Control method and device for collaborative robot for fastening bolts of high-voltage live equipment

Abstract

The present application provides a control method and device for a collaborative robot for tightening bolts on high-voltage live equipment, comprising: controlling the collaborative robot to carry a tightening tool into the operating area of a target bolt in the high-voltage live equipment; obtaining a real-time image of the operating area, and performing a three-dimensional reconstruction of the operating area of the target bolt to obtain three-dimensional reconstruction data of the operating area; aligning the center of the tightening tool approximately with the center of the target bolt based on the real-time image of the operating area and the three-dimensional reconstruction data of the operating area; obtaining a coaxial image of the operating area, extracting an image of the target bolt from the coaxial image of the operating area, and adjusting the tightening tool based on the deviation between the image of the target bolt and the coaxial image of the operating area, so that the center of the target bolt is aligned with the center of the tightening tool. The present application can precisely control the movement of a collaborative robot that performs uninterrupted power tightening on bolts of high-voltage live equipment and effectively provide insulation protection between phases.

Description

Technical Field

[0001] The present application belongs to the field of intelligent manufacturing technology, and particularly relates to the field of substation equipment technology, and specifically relates to a control method and equipment for a collaborative robot for tightening bolts on high-voltage live equipment. Background Art

[0002] Loose hexagonal bolts are a common equipment defect during the daily operation and maintenance of substations. As a critical hub in the power system, the stable operation of substation equipment is crucial to the safety and reliability of the entire power supply. However, due to the long-term influence of various external factors such as equipment vibration, temperature fluctuations, and electromagnetic forces, hexagonal bolts are prone to loosening after long-term operation. If this loosening is not promptly addressed, it may lead to a series of safety hazards, such as increased contact resistance at equipment connections, resulting in localized overheating. In severe cases, it may even cause electrical fires or equipment damage, seriously impacting the normal operation of the power system.

[0003] Currently, there is a lack of mature and effective live-tightening solutions for the common defect of loose hexagonal bolts in substation equipment during actual maintenance operations. Traditional troubleshooting methods often require powering off the equipment, followed by manual tightening by professionals. However, this approach has numerous drawbacks. First, power outages can impact the power supply, especially in environments with high power reliability requirements. Power outages can cause production interruptions and inconveniences. Second, manual tightening operations are not only labor-intensive but also pose certain safety risks. For example, when working at height or in confined spaces, workers are prone to accidents.

[0004] In recent years, with the continuous development and advancement of robotics technology, collaborative robots, thanks to their high flexibility, precise operation capabilities, and excellent human-robot collaboration, have been widely used in various precision and hazardous operations. For example, in industries such as automotive manufacturing, aerospace, and electronic equipment production, collaborative robots have demonstrated significant advantages, effectively improving production efficiency, reducing labor intensity, and ensuring operator safety. However, the application of collaborative robots in bolt tightening operations in high-voltage, live environments faces numerous challenges, and a mature application solution is currently lacking.

[0005] High-voltage, live environments are unique and complex. On the one hand, the presence of high voltage necessitates that robots possess excellent insulation properties during operation to prevent electric shock accidents and electromagnetic interference with surrounding equipment. On the other hand, when operating near high-voltage equipment, special attention must be paid to protecting interphase insulation to avoid serious accidents such as interphase short circuits caused by improper operation. Furthermore, the layout of substation equipment is often complex, and the position and posture of bolts vary. This requires collaborative robots to possess high-precision positioning and motion control capabilities to ensure accurate and flawless bolt tightening tasks.

[0006] To sum up, the existing methods for eliminating loose hexagonal bolts in substation equipment have many shortcomings, and the application of collaborative robots in high-voltage live bolt tightening operations is still in the exploratory stage. There is an urgent need for a collaborative robot motion control method that can achieve precise tightening of high-voltage live equipment bolts while ensuring phase-to-phase insulation, so as to improve the efficiency and safety of substation equipment maintenance and ensure the stable operation of the power system. Summary of the Invention

[0007] The present application provides a control method and device for a collaborative robot for tightening bolts of high-voltage live equipment, which is used to accurately control the movement of the collaborative robot that tightens bolts of high-voltage live equipment without disconnecting the power.

[0008] In a first aspect, the present application provides a control method for a collaborative robot for tightening bolts of high-voltage live equipment, comprising: controlling the collaborative robot to carry a tightening tool into the working area of a target bolt in the high-voltage live equipment; obtaining a real-time image of the working area, and performing three-dimensional reconstruction of the working area of the target bolt to obtain three-dimensional reconstruction data of the working area; aligning the center of the tightening tool approximately with the center of the target bolt based on the real-time image of the working area and the three-dimensional reconstruction data of the working area; obtaining a coaxial image of the working area, extracting the image of the target bolt from the coaxial image of the working area, and adjusting the tightening tool based on the deviation between the image of the target bolt and the coaxial image of the working area, so that the center of the target bolt is aligned with the center of the tightening tool.

[0009] In an implementation of the first aspect, the approximately aligning the center of the tightening tool with the center of the target bolt based on the real-time image of the work area and the three-dimensional reconstruction data of the work area includes: fitting the plane of the real-time image of the work area using the three-dimensional reconstruction data of the work area; adjusting the orientation of the central axis of the tightening tool based on the plane; detecting the target bolt in the real-time image of the work area using a preset visual detection algorithm; and controlling the translation of the central axis of the tightening tool so that the center of the tightening tool is approximately aligned with the center of the target bolt.

[0010] In an implementation of the first aspect, adjusting the orientation of the central axis of the tightening tool based on the plane includes: obtaining a normal vector of the plane based on the fitted plane; obtaining an angle between the plane normal vector of the real-time image of the working area and the central axis of the tightening tool; and detecting whether the angle is greater than a first difference tolerance threshold: if so, changing the orientation of the central axis of the tightening tool to the direction of the normal vector of the plane, and updating the three-dimensional reconstruction of the working area of the target bolt based on the real-time image of the current working area; if not, maintaining the current orientation of the central axis of the tightening tool.

