image optimization on the image, and then transmit it to the image processing unit, such as Figure 3 shown.

[0069] In one embodiment of the present invention, the setting parameters of the high-definition camera are focal length 6mm, fixed focus; resolution L ppi *H ppi It is 1920×1080, the horizontal field of view is 54.5°, and the vertical field of view β is 30.4°.

[0070] Step 4: The image processing unit identifies multiple circles of docking marks in the image through binarization, edge extraction, and feature judgment, selects one circle of docking marks as the identification mark, and calculates the spatial position relationship between the supply equipment and the equipment to be supplied by obtaining the coordinate position information of the identification mark.

[0071] When the distance is far, the outermost circle of docking marks can be used as identification marks to calculate the angle and distance; as the distance gets closer, the middle circle and inner circle of docking marks can be used as identification marks for calculation; in actual application, a clear and complete circle of docking marks can be selected for calculation based on the results of image recognition.

[0072] Taking the outermost circle docking mark as the identification mark for explanation, the calculation method is as follows:

[0073] For the left vertical mark in the identification mark, its length is L1, and the upper and lower ends are at the vertical coordinates Y1 and Y2 in the image. According to the geometric relationship of the pinhole imaging principle of the high-definition camera, it can be known that:

[0074] The elevation angle A (in degrees) corresponding to the vertical mark on the left is:

[0075] A=(Y2-Y1)β / H ppi

[0076] Then the distance S (in meters) between the HD camera and the vertical mark on the left is:

[0077] S=L1 / tan(A)

[0078] like Figure 4 As shown in the figure, when the angles of the supply equipment and the equipment to be supplied are not aligned, there are two situations:

[0079] (1) The two vertical marks in the identification mark are located on the left and right sides of the center of the image.

[0080] Use the above calculation method to obtain the height angle A corresponding to the two vertical marks on the left and right sides of the identification mark n and A f , and the distance S between the HD camera and the two vertical markers on the left and right sides n and S f ; It is known that the length of the horizontal mark in the identification mark is L2, then the angle K (unit: °) between the plane where the docking interface is located and the optical axis of the camera is:

[0081] K=sin -1 ((S n ×sin(A n )+S f ×sin(A f )) / L2)

[0082] (2) The two vertical marks in the identification mark are located on the same side of the center of the image screen.

[0083] Use the above calculation method to obtain the height angle A corresponding to the two vertical marks on the left and right sides of the identification mark n and A f , and the distance S between the HD camera and the two vertical markers on the left and right sides n and S f ; Note that the horizontal coordinates of the two vertical marks on the left and right sides of the image are X n and X f According to the geometric relationship of the pinhole imaging principle of high-definition cameras, the calculation formula for the azimuth angle corresponding to the vertical mark is:

[0084]

[0085] Where i = n or f; according to the horizontal coordinate X n and X f The corresponding azimuth angle B can be obtained n and B f , where |B f |Greater than|B n |.

[0086] Then it is known that S n , S f, the length of the horizontal mark in the identification mark is L2 and the angle corresponding to L2 (|B f |-|B n |, use the triangle sine theorem to calculate S n Corresponding angle C:

[0087] C=sin -1 ((sin(|B f |-|B n |)×S n / L2))

[0088] Finally, the angle K (unit: °) between the plane where the docking interface is located and the optical axis of the high-definition camera is obtained as follows:

[0089] K=180-|B f |-C

[0090] The distance S between the HD camera and the vertical mark on the left, and the angle K between the plane where the docking port is located and the optical axis of the HD camera are used as the distance and angle (angle between the vertical axes) between the supply equipment and the equipment to be supplied.

[0091] Step 5: The image processing unit sends the calculated distance and angle between the supply equipment and the equipment to be supplied to the docking control unit; the docking control unit formulates a docking control strategy according to the distance and angle and performs action control during docking.

[0092] During the docking process, the speed and direction of the supply equipment need to be controlled. To simplify the control, the speed of the supply equipment is stabilized at a relatively low level, 0.1m / s to 0.3m / s, and the direction of movement is the main control variable. The docking control strategy is divided into two types:

[0093] Strategy 1: When the angle K between the supply equipment and the equipment to be supplied is small, for example, within ±3°, the docking control unit controls the supply equipment to slowly approach the equipment to be supplied, and slightly adjusts the direction of the supply equipment until the angle K is 0 or close to 0, indicating that the supply equipment and the equipment to be supplied are aligned. At this time, the supply equipment is controlled to move straight backward until the supply port and the docking port are docked, such as Figure 6 As shown;

