Automatic butt joint method for socket circular pipeline based on binocular vision

Abstract

This invention discloses an automatic docking method for socket-type circular pipes based on binocular vision, comprising the following steps: pre-operation inspection; installing pipe assembly fixtures onto a first pipe assembly and a second pipe assembly; acquiring images at the docking ports and selecting control points from the contour lines of the images at each docking port; establishing a spatial coordinate system at the docking ports of the two pipe assemblies, and estimating the spatial pose of the docking ports of the two pipe assemblies based on the spatial coordinates of the control points; calculating the number of pipes to be docked; matching the docking ports of the pipes in the two pipe assemblies, calculating the angular deviation of each axis of the coordinate system, and planning a pipe attitude adjustment scheme; setting an angular deviation threshold; when the angular deviation of the pose is less than the angular deviation threshold, executing the axial feed of the selected pipes in the first pipe assembly; traversing all pipes, and ending the operation when all pipes are detected to be docked.

Description

Technical Field

[0001] This invention relates to the field of pipeline connection construction, and specifically to an automatic connection method for socket-type circular pipelines based on binocular vision. Background Technology

[0002] Socket-type pipes are pipe fittings used for connection. Their pipe connection ports can be divided into pipe sockets and pipe spigots. Pipe connection is achieved by inserting the pipe spigot into the pipe socket. Socket-type round pipes are widely used in power conduits and drainage pipes.

[0003] In the group laying and installation of urban power pipelines, workers near the pipeline group observe the pipeline's posture and report back to the control room. Workers in the control room then use a robotic arm equipped with a pipeline clamp to perform the pipeline insertion action. Existing technology CN116771984A provides an automatic identification and docking method for circular pipelines in a port setting. It utilizes an integrated gimbal-rotated 2D laser scanner and motor-driven pitch angle changes to scan and acquire 3D point cloud information of the circular pipeline. This information is then processed and a point cloud dataset model is established to complete the pipeline docking. However, this method of acquiring 3D point cloud information using a 2D laser scanner has high power consumption, and the large measurement range of the laser radar results in weak filtering of irrelevant image information, which is not conducive to the extraction and processing of image information at the pipeline docking port. Existing technology CN116748859A provides an automatic leveling and docking method for pump body pipelines. It uses a camera to acquire images of the pipeline bolt holes, a tilt sensor to detect the pipeline's horizontal state, and a position sensor to detect the distance between the docking pipelines, thus completing the docking of the pipeline and pump body. However, this invention uses a variety of sensors and has high technical requirements. Summary of the Invention

[0004] The purpose of this invention is to provide an automatic docking method for socketed circular pipes based on binocular vision. The method involves acquiring images of the pipe docking ports using a binocular camera, recognizing and completing the image contours, filtering redundant environmental image information, and completing the automatic docking of the pipes.

[0005] To achieve the above objectives, the present invention provides an automatic docking method for socketed circular pipes based on binocular vision, comprising the following steps:

[0006] Step S1: Pre-operation inspection, install the pipe assembly clamps onto the first and second pipe assemblies;

[0007] Step S2: Acquire images of the docking ports of the first and second pipe groups and select control points from the contour lines of the images at each docking port; establish a spatial coordinate system at the docking ports of the two pipe groups, and estimate the spatial pose of the docking ports of the first and second pipe groups based on the spatial coordinates of the control points.

[0008] Step S3: Based on the image information processed in step S2, calculate the number of pipes that need to be connected; select a connection port in the second pipe group and match it with the corresponding connection port in the first pipe group, calculate the angular deviation of each axis of the coordinate system, plan the pipe attitude adjustment scheme, and perform secondary adjustment on the pipe attitude of the first pipe group;

[0009] Step S4: Set an angle deviation threshold. When the angle deviation between the pose of the first pipe group docking port and the pose of the second pipe group docking port is less than the angle deviation threshold, the insertion condition is met, and the axial feed of the selected pipe in the first pipe group is executed. If the insertion condition is not met, repeat step S3.

[0010] Step S5: Based on the image information, traverse all pipes. When an incompletely connected pipe is detected in the image, repeat steps S3 to S4. When all pipes are detected to be fully connected, end the operation.