[0011] In an implementation of the first aspect, extracting an image of a target bolt from the coaxial image of the work area and adjusting the tightening tool based on the deviation between the image of the target bolt and the coaxial image of the work area include: detecting the center point and the orientation of each side of the target bolt in the coaxial image of the work area; detecting whether the deviation between the center point position of the target bolt and the center point of the coaxial image of the work area is less than a second tolerance threshold, and detecting whether the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the work area is less than a third tolerance threshold; if the center point position of the target bolt is less than the horizontal axis of the coaxial image of the work area, the tightening tool is adjusted according to the deviation between the center point position of the target bolt and the horizontal axis of the coaxial image of the work area. If the deviation between the center points of the coaxial images is less than a second tolerance threshold, and the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the working area is less than a third tolerance threshold, the position of the tightening tool is maintained unchanged; if the deviation between the center point of the target bolt and the center point of the coaxial image of the working area is not less than the second tolerance threshold and / or the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the working area is not less than the third tolerance threshold, the central axis of the tightening tool is controlled to translate based on preset pixels and a preset distance step size so that the center of the tightening tool is close to the center of the target bolt.

[0012] In an implementation of the first aspect, the method further includes: pre-dividing the deviation into multiple deviation levels according to the size of the deviation; when the deviation gradually decreases and a change in the deviation level is detected, halving the current pixel and distance step, and controlling the central axis translation of the tightening tool based on the halved pixel and distance step, so that the center of the tightening tool is close to the center of the target bolt.

[0013] In an implementation of the first aspect, the method further includes: rotating the central axis of the tightening tool according to a preset rotation step size so that the tightening tool approaches the horizontal axis along the direction of the side at which the horizontal axis angle between the target bolt and the coaxial image of the working area is the smallest.

[0014] In an implementation of the first aspect, the method further includes: when the deviation gradually decreases and a change in the deviation level is detected, halving the current rotation step length, and controlling the rotation of the central shaft of the tightening tool based on the halved rotation step length.

[0015] In an implementation method of the first aspect, it also includes: when performing three-dimensional reconstruction on the working area of the target bolt, it also includes: determining the positions of the working phase and the protective phase based on the three-dimensional reconstruction data of the working area; determining the safe phase-to-phase insulation distance based on the voltage level; on the straight line connecting the working phase and each of the protective phases, taking the rectangular area enclosed by one times the safe phase-to-phase insulation distance in the left and right directions of the protective phase and twice the safe phase-to-phase insulation distance in the front, back, and up and down directions of the protective phase as a protection area, and generating a three-dimensional virtual fence for the protection area.

[0016] In a second aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the control method for a collaborative robot for bolt tightening high-voltage live equipment as described in any one of the first aspects of the present application.

[0017] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory; the memory stores program instructions; the processor is used to run the program instructions to execute the control method for a collaborative robot for bolt tightening high-voltage live equipment described in any one of the first aspects of the present application.

[0018] As described above, the control method for a collaborative robot for tightening bolts on high-voltage live equipment described in this application has the following beneficial effects:

[0019] This application can precisely control the movement of a collaborative robot that tightens bolts of high-voltage live equipment without disconnecting the power supply, and effectively provide insulation protection between phases. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Shown is a flow chart of a control method for a collaborative robot for tightening bolts on high-voltage live equipment as described in an embodiment of the present application.

[0021] Figure 2 Shown is a flow chart of a method for controlling a collaborative robot for tightening bolts of high-voltage live equipment described in an embodiment of the present application, in which the center of the tightening operator is approximately aligned with the center of the target bolt.

[0022] Figure 3 Shown is a flowchart of adjusting the orientation of the central axis of a tightening tool in a control method for a collaborative robot for tightening bolts on high-voltage live equipment as described in an embodiment of the present application.

[0023] Figure 4 Shown is a flowchart of adjusting the position of a tightening operator in a control method for a collaborative robot for tightening bolts on high-voltage live equipment as described in an embodiment of the present application.

[0024] Figure 5 Shown is a schematic diagram of further adjusting the translation of the tightening operator according to the deviation in the control method for a collaborative robot for tightening bolts of high-voltage live equipment described in an embodiment of the present application.

[0025] Figure 6 Shown is a schematic diagram of further adjusting the rotation of a tightening tool according to deviation in a control method for a collaborative robot for tightening bolts of high-voltage live equipment as described in an embodiment of the present application.

[0026] Figure 7 Shown is a schematic diagram of correcting the center of the tightening operator to align with the center of the target bolt in the control method of the collaborative robot for tightening bolts of high-voltage live equipment described in an embodiment of the present application.

[0027] Figure 8 Shown is a schematic diagram of phase-to-phase insulation protection in a control method for a collaborative robot for tightening bolts on high-voltage live equipment as described in an embodiment of the present application.

[0028] Figure 9 Shown is a schematic diagram of a three-dimensional virtual fence for phase-to-phase insulation protection in the control method of a collaborative robot for bolt tightening high-voltage live equipment described in an embodiment of the present application.

[0029] Figure 10 Shown is a principle block diagram of the electronic device described in an embodiment of the present application. DETAILED DESCRIPTION

[0030] The following describes the embodiments of the present application through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0031] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. Therefore, the illustrations only show components related to the present application and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0032] An embodiment of the present application provides a control method for a collaborative robot for tightening bolts of high-voltage live equipment, which is used to accurately control the movement of the collaborative robot that tightens bolts of high-voltage live equipment without disconnecting the power supply.

[0033] This embodiment uses three-dimensional vision to roughly locate the working area of the target bolt, and uses coaxial vision inside the tightening operator to precisely locate the center of the target bolt and the center of the tightening operator. The movement of the collaborative robot that performs uninterrupted power-on tightening of bolts on high-voltage live equipment can be precisely controlled. At the same time, in this application, the positions of the working phase and the protection phase are determined based on the three-dimensional reconstruction data of the working area, the safe phase-to-phase insulation distance is determined for the protection phase, and a three-dimensional virtual fence is generated, thereby effectively performing insulation protection between phases when the collaborative robot is operating.

[0034] The following is a combination of the appended examples of the present application Figure 1 To the attached Figure 10 , the technical solutions in the embodiments of the present application are described in detail.