[0094] Strategy 2: When the angle K between the supply equipment and the equipment to be supplied is larger, for example, greater than ±3°, the movement speed and the turning direction of the supply equipment are controlled to remain stable until the angle K is within ±3°, which means that the supply equipment is basically aligned. At this time, the distance between the supply equipment and the equipment to be supplied is about 2m, and then docking is performed according to Strategy 1. Figure 7 As shown,

[0095] In strategy 2, the supply equipment is controlled to move along a specific arc trajectory (which can be called the target trajectory). When the control strategy ends, the two targets can be accurately positioned and basically aligned at the angle. The radius R of the arc is:

[0096] R=0.5×(S 2 ×cos 2 (K)+4-4×S×cos(K)+S 2 ×sin 2 (K)) / S / sin(K)

[0097] Depend on Figure 8 As shown, the supply equipment moves along a specific arc trajectory, and the turning radius needs to be controlled; assuming that the supply equipment is rear-wheel driven, the front wheels are responsible for steering, and the front and rear wheelbase is X, then the supply equipment controls the steering angle Θ:

[0098] Θ=2*sin -1 (0.5*X / R)

[0099] There is a deviation between the initial longitudinal axis of the supply equipment and the target trajectory, and the deviation angle is also -Θ. Therefore, at the beginning of the movement, the longitudinal axis of the supply equipment should be quickly rotated through -Θ to make the longitudinal axis of the vehicle tangent to the target trajectory. Then, the supply equipment is quickly controlled to control the steering angle to Θ and maintain it; along the target trajectory, approach the equipment to be supplied. When the alignment angle deviation is ±3°, switch to control strategy 1.

[0100] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An intelligent automatic docking device based on monocular vision, characterized in that: include: Docking mark, high-definition camera, image processing unit and docking control unit, wherein: The docking mark is arranged on the plane where the docking port of the equipment to be supplied is located, and is used as a visual distance and direction finding marker; the high-definition camera is arranged on one side of the supply port of the supply equipment, and is used to collect the image of the plane where the docking port of the equipment to be supplied is located and send it to the image processing unit; the image processing unit is arranged on the supply equipment, and is used to process and analyze the image, calculate the distance between the high-definition camera and the plane where the docking port is located, and determine the angle between the optical axis of the high-definition camera and the normal of the plane where the docking port is located; the docking control unit is used to generate a docking control strategy and action control of the supply equipment during docking according to the spatial position relationship between the supply equipment and the equipment to be supplied.

2. The intelligent automatic docking device based on monocular vision according to claim 1 is characterized in that: The docking mark is arranged on the plane where the docking port of the equipment to be supplied is located, specifically outside the docking port; the docking mark is centered on the docking port, and at least one circle is set outside the docking port, and each circle of docking marks is a square mark composed of a pair of horizontal marks and a pair of vertical marks.

3. The intelligent automatic docking device based on monocular vision according to claim 1, characterized in that: The high-definition camera is arranged near the supply port at the tail of the equipment to be supplied, and its field of view should be able to fully capture the docking mark and the docking port.

4. An intelligent automatic docking method based on monocular vision, characterized in that: include: Step 1: Before the docking operation, the supply equipment is located at one side of the rear of the equipment to be supplied, the tails of the supply equipment and the equipment to be supplied are at a preset distance, and the longitudinal axes of the supply equipment and the equipment to be supplied do not overlap; the position of the equipment to be supplied is adjusted so that the docking port of the equipment to be supplied appears in the field of view of the high-definition camera; Step 2: Arrange the docking mark around the docking port of the equipment to be supplied; the docking mark is centered on the docking port, and is composed of a pair of horizontal marks and a pair of vertical marks to form a square mark, and the docking mark is arranged in multiple circles; Step 3, using a high-definition camera to collect the image of the plane where the replenishment equipment interface is located, performing distortion correction and image optimization processing on the image, and then transmitting it to the image processing unit; Step 4: The image processing unit identifies multiple circles of docking marks in the image, selects one circle of docking marks as the identification mark, and calculates the spatial position relationship between the supply equipment and the equipment to be supplied by obtaining the coordinate position information of the identification mark, and obtains the distance and angle between the supply equipment and the equipment to be supplied; Step 5: The image processing unit sends the calculated distance and angle between the supply equipment and the equipment to be supplied to the docking control unit; the docking control unit formulates a docking control strategy according to the distance and angle and performs action control during docking.

