Tunnel inner rod replacement robot image recognition and spatial positioning method and system

Abstract

The application provides a tunnel rod changing robot image recognition and spatial positioning method and system, and belongs to the technical field of engineering equipment automation, and comprises the following steps: collecting image information of a target docking area of a drilling machine; recognizing a triangular calibration plate arranged at a target docking position of the drilling machine based on the image, extracting pixel coordinates of a plurality of feature points on the triangular calibration plate in an image coordinate system; converting the feature points into spatial coordinates, and solving a spatial pose of the triangular calibration plate in a preset spatial reference coordinate system; determining a spatial axis direction and an endpoint spatial position of a drill rod at the target docking position based on the spatial pose of the triangular calibration plate and a fixed spatial geometric relationship between the triangular calibration plate and the drill rod, which is established in advance, and converting the spatial pose of the drill rod into a base coordinate system of a mechanical arm to obtain a target docking posture of the drill rod; and based on the target docking posture of the drill rod, a drill rod to be changed is sent to the target position of the drilling machine to complete rod changing. The application can improve the stability and robustness of automatic docking of the drill rod.

Description

Technical Field

[0001] This invention belongs to the field of engineering equipment automation technology, and in particular relates to an image recognition and spatial positioning method and system for a pole-changing robot in a tunnel. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In tunnel excavation, mining, and underground engineering construction, drilling rigs frequently need to replace and connect drill rods depending on the construction conditions. Drill rod replacement is a crucial step in the drilling process, and its efficiency and safety directly impact the overall construction progress and personnel safety. Currently, drill rod replacement is mostly performed manually or semi-automatically on construction sites. Operators move the drill rods from their storage location to the vicinity of the drilling rig, where the rig's clamping device then picks up and connects the rods. This method is labor-intensive, operates in harsh environments, and poses high safety risks. Its efficiency is heavily influenced by the operator's skill level, resulting in poor stability. The difficulty increases significantly in confined, dusty, or low-visibility environments.

[0004] With the development of automation and intelligent technologies, some engineering equipment has begun to incorporate industrial robotic arms to perform drill pipe handling tasks. Existing solutions typically employ one of the following technical approaches: first, using high-precision mechanical positioning and fixing fixtures to strictly constrain the positional relationship between the drilling rig and the drill pipe storage device; second, guiding the robotic arm to complete the drill pipe handling and feeding operations through manual teaching or preset trajectories. However, these solutions still have significant limitations in practical engineering applications. On the one hand, the construction site environment is complex, and the position, posture, and surrounding components of the drilling rig may change due to construction vibrations, installation errors, or uneven foundation settlement, making it difficult to maintain the accuracy of preset trajectories or fixing fixtures over a long period. On the other hand, manual teaching relies on human intervention, making it difficult to meet the needs of continuous operation and unmanned construction. Furthermore, at the target docking position of the drilling rig, structural components are usually installed to guide the drill pipe docking, such as triangular plates, guide blocks, or other geometrically shaped calibration structures. These structural components have advantages in engineering, such as simple structure, relatively stable position, and easy identification. However, in the existing technology, there are few applications of visual recognition and spatial modeling for such structural components, and a mature and systematic solution that can be directly used for automatic drill pipe docking has not yet been formed. Summary of the Invention

[0005] To overcome the shortcomings of the existing technology, this invention proposes an image recognition and spatial positioning method and system for a drill rod changing robot in a tunnel. It can automatically identify the target position of the drilling rig and construct the automatic drill rod docking posture by utilizing on-site visual information, so as to solve the problems of low drill rod docking accuracy, inability to operate continuously and stable calibration, and improve the automation level and engineering applicability of drill rod changing operation.

[0006] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention discloses an image recognition and spatial positioning method for a pole-changing robot in a tunnel, comprising: Acquire image information of the target docking area of ​​the drilling rig; For the image recognition, a triangular calibration plate is set at the target docking position of the drilling rig, and the pixel coordinates of multiple feature points on the triangular calibration plate in the image coordinate system are extracted; The feature points are transformed into spatial coordinates, and the spatial pose of the triangulation plate in the preset spatial reference coordinate system is calculated. Based on the spatial pose of the triangular calibration plate and the pre-established fixed spatial geometric relationship between the calibration plate and the drill rod, the spatial axis direction and endpoint spatial position of the drill rod at the target docking position are determined, and the spatial pose of the drill rod is converted to the robot arm base coordinate system to obtain the target docking posture of the drill rod. Based on the target docking posture of the drill pipe, the drill pipe to be replaced is sent into the target position of the drilling rig to complete the pipe replacement.