[0035] This embodiment provides a control method for a collaborative robot for fastening bolts on high-voltage live equipment. Figure 1 , which is a flow chart showing a control method for a collaborative robot for tightening bolts on high-voltage live equipment according to an embodiment of the present application. Figure 1 As shown, in this embodiment, the control method of the collaborative robot for tightening bolts of high-voltage live equipment includes the following steps S100 to S400.

[0036] S100: Control the collaborative robot to carry the tightening device into the working area of the target bolt in the high-voltage live equipment;

[0037] S200, acquiring a real-time image of the working area, and performing three-dimensional reconstruction of the working area of the target bolt to obtain three-dimensional reconstruction data of the working area; wherein the real-time image of the working area is collected by a camera installed on the outer surface of the tightening tool;

[0038] S300, approximately aligning the center of the tightening tool with the center of the target bolt based on the real-time image of the work area and the three-dimensional reconstruction data of the work area;

[0039] S400, acquiring a coaxial image of the work area, extracting an image of a target bolt from the coaxial image of the work area, and adjusting the tightening tool based on a deviation between the image of the target bolt and the coaxial image of the work area so that the center of the target bolt is aligned with the center of the tightening tool; wherein the coaxial image of the work area is captured by a camera installed inside the tightening tool.

[0040] In this embodiment, after controlling the collaborative robot to carry the tightening operator into the operating area of the target bolt in the high-voltage live equipment, the operating area of the target bolt is three-dimensionally reconstructed through the real-time image of the operating area captured by the camera installed on the outer surface of the tightening operator, and the three-dimensional reconstruction data of the operating area is obtained. The target bolt is roughly positioned based on the three-dimensional reconstruction data. Afterwards, the camera installed inside the tightening operator captures the coaxial image of the operating area, and the target bolt is precisely positioned based on the coaxial image of the operating area, so that the center of the target bolt is aligned with the center of the tightening operator, thereby realizing precise control of the movement of the collaborative robot that performs uninterrupted power tightening on the bolts of the high-voltage live equipment.

[0041] The following is a detailed description of the above-mentioned S100 to S400 in the control method of the collaborative robot for tightening bolts of high-voltage live equipment in this embodiment.

[0042] S100, controlling the collaborative robot to carry the tightening operator into the operating area of the target bolt in the high-voltage live equipment.

[0043] In this embodiment, the tightening device is mounted on the collaborative robot and is used to tighten bolts on high-voltage live equipment. The collaborative robot controls the operation of the tightening device. The collaborative robot controls the tightening device by executing corresponding control instructions generated by the control method for a collaborative robot for tightening bolts on high-voltage live equipment in this embodiment via an external control device.

[0044] For example, the collaborative robot includes multiple control shafts connected in sequence along different directions and rotatable relative to each other. By driving the coordinated movement of each control shaft, the tightening device is moved to the bolt position of the high-voltage live equipment and controlled to perform the tightening operation. The collaborative robot's design with multiple control shafts connected in sequence along different directions enables the robot to move flexibly in three-dimensional space, meeting complex operational requirements. This embodiment does not limit the specific structure of the collaborative robot.

[0045] In this embodiment, the tightening operator includes an insulating tightening rod, a fastening joint, and a bolt positioning device. Specifically, the fastening joint is installed at the front end of the insulating tightening rod, and is used to match and connect with the bolts of the high-voltage live equipment, and tighten the bolts of the high-voltage live equipment by rotating the insulating tightening rod; the rear end of the insulating tightening rod is connected to the collaborative robot, and the movement or rotation of the insulating tightening rod is controlled by the collaborative robot; the bolt positioning device includes a monocular camera installed on the outer surface of the insulating tightening rod, which is used to perform three-dimensional visual positioning of the bolts of the high-voltage live equipment, and a monocular camera installed inside the insulating tightening rod, which is used to coaxially position the center of the bolt of the high-voltage live equipment and the center of the insulating tightening rod.

[0046] In this embodiment, the collaborative robot is controlled to carry a tightening operator into the operating area of the target bolt in the high-voltage live equipment.

[0047] Specifically, in this embodiment, by inputting various control instructions to the collaborative robot, such as start, stop, speed adjustment, rotation direction and rotation angle of each control axis, the collaborative robot is controlled to carry the tightening operator into the operating area of the target bolt in the high-voltage live equipment.

[0048] In addition, in this embodiment, a real-time image of the working area can also be obtained, and based on the obtained real-time image of the working area, it can be determined whether the tightening tool carried by the collaborative robot has entered the working area of the high-voltage live equipment. Specifically, in this embodiment, one implementation method for determining whether the tightening tool carried by the collaborative robot has entered the working area of the high-voltage live equipment based on the obtained real-time image of the working area is:

[0049] An image feature library of high-voltage live equipment and bolts is established in advance, and the real-time image of the working area is compared with the images of the high-voltage live equipment and bolts in the image feature library. When it is detected that the real-time image of the working area matches the image in the image feature library, it is determined that the tightening device carried by the collaborative robot has entered the working area of the high-voltage live equipment.

[0050] For example, real-time images of the work area are acquired. To improve image quality and recognition accuracy, these images are preprocessed, such as through denoising, enhancement, and filtering. An image recognition algorithm is then used to extract the features of the target bolt to be tightened, such as shape, size, color, and texture, from the preprocessed real-time images. These features are then compared with a pre-set bolt feature library to accurately identify the target bolt. Based on the identified target bolt's position and posture, a motion path is planned for the collaborative robot to approach the target bolt, ensuring smooth and accurate approach. Finally, based on the planned path, the collaborative robot's control axes (such as joints) are controlled to move at a specific speed and direction, driving the tightening tool to approach the target bolt and enter the work area.

[0051] In addition, in other embodiments, an image recognition model can also be trained using a pre-established image feature library of high-voltage live equipment and bolts, so that the trained image recognition model can quickly identify real-time images of the work area to quickly determine whether the tightening tool carried by the collaborative robot has entered the work area of the high-voltage live equipment.

[0052] After the collaborative robot enters the operating area of high-voltage live equipment, this embodiment further includes: amplifying the real-time operating image. This amplification of the real-time operating image may employ, for example, an interpolation algorithm to calculate the values of newly added pixels in the amplified image. Common interpolation algorithms include nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation. Bicubic interpolation offers a good balance, achieving a higher magnification while maintaining good image quality.