5. The intelligent automatic docking method based on monocular vision according to claim 4 is characterized in that: The step of calculating the spatial position relationship between the replenishment equipment and the equipment to be replenished by acquiring the coordinate position information of the identification mark includes: For the left vertical mark in the identification mark, its length is L1, and its upper and lower ends are at the vertical coordinates Y1 and Y2 in the image screen. Then the height angle A corresponding to the left vertical mark is: A=(Y2-Y1)β / H ppi Where β is the vertical field of view of the HD camera, H ppi It is the vertical resolution of the HD camera image; The distance S between the HD camera and the vertical sign on the left is: S=L1 / tan(A) Based on the distance S between the high-definition camera and the vertical mark on the left, determine the angle K between the plane where the docking interface is located and the optical axis of the high-definition camera.

6. The intelligent automatic docking method based on monocular vision according to claim 5 is characterized in that: The method of determining the angle K between the plane where the docking interface is located and the optical axis of the high-definition camera based on the distance S between the high-definition camera and the left vertical mark includes: When the two vertical marks in the identification mark are located on the left and right sides of the center of the image: Calculate the height angle A corresponding to the two vertical marks on the left and right sides of the identification mark n and A f , and the distance S between the HD camera and the two vertical markers on the left and right sides n and S f ; It is known that the length of the horizontal mark in the identification mark is L2, then the angle K between the plane where the interface is located and the camera optical axis is: K=sin -1 ((S n ×sin(A n )+S f ×sin(A f )) / L2) When the two vertical marks in the identification mark are located on the same side of the center of the image screen: Note that the horizontal coordinates of the two vertical marks on the left and right sides of the image are X n and X f , the calculation formula of the azimuth corresponding to the vertical mark is: Where i = n or f; according to the horizontal coordinate X n and X f The corresponding azimuth angle B can be obtained n and B f ; Calculate the distance S between the HD camera and the vertical mark on the left n Corresponding angle C: <h2 style=";text-align:left;direction:ltr">C = sin<h2 style=";text-align:left;direction:ltr"> -1 <h2 style=";text-align:left;direction:ltr"> ((sin(|B<h2 style=";text-align:left;direction:ltr"> f <h2 style=";text-align:left;direction:ltr"> |-|B<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> |)×S<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> / L2)) Finally, the angle K between the plane where the docking interface is located and the optical axis of the high-definition camera is obtained as: K=180-|B f |-C The distance S between the HD camera and the vertical mark on the left, and the angle K between the plane where the docking port is located and the optical axis of the HD camera are used as the distance and angle between the supply equipment and the equipment to be supplied.

7. The intelligent automatic docking method based on monocular vision according to claim 4 is characterized in that: The image processing unit identifies the multi-circle docking mark in the image screen by means of binarization processing, edge extraction and feature judgment.

8. The intelligent automatic docking method based on monocular vision according to claim 6, characterized in that: During the docking process, the movement speed of the supply equipment is stabilized within the preset speed range. The docking control strategy is divided into two types: Strategy 1: When the angle K between the supply equipment and the equipment to be supplied is within the preset angle range, the docking control unit controls the supply equipment to approach the equipment to be supplied at a preset speed range, and the direction of the supply equipment is slightly adjusted until the angle K is 0 or close to 0, indicating that the supply equipment and the equipment to be supplied are aligned. At this time, the supply equipment is controlled to move straight backward until the supply port and the docking port are docked; Strategy 2: When the angle K between the supply equipment and the equipment to be supplied is outside the preset angle range, control the movement speed and turning direction of the supply equipment to remain stable until the angle K is within the preset angle range and is basically aligned, and then dock according to Strategy 1.

9. The intelligent automatic docking method based on monocular vision according to claim 8, characterized in that: In Strategy 2, the supply equipment is controlled to move along a specific arc trajectory. When the control strategy ends, the two goals of accurate positioning and basic angle alignment can be achieved; the radius R of the arc is: R=0.5×(S 2 ×cos 2 (K)+4-4×S×cos(K)+S 2 ×without 2 (K)) / S / sin(K)。 10. The intelligent automatic docking method based on monocular vision according to claim 9, characterized in that: The supply equipment moves along a specific arc trajectory, and the turning radius needs to be controlled; assuming that the supply equipment is rear-wheel driven, the front wheels are responsible for steering, and the front and rear wheelbase is X, then the supply equipment controls the steering angle Θ: Θ=2*sin -1 (0.5*X / R) There is a deviation between the initial longitudinal axis of the supply equipment and the specific arc trajectory, and the deviation angle is also -Θ. Therefore, at the beginning of the movement, the longitudinal axis of the supply equipment should be quickly rotated through -Θ to make the longitudinal axis of the vehicle tangent to the target trajectory, and then the supply equipment should be quickly controlled to control the steering angle to Θ and maintain it; approach the equipment to be supplied along the specific arc trajectory, and when the angle K is within the preset angle range, switch to control strategy 1.

Images