[0011] Optionally, step S1 includes:

[0012] Step S1.1: Before operation, complete the function check of the pipe assembly fixture, including: the pose estimation function of the docking port, the solution of the attitude adjustment scheme, and the clarity of the acquired image;

[0013] Step S1.2: Fix the pipe assembly clamp to the first pipe assembly and the second pipe assembly, use the robotic arm to drive the pipe assembly clamp, initially adjust the position of the docking port of the first pipe assembly and the docking port of the second pipe assembly, and send a prompt to the staff in the operating room.

[0014] Optionally, the pipe assembly clamp includes:

[0015] Attitude control hydraulic device, axial feed hydraulic device, two pipe limit mechanisms, binocular camera and control system;

[0016] The attitude control hydraulic device is installed on the first pipeline group, the axial feed hydraulic device is connected to one end of the attitude control hydraulic device, the pipeline limiting mechanism is connected to the other end of the attitude control hydraulic device, and the binocular camera is installed on the pipeline limiting mechanism of the first pipeline group.

[0017] The two pipe limiting mechanisms are respectively installed at the docking port of the first pipe group and the docking port of the second pipe group to limit the docking ports of each pipe in the first pipe group and the second pipe group.

[0018] The control system is communicatively connected to the binocular camera, the attitude control hydraulic device, and the axial feed hydraulic device. It receives image signals from the binocular camera, processes the images, designs a pipeline attitude adjustment scheme, and transmits the adjustment signals to the attitude control hydraulic device and the axial feed hydraulic device to complete the insertion and insertion of the two sets of pipelines.

[0019] Optionally, in steps S2 and S5, the binocular camera is used to acquire real-time images at the docking port of the first pipe group and the second pipe group. The binocular camera consists of two monocular cameras, which are respectively disposed on both sides of the pipe limiting mechanism on the first pipe group.

[0020] Optionally, the spatial coordinate system in step S2 is selected with the center point of the ellipse in the image as the origin of the spatial coordinate axis, the pipeline axis as the z-axis, the major axis of the ellipse as the x-axis, and the minor axis of the ellipse as the y-axis. The spatial coordinate system of the docking port of the second pipeline group is represented by x1, y1, z1, and the spatial coordinate system of the docking port of the first pipeline group is represented by x2, y2, z2.

[0021] Optionally, step S4 utilizes the axial feed hydraulic device to feed a single pipe in the first pipe group along its axial direction. The axial feed hydraulic device consists of several independent hydraulic pistons, each of which is fixed to a non-connecting port on each pipe in the first pipe group.

[0022] Optionally, in step S3, the attitude control hydraulic device is used to make a secondary adjustment to the attitude of the first pipe group. The attitude control hydraulic device consists of three independent hydraulic pistons, which can adjust the spatial position of the first pipe group in three degrees of rotation.

[0023] Optionally, the method for constructing the pipeline attitude adjustment scheme in step S3 is as follows:

[0024] The pose information of the second pipe group docking port and the first pipe group docking port can be represented as follows: and And its calculation method is as follows:

[0025]

[0026] In the formula, The elements are derived from the spatial coordinates of the control points on the pipe port outline in step S2.

[0027] The orientation of the docking port of the second pipe group relative to the docking port of the first pipe group is as follows: The attitude can be represented in the RPY coordinate system as RPY(φ,θ,ψ)=Rot(z2,φ)Rot(y2,θ)Rot(x2,ψ); according to =RPY(φ,θ,ψ) can be used to solve for the adjustment angles of each axis:

[0028]

[0029] In the formula, , , , , , , All indicate the result after calculation. The elements in the diagram, ψ, θ, and φ, represent the rotation angles required in the x2, y2, and z2 axes of the docking port of the first pipe group when the insertion action is completed and the docking port of the first pipe group moves to the docking port of the second pipe group.

[0030] The angles for adjusting the attitude of the first pipe group docking port of ψ, θ, and φ are used as the pipe attitude adjustment scheme.

[0031] Optionally, in step S2, the images at the docking ports of the first and second pipe groups are filtered, edge detected, feature extracted and matched to obtain a binary map containing partial contour lines of the docking ports of the first and second pipe groups and a depth map containing key point distance information, and the contours of the docking ports of the two pipe groups are matched and completed.