[0053] Furthermore, the magnification of the real-time operation images can also include super-resolution reconstruction based on deep learning. For example, deep learning techniques such as generative adversarial networks (GANs) or convolutional neural networks (CNNs) can be used for super-resolution reconstruction. After training on large amounts of image data, these networks can significantly improve image resolution without adding excessive noise. By magnifying the real-time operation images, operators can clearly observe the details of the collaborative robot after it enters the work area, which helps improve work efficiency, reduce operational risks, and enhance work quality.

[0054] This embodiment also includes real-time detection of the distance between the collaborative robot's tightening tool and high-voltage live equipment. By comparing the safety distance threshold, it is possible to further accurately determine whether the collaborative robot's tightening tool has entered the operating area of high-voltage live equipment and determine the distance to the high-voltage live equipment, thereby enabling more precise control of the collaborative robot. In this embodiment, when it is confirmed that the collaborative robot's tightening tool is about to enter or has already entered the operating area of high-voltage live equipment, an early warning signal is generated to provide an early warning. The early warning may be provided in the form of sound, symbols, or text.

[0055] S200, obtaining a real-time image of the working area, and performing three-dimensional reconstruction on the working area of the target bolt to obtain three-dimensional reconstruction data of the working area; wherein the real-time image of the working area is collected by a camera installed on the outer surface of the tightening operator.

[0056] In this embodiment, the working area of the target bolt is three-dimensionally reconstructed using real-time images of the working area captured by a camera installed on the outer surface of the tightening operator, three-dimensional reconstruction data of the working area is obtained, and the target bolt is roughly positioned based on the three-dimensional reconstruction data.

[0057] In this embodiment, a camera on the exterior of the tightening device is used to initially locate the target bolt before tightening. The camera, mounted on the exterior of the tightening device with a field of view covering the target bolt area, employs, for example, a wide-angle lens or a monocular camera. The camera provides a quick and approximate location of the target bolt, helping the collaborative robot initially align with the target bolt for more precise positioning later.

[0058] Preferably, at least two cameras are arranged axially staggered on the outer surface of the tightening tool. This arrangement enables the cameras to capture the target bolt's position from different heights (or depths), thereby facilitating more accurate positioning. If ambient lighting is insufficient, a ring-shaped LED light source or structured light projection (such as laser stripes or coded gratings) can be added to the outer surface of the tightening tool to assist in imaging.

[0059] In this embodiment, by controlling the left and right rotation of the tightening operator, a camera installed on the outer surface of the tightening operator scans the working area and obtains three-dimensional reconstruction data of the working area. The three-dimensional reconstruction data of the working area can be used to locate the target bolt with millimeter or sub-millimeter accuracy.

[0060] In this embodiment, a camera mounted on the exterior of the tightening tool simultaneously captures real-time images of the work area during its rotation. To ensure the continuity and integrity of the real-time images of the work area, the camera is preconfigured for frame rate, exposure time, and rotation speed. For example, if the rotation speed is fast, the frame rate should be increased accordingly to avoid image blur.

[0061] The real-time images of the work area collected may have problems such as noise and distortion. In this embodiment, the real-time images of the work area are preprocessed, and the preprocessing includes but is not limited to image denoising, distortion correction, grayscale and other operations to improve image quality.

[0062] In this embodiment, the process of obtaining the three-dimensional reconstruction data of the working area includes:

[0063] 1) Extract feature points, such as corners and edges, from the preprocessed real-time images of the work area, and use feature matching algorithms, such as SIFT, SURF, or ORB, to find matching points between real-time images of the work area from different perspectives.

[0064] 2) Based on the 3D reconstruction algorithm and the feature matching results, the matching points between the real-time images of the work area from different perspectives are 3D reconstructed to obtain the 3D reconstruction data of the work area.

[0065] The 3D reconstruction algorithm can employ multi-view geometry, stereo vision, or deep learning methods. Multi-view geometry methods reconstruct sparse 3D point clouds through feature point matching and camera calibration from multi-view images. Stereo vision methods utilize parallax information from two or more cameras to calculate depth information and reconstruct dense 3D point clouds. Deep learning methods (such as monocular depth estimation) train deep learning models to predict depth information from a single image, then combine multi-view images for 3D reconstruction.

[0066] For example, a stereo vision method is used in combination with feature matching results to perform three-dimensional reconstruction of matching points between real-time images of the work area from different perspectives. The principle of obtaining three-dimensional reconstruction data of the work area is as follows: the stereo vision method calculates depth information through the parallax information of two or more images, thereby reconstructing three-dimensional data, including the following steps:

[0067] 1) Image Correction: Geometrically correct the real-time images of the work area collected from different perspectives to align them on the same plane. This image correction includes camera calibration and stereo image correction. Camera calibration uses a calibration plate to calibrate the camera's intrinsic parameters and distortion parameters. Stereo image correction uses a correction matrix to correct the stereo image pairs.

[0068] 2) Disparity calculation: The depth map is calculated from the disparity map. The disparity map is calculated using a block matching or semi-global matching (SGM) algorithm.

[0069] 3) 3D reconstruction: Use depth information and camera parameters to reconstruct a 3D point cloud.

[0070] The 3D point clouds reconstructed from different perspectives are then fused to form a complete 3D reconstruction of the work area. The ICP (Iterative Closest Point) algorithm can be used to align and optimize the point clouds, improving the accuracy of the 3D reconstruction.

[0071] Specifically, one implementation method for point cloud alignment and optimization using the ICP (Iterative Closest Point) algorithm includes:

[0072] 1) Initial transformation: If the approximate alignment relationship between the two point clouds is known, select an initial transformation matrix (usually the identity matrix), or obtain a rough initial transformation through other methods (such as feature matching).

[0073] 2) Closest point matching: For each point in the source point cloud, find the closest point in the target point cloud to form a point pair. For example, use Open3D's ICP function for closest point matching.

[0074] 3) Calculate the transformation: Minimize the sum of the squared distances between point pairs to find the optimal rotation matrix and translation vector. For example, use Open3D's ICP function to calculate the optimal transformation matrix.