[0007] Secondly, this invention discloses an image recognition and spatial positioning system for a tunnel pole-changing robot, comprising: A visual sensing device is used to acquire image information of the drilling rig target docking area; The coordinate extraction module is used to extract the pixel coordinates of multiple feature points on the triangular calibration plate set at the target docking position of the drilling rig in the image coordinate system. The spatial transformation module is used to transform the feature points into spatial coordinates and calculate the spatial pose of the triangulation plate in the preset spatial reference coordinate system. The attitude determination module is used to determine the spatial axis direction and endpoint spatial position of the drill rod at the target docking position based on the spatial pose of the triangular calibration plate and the fixed spatial geometric relationship between the calibration plate and the drill rod, and to convert the spatial pose of the drill rod to the robot arm base coordinate system to obtain the target docking attitude of the drill rod. The docking module is used to send the drill rod to be replaced into the target position of the drilling rig based on the target docking posture of the drill rod to complete the rod replacement.

[0008] Thirdly, the present invention discloses an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when run by the processor, complete the steps of the above-mentioned image recognition and spatial positioning method for the tunnel pole-changing robot.

[0009] Fourthly, the present invention discloses a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-described method for image recognition and spatial positioning of a pole-changing robot in a tunnel.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention achieves adaptive spatial positioning of the target location through visual recognition of the calibration structure at the target position of the drilling rig, reducing the dependence on on-site installation accuracy; and achieves precise drilling rod insertion by constructing the target spatial axis of the drill rod and solving the coordinates of the drill rod endpoints, avoiding reliance on the inverse kinematics model of the robotic arm. This invention eliminates the need for direct visual recognition of the drill pipe body and does not rely on the underlying kinematic modeling and inverse kinematics solution of the robotic arm. It can improve the stability and robustness of automatic drill pipe docking and is suitable for automatic drill pipe supply and docking operations in complex engineering environments.

[0011] The system structure of this invention has a high degree of modularity, is applicable to various drilling rig models and construction scenarios, and has good engineering versatility; it can realize the automation and unmanned operation of drill rod changing, significantly improving the safety and efficiency of operation.

[0012] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0013] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0014] Figure 1 This is a flowchart of the image recognition and spatial positioning method for the tunnel pole-changing robot described in Embodiment 1 of the present invention.

[0015] Figure 2 This is a schematic diagram showing the relative positions of the calibration plate and the drill rod target position as described in Embodiment 1 of the present invention.

[0016] Figure 3 This is a schematic diagram of the rod-changing robot structure described in Embodiment 2 of the present invention.

[0017] In the diagram: 1. Six-axis industrial robotic arm; 2. Left lever box; 3. Control module; 4. Vision sensing device; 5. Right lever box; 6. Movable base. Detailed Implementation

[0018] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0020] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0021] Example 1 In one or more embodiments, a method for image recognition and spatial positioning of a drill rod changing robot in a tunnel is disclosed. This method automatically constructs a drill rod target docking posture model by visually recognizing and spatially positioning the calibration structure at the target location of the drilling rig, thereby enabling the automatic delivery of the drill rod from its storage location to the target location on the drilling rig. Figure 1 As shown, it includes the following steps: Step S1: Control the six-axis industrial robotic arm mounted on the movable base to move to the preset photo-taking position, and collect image information of the drilling rig target docking area through the visual sensing device mounted on the end of the robotic arm or the base.

[0022] Step S2: Process the acquired image, identify the triangular calibration plate set at the target docking position of the drilling rig, and extract the pixel coordinates of multiple feature points on the triangular calibration plate in the image coordinate system.

[0023] The process of image processing, calibration board identification, and feature point extraction specifically includes: Step S2-1: Convert the acquired target area image to the HSV color space, perform mask segmentation using a preset color threshold, and perform morphological opening and closing operations on the mask image to remove background noise and fill the holes inside the target area.

[0024] Step S2-2: Extract the contour of the denoised mask image. Using the area screening method, select candidate contours with an area greater than a set threshold. Then, use a polygon approximation algorithm to obtain the three rough vertex coordinates of the largest triangle contour. At the same time, use a blob detection algorithm to extract the rough pixel coordinates of the independent physical markers set inside the calibration plate.