[0032] Optionally, the docking ports of the first pipe group are all pipe spigots, and the docking ports of the second pipe group are all pipe sockets.

[0033] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:

[0034] (1) The method of the present invention identifies the elliptical contour of the pipeline docking port by using Hough transform and boundary clustering algorithm, and performs contour completion on it, filtering out irrelevant non-pipeline docking port image information, which greatly reduces the information processing work of the control system and improves the working efficiency and docking accuracy of automatic pipeline docking.

[0035] (2) The method of the present invention uses a binocular camera to acquire image information of the pipe docking port, and uses the Canny algorithm, ORB (Oriented FAST and Rotated BRIEF) algorithm and EPnP (Efficient Perspective-n-Point) algorithm to obtain the pose estimation of the pipe docking port, plan the pipe adjustment scheme and execute it, reduce the risk of manual construction, reduce the types of sensors used, and facilitate the installation and laying of the socket pipe assembly. Attached Figure Description

[0036] Figure 1This is a flowchart of the automatic docking method for socketed circular pipes based on binocular vision according to the present invention.

[0037] Figure 2 This is a schematic diagram of the control system of the automatic docking method for socketed circular pipes based on binocular vision according to the present invention.

[0038] Figure 3 This is a schematic diagram of the pipe assembly clamp structure for an automatic docking method for socketed circular pipes based on binocular vision according to the present invention.

[0039] Figure 4 This is a schematic diagram illustrating the establishment of the pipe docking port coordinate system in an automatic docking method for socketed circular pipes based on binocular vision according to the present invention.

[0040] In the diagram: 1-Control system, 2-Binocular camera, 3-Attitude control hydraulic device, 4-Axial feed hydraulic device, 5-First pipe group, 51-Pipe socket, 52-Pipe spigot, 6-Pipe limiting mechanism, 7-Fixing component. Detailed Implementation

[0041] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] In the description of this invention, it should be noted that the terms "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0043] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0044] Figure 3This is a schematic diagram of the pipe assembly clamp structure for the binocular vision-based automatic docking method for socketed circular pipes according to the present invention. When docking two sets of pipes, the docking ports of the first pipe set 5 are all pipe sockets 52, and the docking ports of the second pipe set (the...) are... Figure 3 The docking ports of the pipe group (not shown) are all pipe sockets 51. Some parts of the pipe group clamp are fixed to the first pipe group 5, and the remaining parts of the pipe group clamp are fixed to the second pipe group. The user controls the robotic arm to clamp the pipe group clamp and move the first pipe group 5 to complete the insertion and insertion of each pipe socket 52 of the first pipe group into each pipe socket 51 of the second pipe group.

[0045] The pipe assembly clamp includes: a posture control hydraulic device 3, an axial feed hydraulic device 4, two pipe limiting mechanisms 6, a binocular camera 2, a control system 1, and a fixing component 7. The posture control hydraulic device 3, mounted on the first pipe assembly 5, consists of three independent hydraulic pistons, capable of adjusting the spatial position of the first pipe assembly 5 with three rotational degrees of freedom. The axial feed hydraulic device 4, located at one end of the posture control hydraulic device 3, consists of several independent hydraulic pistons, each fixed to the non-connection port side of each pipe in the first pipe assembly 5, enabling feed along the axial direction of a single pipe during the connection process. Two pipe limiting mechanisms 6 are respectively installed at the pipe inlet 52 of the first pipe group 5 (i.e., the other end of the attitude control hydraulic device 3) and at the pipe socket 51 of the second pipe group. Each pipe limiting mechanism 6 is a metal frame structure with several baffles. The baffles are installed between individual pipes in the first or second pipe group to prevent pipes from overlapping and causing mismatching of pipe docking ports. During pipe docking, the mechanism can limit the docking ports of each pipe in the first and second pipe groups to prevent over-insertion or under-insertion. Furthermore, as... Figure 2 As shown, the binocular camera 2 consists of two monocular cameras, which are respectively installed on both sides of the pipe limiting mechanism 6 on the first pipe group 5. These cameras can capture real-time images of the pipe socket 51 of the second pipe group and the pipe spigot 52 of the first pipe group 5 to further detect the connection between the pipes. The fixing member 7 is installed on the attitude control hydraulic device 3 and can fix the pipe group clamp on the first pipe group 5.