[0075] 4) Update point cloud: Apply the calculated transformation to the source point cloud and update the position of the source point cloud.

[0076] 5) Iteration Termination: When the change in the transformation matrix is less than a matching threshold, or when the number of iterations reaches an upper limit, the iterations terminate. The ICP algorithm's termination criteria are controlled by setting the maximum number of iterations and the convergence matching threshold. The ICP algorithm typically requires multiple iterations to converge to the optimal solution. In practical applications, a larger number of iterations can be used to ensure that the algorithm fully optimizes point cloud alignment.

[0077] If there are multiple sets of point cloud data, you can first perform ICP alignment on each pair of point clouds, and then fuse the aligned point clouds to form a complete 3D work area data. Through multiple alignment and fusion, the accuracy of the reconstructed 3D work area data can be gradually improved.

[0078] In this embodiment, a suitable initial transformation is selected for the application of the ICP algorithm, and a reasonable number of iterations and matching threshold are set to further improve the accuracy and efficiency of the alignment.

[0079] Through step S200, the tightening tool can be controlled to rotate left and right, and the working area can be scanned by a camera for three-dimensional reconstruction, so as to facilitate the subsequent positioning of the target bolt with millimeter or sub-millimeter accuracy through step S300.

[0080] S300: Approximately aligning the center of the tightening tool with the center of the target bolt based on the real-time image of the work area and the three-dimensional reconstruction data of the work area.

[0081] Figure 2 The flowchart of the control method of the collaborative robot for tightening bolts of high-voltage live equipment described in the embodiment of the present application is shown as follows: Figure 2 As shown, in one implementation of this embodiment, the step of approximately aligning the center of the tightening tool with the center of the target bolt based on the real-time image of the work area and the three-dimensional reconstruction data of the work area includes the following steps S310 to S340:

[0082] Step S310 : fitting the plane of the real-time image of the working area using the three-dimensional reconstruction data of the working area.

[0083] In this embodiment, in order to improve computational efficiency, the point cloud of the three-dimensional reconstructed data may be downsampled, and noise points in the point cloud may be removed to improve fitting accuracy.

[0084] In this embodiment, a specific implementation method of fitting the plane of the real-time image of the working area using the three-dimensional reconstruction data of the working area includes:

[0085] 1) Extracting planes from point clouds: For example, use the RANSAC (Random Sample Consensus) algorithm to fit planes in point clouds. RANSAC is a commonly used robust fitting method that effectively handles noise and outliers.

[0086] 2) Calculate the projection matrix of the point cloud to the two-dimensional plane based on the fitted plane equation. Assume that the plane equation is ax + by + cz + d =0, the point cloud can be projected onto z =0 on the plane.

[0087] 3) Fitting a two-dimensional plane: on a two-dimensional plane, a least square method is used to fit the plane of the real-time image of the working area.

[0088] Step S320: adjusting the orientation of the central axis of the tightening tool based on the plane.

[0089] Figure 3 The flowchart of adjusting the direction of the central axis of the tightening tool in the control method of the collaborative robot for tightening bolts of high-voltage live equipment described in the embodiment of the present application is shown. Figure 3 As shown, in an implementation of this embodiment, adjusting the orientation of the central axis of the tightening tool based on the plane includes the following steps S321 to S325.

[0090] Step S321, obtaining a normal vector of the plane based on the fitted plane;

[0091] The equation of a plane is usually expressed as: ax + by + cz + d =0 where ( a , b , c ) is the normal vector of the plane. Therefore, after fitting the plane, the equation of the plane can be obtained, and the normal vector is directly obtained from the coefficients of the equation.

[0092] Step S322: Acquire the angle between the plane normal vector of the real-time image of the working area and the central axis of the tightening tool.

[0093] The direction of the tightening tool's central axis can be represented by its direction vector in three-dimensional space. If the tightening tool's central axis is parallel to the z-axis, its direction vector can be represented as [0, 0, 1]. If the tightening tool's direction is arbitrary, the direction vector must be derived from its position and posture information.

[0094] Use the dot product formula of the vector to calculate the angle between the plane normal vector and the center axis direction vector of the tightening tool. The dot product formula of the vector is:

[0095] , where u and v are the plane normal vector and the center axis vector of the tightening tool respectively. θ is the angle between them, ∥u∥ and ∥v∥ are the moduli of the vectors. Through the above steps, the angle between the plane normal vector of the real-time image of the working area and the central axis of the tightening tool can be calculated.

[0096] Step S323, detect whether the angle is greater than a first difference tolerance threshold: if so, proceed to step S324: change the orientation of the central axis of the tightening tool to the direction of the normal vector of the plane, and update the three-dimensional reconstruction data of the working area of the target bolt based on the real-time image of the current working area; if not, proceed to step S325: maintain the current orientation of the central axis of the tightening tool.

[0097] In this embodiment, the direction of the central axis of the tightening tool is changed to the direction of the normal vector of the plane, which includes:

[0098] 1) Generate the rotation matrix from the direction vector of the central axis of the tightening tool to the normal vector of the plane:

[0099] First, the rotation axis from the direction vector of the central axis of the tightening tool to the normal vector of the plane is determined, and the rotation angle of the direction vector of the central axis of the tightening tool around the rotation axis is calculated. Then, a preset formula (such as the Rodriguez formula) is used to generate a rotation matrix.

[0100] 2) Convert the rotation matrix into the target pose (Euler angles or quaternions) at the end of the collaborative robot's control axis, obtain the collaborative robot's joint control angles, configure the rotation speed, and plan a collision-free rotation path.

[0101] Therefore, in this embodiment, the 3D reconstruction results are used to fit the plane in the current field of view, and the angle between the plane normal vector and the central axis of the tightening tool is calculated. If the angle is greater than a set tolerance threshold, the assisting robot is controlled to change the direction of the central axis of the tightening tool to the direction of the plane normal vector.

[0102] Step S330: Detect the target bolt in the real-time image of the working area using a preset visual detection algorithm.