[0025] Steps S2-3: To improve spatial positioning accuracy, the target region image is converted into a grayscale image. Based on the three coarse vertex coordinates and the coarse pixel coordinates of the independent physical identifier as initial iteration points, the gradient is calculated on the grayscale image to perform sub-pixel level feature extraction, thereby obtaining the sub-pixel high-precision coordinates of the three vertices and the independent physical identifier. Specifically, it includes the following steps: (1) Coarse vertex coordinates obtained by polygon approximation algorithm Centered on the image, a local search window of a predetermined size is set on the grayscale image, and the gradient operator is used to calculate the value of each pixel within the search window. ,exist direction and The first-order partial derivative of the direction yields the spatial gradient vector at that point. .

[0026] (2) Gradient features in the local neighborhood of the corner point, set any point in the search window gradient vector And from that point to the true sub-pixel corner vector They are mutually orthogonal, meaning they satisfy the mathematical constraint that their dot product is zero. .

[0027] (3) Construct a diagram of the real sub-pixel corners by combining all pixels within the search window. The system of overdetermined linear equations is used. The least squares method is employed to solve the system of equations, calculate the sub-pixel level displacement compensation of the corner points, and update the current corner point coordinates accordingly.

[0028] (4) Use the updated corner coordinates as the new initial iteration point and repeat steps (1)-(3). When the displacement compensation calculated in two consecutive iterations is less than the preset small threshold or the maximum allowed number of iterations is reached, stop the iteration and output the final converged coordinates as the sub-pixel high-precision coordinates of the vertex.

[0029] (5) For an independent physical marker set inside the calibration board, the sub-pixel high-precision center coordinates of the physical marker are calculated using the gray-scale centroid method with its coarse pixel coordinates as the center.

[0030] Step S2-4: Use the sub-pixel high-precision coordinates of the three vertices and the independent physical identifier as four coplanar and non-collinear feature points for PnP pose calculation.

[0031] Specifically, based on the pixel coordinates of the aforementioned feature points, combined with the triangle ruler... Given the geometric model, the PnP method is used to solve the triangle in the camera coordinate system. pose in According to the rigid body constraint relationship, the direction of the drill pipe axis is parallel to the direction of the height line of the triangular plate.

[0032] in, Let be the rotation matrix of the triangulation plate coordinate system relative to the camera coordinate system. Let be the translation vector of the triangulation plate coordinate system relative to the camera coordinate system.

[0033] Step S3: Based on the calibration parameters of the camera device, convert the pixel coordinates of the feature points into three-dimensional spatial coordinates in the camera coordinate system, and calculate the spatial pose of the triangulation plate in the preset spatial reference coordinate system.

[0034] In this invention, the drill rod and the isosceles triangle plate positioned at the target docking point of the drilling rig together form a rigid body system, and there is a fixed spatial geometric relationship between the two. Specifically, the axis of the drill rod is parallel to the height of the isosceles triangle plate, but they do not coincide in space, and there is a fixed three-dimensional spatial offset relationship between them.

[0035] like Figure 2 As shown, the relative positional relationship between the drill pipe and the isosceles triangle was calibrated during the system installation and commissioning phase, and the calibration result remained unchanged during subsequent operations. This invention achieves indirect sensing of the drill pipe's spatial attitude through this rigid body constraint relationship.

[0036] Before the task is executed, a hand-eye calibration will be performed during system installation and debugging to obtain the camera coordinate system. coordinate system of the robotic arm end effector The fixed external parameter relationship between them is The results directly determine the accuracy and stability of the system's online operation.

[0037] In the isosceles triangle coordinate system In the above, the three vertices of an isosceles triangle are defined as follows: Using the altitude direction of the isosceles triangle as a reference direction, the unit vector of the altitude direction is defined as... The direction vector In the triangular coordinate system The direction of the drill pipe axis remains constant and is used for subsequent construction.

[0038] In the triangular coordinate system The position of the drill pipe near the end of the triangular plate is directly calibrated, and its coordinates are expressed as follows: The coordinates mentioned above include all spatial offsets of the drill pipe relative to the triangular plate in the front-back, left-right, and up-down directions, where This indicates that the drill pipe and the triangular plate are not coplanar.

[0039] Step S4: Based on the spatial pose of the triangular calibration plate and the pre-established fixed spatial geometric relationship between the calibration plate and the drill rod, i.e., the preset structural offset relationship, such as... Figure 2 The drill rod and the triangular plate have a fixed offset relationship established during installation and will not be changed. The spatial axis direction and endpoint spatial position of the drill rod at the target docking position are determined, and the spatial pose of the drill rod is converted to the robot arm base coordinate system to obtain the target docking posture of the drill rod.

[0040] Step S4-1: Based on the pose of the triangulation plate obtained in step S2 and the spatial geometric relationship obtained in step S3, the unit direction vector of the drill pipe's spatial axis in the camera coordinate system is: .