[0046] like Figure 2As shown, the control system 1 is a remote control system with an internal control chip. It is communicatively connected to the binocular camera 2, the attitude control hydraulic device 3, and the axial feed hydraulic device 4. The control system 1 can receive image signals from the binocular camera 2, process the image signals, and further design a pipeline attitude adjustment scheme. The adjustment signal is transmitted to the attitude control hydraulic device 3 and the axial feed hydraulic device 4. The attitude control hydraulic device 3 adjusts the pipeline posture, and the axial feed hydraulic device 4 propels the pipeline forward to complete the insertion of the two sets of pipelines.

[0047] like Figure 1 As shown, the automatic docking method for socketed circular pipes based on binocular vision includes:

[0048] Step S1, pre-operation inspection: The robotic arm moves the pipe assembly clamp fixed to the first pipe assembly 5, thereby moving the first pipe assembly 5 as well, initially adjusting the position of the pipe spigot at the docking port of the first pipe assembly 5 relative to the pipe socket at the docking port of the second pipe assembly. Specifically, this includes:

[0049] Step S1.1: Before operation, complete the inspection of the control system, the calibration of the binocular camera, and the inspection of the binocular camera function. The inspection of the binocular camera function includes: the pose estimation function of the pipe socket and the pipe spigot, the solution of the attitude adjustment scheme, and the image clarity of the binocular camera.

[0050] Step S1.2: Fix the pipe assembly clamp to the pipe spigot side of the first pipe assembly 5, and use the robotic arm to move the pipe spigot 52 of the first pipe assembly 5 to a position close to the pipe socket 51 of the second pipe assembly. The control system 1 sends a prompt to the staff in the operating room.

[0051] Step S2: Acquire images using a binocular camera, control the system to process the images, and estimate the pose of the pipe socket of the second pipe group and the pipe spigot of the first pipe group.

[0052] Step S2.1: Images of the two sets of pipe docking ports are acquired by the binocular camera 2 and transmitted to the control system 1;

[0053] Step S2.2 involves processing the image at the pipe inlet 52 of the first pipe group 5. The image processing includes filtering, edge detection, feature extraction, and matching. The processed image is a binary image containing a portion of the outline of the pipe inlet of the first pipe group, and a depth map containing key point distance information.

[0054] The filtering and edge detection employ the Canny algorithm to eliminate unnecessary interference information in the image, obtaining a binary image containing the partial contour line of the pipe docking port. The Canny algorithm includes the following steps: smoothing the image using Gaussian filtering to remove noise; calculating the gradient magnitude and direction of each pixel in the image using the Sobel operator; determining whether a pixel is the maximum value of the maximum gradient in the same direction among its surrounding pixels, thereby suppressing non-maximum values ​​and eliminating the adverse effects of edge detection; applying double threshold detection to determine real and potential edges; and completing edge detection by suppressing isolated weak edges.

[0055] The feature extraction and matching employs the ORB (Oriented FAST and Rotated BRIEF) algorithm, which matches images captured by two monocular cameras to determine each pipe socket in the image and its matching pipe spout, and calculates the distance from each point in the scene to the camera, obtaining a depth map containing keypoint distance information. The ORB algorithm's steps include: using the FAST (Features from Accelerated Segments Test) algorithm to find keypoints in the image; using the adjusted BRIEF (Binary Robust Independent Elemantary Features) algorithm to create binary feature vectors based on the image keypoints; and calculating the Hamming distance between the feature vectors of the two monocular camera images to match feature points.

[0056] Step S2.3: Based on the local features extracted from the image, the control system is used to match and complete the contour of the pipe docking port in the image.

[0057] By invoking the Hough Transform in the control system, the incomplete elliptical contour of the pipe docking port in the image from step S2.1 is completed, and the center point position, the lengths of the major and minor axes, and the deflection angle of the elliptical axis are extracted. The Hough Transform includes: randomly selecting three points and all points in their neighborhood of the pipe docking port from the binary image containing the partial contour line after edge detection, and fitting an ellipse using the least squares method; randomly selecting additional points from the edge points and determining whether these points lie on the fitted ellipse to verify the accuracy of the ellipse.