[0103] In this embodiment, the necessary libraries for a visual detection algorithm (YOLO or other object detection algorithms) are configured, such as OpenCV and PyTorch. Pretrained visual detection models (such as YOLOv5 and YOLOv7) are used as a foundation, and training is performed using a custom bolt dataset. The bolt dataset contains images of bolts under various angles, lighting, and occlusion conditions, with bounding boxes annotated. The trained visual detection model is then used to detect target bolts in real-time images of the work area.

[0104] For example, a real-time image of the work area is fed into the YOLO model, which outputs target bolt detection results, including but not limited to the bolt image's bounding box, category, and confidence score. Detection results classified as "bolt" are filtered, and boxes with low confidence scores (e.g., a threshold < 0.5) are filtered out. In this embodiment, bounding boxes for all detected bolts are drawn on the real-time image of the work area and annotated with confidence scores. Furthermore, each bolt is assigned a unique ID, such as sorted by confidence score or numbered from left to right.

[0105] In this embodiment, target bolts can be selected using a mouse or touchscreen. For example, a user can click a bolt area in the image, and the system will determine whether the clicked location is within a bounding box. Furthermore, this embodiment allows direct selection of target bolts from the real-time image of the work area by entering a number on the keyboard (e.g., entering "1" selects the first bolt).

[0106] Through the above steps, this embodiment can use a visual detection algorithm (such as YOLO) to detect the target bolts on the target plane and allow the user to select the target bolts in an interactive manner.

[0107] Step S340 , controlling the central axis of the tightening tool to translate so that the center of the tightening tool is approximately aligned with the center of the target bolt.

[0108] The translation between the center of the tightening tool and the center of the target bolt is calculated. Based on the calculated translation, a control path is planned for the collaborative robot to achieve translation of the tightening tool so that the center of the tightening tool is approximately aligned with the center of the target bolt.

[0109] To plan the control path for the collaborative robot, a path optimization algorithm based on the artificial potential field method is used to construct a composite potential field function with the bolt center as the attraction domain and surrounding obstacles as the repulsion domain. The optimal translation path is solved using the gradient descent method. This embodiment also incorporates an elastic collision detection algorithm to predict the joint motion limits of the collaborative robot's control axes and avoid control axis interference.

[0110] S400, acquiring a coaxial image of the work area, extracting an image of a target bolt from the coaxial image of the work area, and adjusting the tightening tool based on a deviation between the image of the target bolt and the coaxial image of the work area so that the center of the target bolt is aligned with the center of the tightening tool; wherein the coaxial image of the work area is captured by a camera installed inside the tightening tool.

[0111] In this embodiment, the camera installed inside the tightening operator is used to perform high-precision positioning of the target bolt before the tightening operation. By coaxially capturing the image of the target bolt inside the tightening operator, the exact position and posture of the bolt can be determined based on the deviation between the coaxial axes, thereby adjusting the position and angle of the tightening operator to ensure that the tightening operator can accurately align with the bolt head. The camera installed inside the tightening operator includes at least two cameras arranged along the inner circumferential direction of the insulated tightening connecting rod in the tightening operator. This layout enables the camera to capture the position information of the coaxial target bolt from different angles, thereby achieving precise positioning of the target bolt. The camera can be installed at the front end or rear end inside the tightening operator, or at both the front and rear ends to provide more comprehensive field of view coverage.

[0112] Figure 4 The flowchart of adjusting the position of the tightening operator in the control method of the collaborative robot for tightening bolts of high-voltage live equipment described in the embodiment of the present application is shown. Figure 4 As shown, in one implementation of this embodiment, extracting the image of the target bolt from the coaxial image of the working area and adjusting the tightening tool based on the deviation between the image of the target bolt and the coaxial image of the working area include the following steps S410 to S450.

[0113] Step S410, detecting the center point and directions of each edge of the target bolt in the coaxial image of the working area;

[0114] Step S420, detecting whether the deviation between the center point position of the target bolt and the center point of the coaxial image of the working area is less than a second tolerance threshold;

[0115] Step S430 , detecting whether the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the working area is less than a third tolerance threshold;

[0116] Step S440: If the deviation between the center point of the target bolt and the center point of the coaxial image of the work area is less than a second tolerance threshold, and the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the work area is less than a third tolerance threshold, then the position of the tightening tool is maintained unchanged.

[0117] Step S450: If the deviation between the center point of the target bolt and the center point of the coaxial image of the working area is not less than a second tolerance threshold and / or the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the working area is not less than a third tolerance threshold, the central axis of the tightening tool is controlled to translate based on preset pixels and a preset distance step so that the center of the tightening tool is aligned with the center of the target bolt.

[0118] Figure 5 The diagram shows a method for controlling a collaborative robot for tightening bolts on high-voltage live equipment according to an embodiment of the present application, wherein the translation of the tightening tool is further adjusted according to the deviation. Figure 5 As shown, in one implementation of this embodiment, the further step is further included: pre-dividing the deviation into a plurality of deviation levels according to the size of the deviation; when the deviation gradually decreases and a change in the deviation level is detected, halving the current pixel and the distance step, and controlling the translation of the central axis of the tightening tool based on the halved pixel and the distance step, so that the center of the tightening tool is close to the center of the target bolt.

[0119] Figure 6 The diagram shows a method for controlling a collaborative robot for tightening bolts on high-voltage live equipment according to an embodiment of the present application, wherein the rotation of the tightening tool is further adjusted according to the deviation. Figure 6 As shown, in one implementation of this embodiment, it also includes: rotating the central axis of the tightening tool according to a preset rotation step, so that the tightening tool approaches the horizontal axis along the direction of the side with the smallest angle between the horizontal axis of the coaxial image of the target bolt and the working area.

[0120] In an implementation of this embodiment, the method further includes: when the deviation gradually decreases and a change in the deviation level is detected, halving the current rotation step length, and controlling the rotation of the central shaft of the tightening tool based on the halved rotation step length.

[0121] Figure 7 The diagram shows a method for controlling a collaborative robot for tightening bolts of high-voltage live equipment according to an embodiment of the present application, wherein the center of the tightening tool is aligned with the center of the target bolt. Figure 7 As shown in the figure, after coarse positioning of the target bolt, the center of the tightening tool is approximately aligned with the center of the bolt. However, due to the large field of view used in the 3D reconstruction data, the pixel accuracy is low, and an error of several millimeters may occur. Therefore, further calibration is performed using the camera inside the tightening tool.

[0122] First, the edge of the coaxial image of the working area is extracted, and the cross-sectional graphics of the target bolt in the coaxial image of the working area are detected, for example Figure 7A regular hexagon is extracted from the image, and the center point and orientation of each edge are extracted. If the deviation between the center of the hexagon and the center and bottom of the image does not exceed the tolerance threshold, and if two edges have angles with the horizontal axis of the image that do not exceed the tolerance threshold, the process ends. Otherwise, the tool axis is translated by the preset pixel-distance step coefficient, bringing the image center closer to the bolt center. Each time the deviation sign changes, the pixel-distance step coefficient is halved. The tool axis is rotated by the preset rotation step coefficient, bringing the direction of the edge with the smallest angle between the bolt and the horizontal axis closer to the horizontal axis. Each time the deviation sign changes, the rotation step coefficient is halved. This method allows the center of the tightening tool to be gradually aligned with the center of the target bolt.

[0123] Figure 8 Shown is a schematic diagram of interphase insulation protection in the control method of a collaborative robot for tightening bolts on high-voltage live equipment according to an embodiment of the present application. Figure 8 As shown, in an implementation of this embodiment, the following steps S510 to S530 are also included when the three-dimensional reconstruction of the working area of the target bolt is performed.

[0124] Step S510, determining the positions of the working phase and the protective phase based on the three-dimensional reconstruction data of the working area;

[0125] Step S520, determining a safe phase-to-phase insulation distance based on the voltage level;

[0126] In step S530, on the straight line connecting the working phase and each of the protection phases, a rectangular area enclosed by a distance from the protection phase in the left and right directions that is twice the safe phase-to-phase insulation distance and a distance from the protection phase in the front, back, and up and down directions that is twice the safe phase-to-phase insulation distance is used as a protection area, and a three-dimensional virtual fence is generated for the protection area.

[0127] Figure 9 This diagram shows a three-dimensional virtual fence for phase-to-phase insulation protection in the control method for a collaborative robot for bolt tightening high-voltage live equipment, as described in an embodiment of this application. In this embodiment, during three-dimensional reconstruction, the positions of the working phase and the protection phase can be interactively selected in the reconstruction results to generate a three-dimensional virtual fence:

[0128] First, the safe interphase insulation distance is determined based on the voltage level. On a straight line approximately connecting the three phases, a rectangular block is defined as the protected area, defined by the safe distance to the left and right of the protected phase and twice the safe distance in front, back, and top and bottom. A three-dimensional virtual fence is generated for this protected area. Each time a cobot's motion path is planned, it is ensured that no part of the cobot intrudes into the protected area formed by the three-dimensional virtual fence.

[0129] In this embodiment, according to different target bolt specifications and tightening requirements, corresponding tightening parameters, including tightening torque, tightening speed, number of rotations, etc., are obtained from the database or process parameter table stored in the collaborative robot, and these tightening parameters are converted into control signals that can be recognized and received by the motor of the tightening operator, and the tightening operation is ready to begin. According to the received tightening parameters, the tightening of the target bolt is precisely controlled. During the tightening process, the torque value of the actual tightening output is monitored in real time and compared with the set tightening torque. The torque output is adjusted in time according to the deviation to ensure the accuracy and stability of the torque. At the same time, the rotation angle and number of turns of the tightening operator during the tightening process are monitored to ensure that the preset tightening requirements are met. When the set tightening torque, number of rotations or angle is reached, the tightening operation is stopped.

[0130] After tightening is complete, sensors (such as torque sensors and displacement sensors) can be used to detect and evaluate the tightening results, checking whether the target bolt has achieved the desired tightening effect, including whether the tightening torque is within the qualified range and whether the bolt has reached the specified preload. Key data from the tightening process (such as actual tightening torque, number of revolutions, position information, etc.) is also recorded and stored in the collaborative robot's database for subsequent quality traceability and data analysis. Based on this large amount of tightening result data and feedback, the collaborative robot control unit can optimize and adjust the tightening process parameters to improve the quality and efficiency of the tightening operation.

[0131] When the center of the tightening tool and the center of the target bolt are aligned after coarse and fine positioning, an alignment completion signal is generated. This signal can be a level signal, a pulse signal, or other form of signal, which is used to start distance measurement.

[0132] The tightening tool can be equipped with a laser displacement sensor, ultrasonic sensor, or infrared sensor to detect the distance between the tightening tool and the target bolt. Alternatively, the distance between the tightening tool and the target bolt can be estimated using images of the high-pressure working area and software algorithms.

[0133] Based on the distance detection results, the tightening torque is corrected in real time. If the measured distance is less than the expected value, it may indicate that the target bolt is already in a good preload state. In this case, the tightening torque can be appropriately reduced to avoid overtightening that may cause bolt damage or equipment failure. Conversely, if the measured distance is greater than the expected value, the tightening torque may need to be increased to ensure that the bolt has sufficient preload.

[0134] Furthermore, distance detection results can be used to precisely control the penetration depth of the tightening tool. For example, when the set tightening depth is reached, the tightening operation can be stopped even if the preset tightening torque has not been reached, preventing the bolt from being over-tightened.

[0135] In this embodiment, the working area of the target bolt is three-dimensionally reconstructed by a camera installed on the outside of the tightening operator, and a coaxial image of the working area is collected by an internal camera. The target bolt is precisely positioned based on the coaxial image of the working area. The three-dimensional reconstruction data and the different perspectives and layout methods provided by the coaxial image of the working area cooperate with each other to achieve accurate and efficient positioning of the target bolt, thereby ensuring the smooth progress of the tightening operation.

[0136] The protection scope of the control method for a collaborative robot for bolt tightening high-voltage live equipment described in the embodiment of the present application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing, or replacing steps in the prior art based on the principles of the present application are included in the protection scope of the present application.

[0137] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the control method for a collaborative robot for bolt tightening high-voltage live equipment provided in any embodiment of the present application is implemented.

[0138] In the embodiments of the present application, any combination of one or more storage media may be used. The storage medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, RAM, ROM, an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or component.