[0058] In a preferred embodiment, the outline of the pipe docking port in the image is completed by calling the boundary clustering algorithm in the control system. The steps of the boundary clustering algorithm include: connecting edge points using the Kovesi boundary connection algorithm; extracting polyline segments from the connected edge curves; detecting and segmenting concave points and corner points, and removing some non-elliptical arc segments; clustering and re-pairing elliptical arcs; and fitting ellipses using the least squares method to complete the outline of the pipe docking port.

[0059] Step S2.4: Perform image processing on the pipe socket 51 of the second pipe group, repeat steps S2.2 to S2.3 to obtain a binary image containing the partial outline of the pipe socket of the second pipe group and a depth image containing key point distance information, and match and complete the outline of its pipe docking port.

[0060] Step S2.5: Based on the completed outline shape of the pipe docking port, establish the spatial coordinate system of the pipe docking port, and use the control system to estimate the spatial pose of the pipe docking port.

[0061] like Figure 4 As shown, based on the two sets of pipe docking port contour shapes supplemented in steps S2.2 to S2.4, a spatial coordinate system is established at the pipe docking ports. The center point of the ellipse in the image is selected as the origin of the spatial coordinate axis, with the pipe axis as the z-axis, the major axis of the ellipse as the x-axis, and the minor axis of the ellipse as the y-axis. The spatial coordinate system of the pipe socket in the second pipe group is represented by x1, y1, z1, and the spatial coordinate system of the pipe spigot in the first pipe group is represented by x2, y2, z2. The EPnP (Efficient Perspective-n-Point) algorithm is used to calculate the pose estimates of the pipe socket and pipe spigot. The steps of the EPnP algorithm include: selecting control points from the supplemented pipe docking port contour lines; calculating the coordinates and projection relationships of the control points in the spatial coordinate system of the binocular camera, thereby obtaining the pose estimates of the pipe socket 51 and pipe spigot 52.

[0062] Step S3: Based on the image information processed in step S2, calculate the number of pipes that need to be connected, formulate an attitude adjustment plan, and execute it.

[0063] Step S3.1: Analyze the images captured by the binocular camera 2 and calculate the number of pipes that need to be connected using the control system. Based on the pose information of the pipe sockets and spigots obtained in step S2.5, select one pipe socket from the second pipe group and match it with the corresponding pipe spigot from the first pipe group, calculate the angular deviation of each axis of the coordinate system, and plan a pipe attitude adjustment scheme.

[0064] The positional information of the pipe socket and pipe spigot can be represented as follows: and And its calculation method is as follows:

[0065] (Equation 1)

[0066] In the formula, The elements are derived from the spatial coordinates of the control points on the pipe docking port outline in steps S2.2 to S2.4.

[0067] The orientation of the pipe socket 51 relative to the pipe spigot 52 Equation 2 can be used to represent the attitude in the RPY coordinate system as RPY(φ,θ,ψ)=Rot(z2,φ)Rot(y2,θ)Rot(x2,ψ). =RPY(φ,θ,ψ) can be used to solve for the adjustment angles of each axis, as shown in Equation 3.

[0068] (Equation 2)

[0069] (Equation 3)

[0070] In the formula, ψ, θ, and φ represent the rotation angles of the pipe spigot 52 in the selected paired pipe in the x2, y2, and z2 axes respectively when the spigot 52 moves towards the socket 51.

[0071] The angles for adjusting the pipe spigot attitude of ψ, θ, and φ are used as the pipe attitude adjustment scheme.

[0072] Step S3.2: Based on the attitude adjustment scheme obtained in step S3.1, the structure of the attitude control hydraulic device, and the positions of the pipe socket 51 and pipe spigot 52 calculated in step S2.5, the control system 1 calculates the displacement of the piston in the axial feed hydraulic device 4 corresponding to the currently selected pipe, and drives the three hydraulic pistons in the attitude control hydraulic device 3 to execute the pipe attitude adjustment scheme, thereby adjusting the angle of the pipe spigot and performing a secondary adjustment on the attitude of the currently selected pipe in the first pipe group. After completing the docking of the currently selected pipe, the robotic arm adjusts the position of the pipe clamp to dock the next group of pipes.