[0139] An embodiment of the present application also provides an electronic device. Figure 10 The electronic device 100 provided in an embodiment of the present application is shown as a schematic diagram of the structure. In some embodiments, the electronic device can be a terminal device such as a computer, a server, a smart phone, a PAD, etc.

[0140] like Figure 10 As shown, the electronic device 100 provided in an embodiment of the present application includes a memory 101 and a processor 102 .

[0141] The memory 101 is used to store computer programs; preferably, the memory 101 includes: ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk, etc., various media that can store program codes.

[0142] Specifically, the memory 101 may include computer-readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. The electronic device 100 may further include other removable / non-removable, volatile / non-volatile computer-readable storage media. The memory 101 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present application.

[0143] The processor 102 is connected to the memory 101 and is used to execute the computer program stored in the memory 101 so that the electronic device 100 executes the control method for the collaborative robot for bolt tightening high-voltage live equipment provided in any embodiment of the present application.

[0144] Optionally, the processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0145] Optionally, the electronic device 100 in this embodiment may further include a display 103. The display 103 is communicatively connected to the memory 101 and the processor 102, and is used to display a GUI interaction interface related to the control method of the collaborative robot for tightening bolts on high-voltage live equipment.

[0146] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A control method for a collaborative robot for tightening bolts on high-voltage live equipment, characterized in that: include: Control the collaborative robot to carry the tightening operator into the target bolt operation area of the high-voltage live equipment; Acquire a real-time image of the working area, and perform three-dimensional reconstruction of the working area of the target bolt to obtain three-dimensional reconstruction data of the working area; wherein the real-time image of the working area is collected by a camera installed on the outer surface of the tightening operator; Approximately aligning the center of the tightening tool with the center of the target bolt based on the real-time image of the working area and the three-dimensional reconstruction data of the working area; Acquiring a coaxial image of the work area, extracting an image of a target bolt from the coaxial image of the work area, and adjusting the tightening tool based on a deviation between the image of the target bolt and the coaxial image of the work area so that the center of the target bolt is aligned with the center of the tightening tool; wherein the coaxial image of the work area is acquired by a camera installed inside the tightening tool; The method of approximately aligning the center of the tightening tool with the center of the target bolt based on the real-time image of the working area and the three-dimensional reconstruction data of the working area includes: Fitting the plane of the real-time image of the working area using the three-dimensional reconstruction data of the working area; adjusting the orientation of the central axis of the tightening tool based on the plane; Detecting the target bolt in the real-time image of the working area using a preset visual detection algorithm; Controlling the central axis of the tightening operator to translate so that the center of the tightening operator is approximately aligned with the center of the target bolt; The adjusting the direction of the central axis of the tightening tool based on the plane includes: Obtaining a normal vector of the plane based on the fitted plane; Acquire the angle between the plane normal vector of the real-time image of the working area and the central axis of the tightening tool; Detect whether the angle is greater than a first difference tolerance threshold: If so, the direction of the central axis of the tightening tool is changed to the direction of the normal vector of the plane, and the three-dimensional reconstruction of the working area of the target bolt is updated based on the real-time image of the current working area; If not, maintaining the current orientation of the central axis of the tightening operator; The step of extracting an image of a target bolt from the coaxial image of the working area and adjusting the tightening tool based on a deviation between the image of the target bolt and the coaxial image of the working area includes: detecting the center point and directions of each edge of the target bolt in the coaxial image of the working area; detecting whether a deviation between a center point of the target bolt and a center point of the coaxial image of the working area is less than a second tolerance threshold, and detecting whether an angle between at least two sides of the target bolt and a horizontal axis of the coaxial image of the working area is less than a third tolerance threshold; If the deviation between the center point of the target bolt and the center point of the coaxial image of the working area is less than a second tolerance threshold, and the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the working area is less than a third tolerance threshold, then the position of the tightening tool is maintained unchanged; If the deviation between the center point of the target bolt and the center point of the coaxial image of the working area is not less than a second tolerance threshold and / or the angle between at least two sides of the target bolt and the horizontal axis of the coaxial image of the working area is not less than a third tolerance threshold, then controlling the central axis of the tightening tool to translate based on preset pixels and a preset distance step size so that the center of the tightening tool is close to the center of the target bolt; Also includes: Dividing the deviation into multiple deviation levels in advance according to the size of the deviation; When the deviation gradually decreases and a change in the deviation level is detected, the current pixel and distance step are halved, and the central axis of the tightening tool is controlled to translate based on the halved pixel and distance step, so that the center of the tightening tool is aligned with the center of the target bolt.

2. The control method for a collaborative robot for tightening bolts on high-voltage live equipment according to claim 1, characterized in that: Also includes: The central axis of the tightening tool is rotated according to a preset rotation step so that the tightening tool approaches the horizontal axis along the direction of the side where the angle between the target bolt and the horizontal axis of the coaxial image of the working area is the smallest.

3. The control method for a collaborative robot for tightening bolts of high-voltage live equipment according to claim 2, characterized in that: Also includes: When the deviation gradually decreases and a change in the deviation level is detected, the current rotation step length is halved, and the rotation of the central shaft of the tightening tool is controlled based on the halved rotation step length.

4. The control method for a collaborative robot for tightening bolts of high-voltage live equipment according to claim 1, characterized in that: Also includes: When performing three-dimensional reconstruction on the operating area of the target bolt, the method further includes: Determining the positions of the working phase and the protective phase based on the three-dimensional reconstruction data of the working area; Determine the safe phase-to-phase insulation distance based on the voltage level; On the straight line connecting the working phase and each protection phase, a rectangular area enclosed by one times the safe phase-to-phase insulation distance in the left and right directions of the protection phase and twice the safe phase-to-phase insulation distance in the front, back, and up and down directions of the protection phase is used as a protection area, and a three-dimensional virtual fence is generated for the protection area.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the control method of a collaborative robot for tightening bolts of high-voltage live equipment according to any one of claims 1 to 4 is implemented.

6. An electronic device, characterized in that: The electronic device comprises: processor and memory; The memory stores program instructions; The processor is used to run the program instructions to execute the control method for a collaborative robot for tightening bolts of high-voltage live equipment as described in any one of claims 1 to 4.

Images