[0073] Step S4: Set the angle deviation threshold and determine whether the pose of the currently selected pipe socket in the first pipe group meets the insertion conditions.

[0074] When the angular deviation between the pipe spigot position in the first pipe group and the pipe socket position in the second pipe group is less than the angular deviation threshold, the insertion condition is met. The control system 1 calculates the independent hydraulic piston movement distance of the axial feed hydraulic device 4 according to step S3.2, and drives the independent hydraulic piston in the axial feed hydraulic device 4 to move the pipe spigot 52 to complete the pipe insertion connection.

[0075] When the angular deviation between the pipe spigot position in the first pipe group and the pipe socket position in the second pipe group is greater than or equal to the angular deviation threshold, step S3 will be repeated to formulate a new attitude adjustment plan and make another judgment.

[0076] Step S5: Based on the image information from the binocular camera 2, the control system 1 traverses all pipes to determine whether the pipes are properly connected. If any incompletely connected pipes are detected in the image, steps S3 to S4 are repeated; when all pipes are found to be properly connected, the operation ends.

[0077] In a preferred embodiment, the method of the present invention provides an automatic docking method for clamping power pipeline groups with specific clamps. This method can also be applied to the accurate identification and automatic docking of similar socketed circular pipes.

[0078] In summary, this invention identifies elliptical contours of pipe docking ports using Hough transform and boundary clustering algorithms, completes the contours, and filters out irrelevant non-pipe docking port image information, greatly reducing the information processing workload of the control system and improving the efficiency and accuracy of automatic pipe docking. Furthermore, it employs a binocular camera to acquire pipe docking port image information and utilizes the Canny algorithm, ORB (Oriented Fast and Rotated BRIEF) algorithm, and EPnP (Efficient Perspective-n-Point) algorithm to obtain the pose estimation of the pipe docking port, plan and execute pipe adjustment schemes, reduce the risks of manual construction, reduce the types of sensors used, and facilitate the installation and laying of socketed pipe assemblies.

[0079] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.

Claims

1. A method for automatic butt joint of socket circular pipes based on binocular vision, characterized in that, Includes the following steps: Step S1: Pre-operation inspection, install the pipe assembly clamps onto the first and second pipe assemblies; Step S2: Acquire images of the docking ports of the first and second pipe groups and select control points from the contour lines of the images at each docking port; establish a spatial coordinate system at the docking ports of the two pipe groups, and estimate the spatial pose of the docking ports of the first and second pipe groups based on the spatial coordinates of the control points; wherein, the spatial coordinate system is selected with the center point of the ellipse in the image as the origin of the spatial coordinate axis, the pipe axis as the z-axis, the major axis of the ellipse as the x-axis, and the minor axis of the ellipse as the y-axis, wherein the spatial coordinate system of the docking port of the second pipe group is represented by x1, y1, z1, and the spatial coordinate system of the docking port of the first pipe group is represented by x2, y2, z2; Step S3: Based on the image information processed in step S2, calculate the number of pipes that need to be connected; select a connection port in the second pipe group and match it with the corresponding connection port in the first pipe group, calculate the angular deviation of each axis of the coordinate system, plan the pipe attitude adjustment scheme, and perform secondary adjustment on the pipe attitude of the first pipe group; Step S4: Set an angle deviation threshold. When the angle deviation between the pose of the first pipe group docking port and the pose of the second pipe group docking port is less than the angle deviation threshold, the insertion condition is met, and the axial feed of the selected pipe in the first pipe group is executed. If the insertion condition is not met, repeat step S3. Step S5: Based on the image information, traverse all pipes. When an incompletely connected pipe is detected in the image, repeat steps S3 to S4. When all pipes are detected to be fully connected, end the operation. The method for constructing the pipeline attitude adjustment scheme in step S3 is as follows: The pose information of the second pipeline group docking port and the first pipeline group docking port can be respectively represented as and , and the calculation method is that In the formula, The elements are derived from the spatial coordinates of the control points on the pipe port outline in step S2. The orientation of the docking port of the second pipe group relative to the docking port of the first pipe group is as follows: The attitude can be represented in the RPY coordinate system as RPY(φ,θ,ψ)=Rot(z2,φ)Rot(y2,θ)Rot(x2,ψ); according to =RPY(φ,θ,ψ) can be used to solve for the adjustment angles of each axis: In the formula, , , , , , , All indicate the result after calculation. The elements in the diagram, ψ, θ, and φ, represent the rotation angles required in the x2, y2, and z2 axes of the docking port of the first pipe group when the insertion action is completed and the docking port of the first pipe group moves to the docking port of the second pipe group. The angles for adjusting the attitude of the first pipe group docking port of ψ, θ, and φ are used as the pipe attitude adjustment scheme.

2. The method of automatic butt joining of socket circular pipes based on binocular vision according to claim 1, characterized in that, Step S1 includes: Step S1.1: Before operation, complete the function check of the pipe assembly fixture, including: the pose estimation function of the docking port, the solution of the attitude adjustment scheme, and the clarity of the acquired image; Step S1.2: Fix the pipe assembly clamp to the first pipe assembly and the second pipe assembly, use the robotic arm to drive the pipe assembly clamp, initially adjust the position of the docking port of the first pipe assembly and the docking port of the second pipe assembly, and send a prompt to the staff in the operating room.

3. The automatic docking method for socket-type circular pipes based on binocular vision according to claim 1, characterized in that, The pipe assembly clamp includes: Attitude control hydraulic device, axial feed hydraulic device, two pipe limit mechanisms, binocular camera and control system; The attitude control hydraulic device is installed on the first pipeline group, the axial feed hydraulic device is connected to one end of the attitude control hydraulic device, the pipeline limiting mechanism is connected to the other end of the attitude control hydraulic device, and the binocular camera is installed on the pipeline limiting mechanism of the first pipeline group. The two pipe limiting mechanisms are respectively installed at the docking port of the first pipe group and the docking port of the second pipe group to limit the docking ports of each pipe in the first pipe group and the second pipe group. The control system is communicatively connected to the binocular camera, the attitude control hydraulic device, and the axial feed hydraulic device. It receives image signals from the binocular camera, processes the images, designs a pipeline attitude adjustment scheme, and transmits the adjustment signals to the attitude control hydraulic device and the axial feed hydraulic device to complete the insertion and insertion of the two sets of pipelines.

4. The method of automatic butt joining of socket circular pipes based on binocular vision according to claim 3, characterized in that, In steps S2 and S5, the binocular camera is used to acquire real-time images at the docking port of the first pipe group and the second pipe group. The binocular camera consists of two monocular cameras, which are respectively set on both sides of the pipe limiting mechanism on the first pipe group.

5. The method of automatic butt-joining of socket-and-spigot circular pipes based on binocular vision according to claim 3, characterized in that, Step S4 utilizes the axial feed hydraulic device to feed a single pipe in the first pipe group along its axial direction. The axial feed hydraulic device consists of several independent hydraulic pistons, each of which is fixed to a non-connecting port on each pipe in the first pipe group.

6. The method of automatic butt-joining of socket-and-spigot circular pipes based on binocular vision according to claim 3, characterized in that, Step S3 uses the attitude control hydraulic device to make a secondary adjustment to the attitude of the first pipe group. The attitude control hydraulic device consists of three independent hydraulic pistons, which can adjust the spatial position of the first pipe group in three degrees of rotation.

7. The method of automatic butt-joining of socket-and-spigot circular pipes based on binocular vision according to claim 1, characterized in that, In step S2, the images at the docking ports of the first and second pipe groups are filtered, edge detected, feature extracted and matched to obtain a binary map containing partial contour lines of the docking ports of the first and second pipe groups and a depth map containing key point distance information. The contours of the docking ports of the two pipe groups are then matched and completed.

8. The method of automatic butt-joining of socket-and-spigot circular pipes based on binocular vision according to claim 1, characterized in that, The first pipe group has pipe spigots as its docking ports, while the second pipe group has pipe sockets as its docking ports.

Images