[0041] Step S4-2, in the triangular coordinate system Below, the coordinates of the starting end point of the drill pipe are known. By transforming the coordinates, we can convert them to the camera coordinate system:

[0042] in, The coordinates of the starting end point of the drill rod are in the camera coordinate system.

[0043] In the triangular coordinate system The coordinates of the other end of the drill pipe are first calculated using length relationships:

[0044] in, The coordinates of the drill pipe end point are: L This refers to the drill pipe length, which is the actual length of a drill pipe in the real world.

[0045] Step S4-3: Transform the endpoint coordinates to the camera coordinate system:

[0046] in, These are the endpoint coordinates of the camera coordinate system.

[0047] At this point, both ends of the drill pipe and the direction of the drill pipe axis are all in the camera coordinate system. The details were fully confirmed.

[0048] Step S4-5: In Eye-in-Hand installation mode, the transformation from the camera coordinate system to the robot arm base coordinate system is as follows:

[0049] in, Ideally, the real-time end-effector pose of the robotic arm should be read from the robotic arm control cabinet in real time. To establish a fixed extrinsic parameter relationship between the camera coordinate system and the robotic arm's end effector coordinate system, preferably, a hand-eye calibration is performed during system installation and debugging before task execution. Using a standard calibration board, the robotic arm, carrying the camera, is controlled to change 10 to 20 different poses above the calibration board and take pictures for each pose. The visual algorithm extracts features from the calibration board and combines this with the robotic arm's pose during each picture capture to accurately solve for the fixed extrinsic parameter relationship. The matrix is ​​then saved in the control program's file; this entire process is a common robotic arm calibration procedure.

[0050] The coordinates of the drill pipe's starting point in the robot arm's base coordinate system are:

[0051] The coordinates of the drill pipe end point in the robot arm's base coordinate system are:

[0052] The direction of the drill pipe axis is:

[0053] in, This is the rotation matrix extracted from the transformation matrix.

[0054] The above method can achieve stable output. drill pipe end point drill pipe axis direction The above results output the coordinates of the drill rod's starting point, the coordinates of its ending point, and the direction of its axis in the robot arm's base coordinate system. These serve as input parameters for the robot arm's motion planning, guiding the robot arm to complete the drill rod gripping, attitude adjustment, and feeding operations.

[0055] Step S5: Control the six-axis industrial robotic arm to grab the drill rod from the drill rod storage device according to the target docking posture of the drill rod, and guide the drill rod to be sent into the target position of the drilling machine along the target docking posture. After the drill rod is sent in, keep the drill rod posture stable and wait for the drilling machine clamping device to complete the drill rod clamping. Then control the robotic arm to release the drill rod and return to the preset initial position.

[0056] Specifically, based on the spatial coordinates of the two ends of the drill rod in the robotic arm's base coordinate system, the target gripping posture and feeding direction of the drill rod are determined. The robotic arm first moves to the pre-gripping posture above the drill rod storage device to complete the drill rod clamping action; then, the drill rod posture is adjusted to ensure that the drill rod axis is consistent with the target docking axis. After the drilling rig completes the clamping, the robotic arm releases the drill rod and retracts along a safe path to the initial standby position, completing one full automatic drill rod changing operation cycle.

[0057] In other embodiments, the visual sensing device may employ a multi-camera configuration to improve the robustness of target recognition. The number and arrangement of the drill pipe storage devices can be adjusted according to construction needs, and this invention does not limit this.

[0058] Through the above specific implementation methods, the present invention realizes automatic identification, grasping and docking of drill rods based on visual recognition and spatial positioning in complex engineering environments, and has good engineering practicality and promotion value.

[0059] This invention utilizes a visual perception device to acquire image information of the drilling rig target area in complex engineering environments. By identifying and spatially modeling the target calibration structure, it calculates the drill rod target docking posture and guides an industrial robotic arm to automatically grasp, transport, and deliver the drill rod from its storage location to the drilling rig target position. It is applicable to automatic drill rod changing and supply operations in underground engineering, mining engineering, tunnel construction, and geological exploration scenarios.

[0060] Example 2 In one or more embodiments, an image recognition and spatial positioning system for a pole-changing robot in a tunnel is disclosed, such as... Figure 3 As shown, it specifically includes a movable base 6, a six-axis industrial robotic arm 1, a drill pipe storage device, a vision sensing device 4, a target calibration plate, and a control module 3. Specifically: The movable base is used to support the six-axis industrial robotic arm and related functional modules. It is equipped with a power module, a control cabinet, and at least one drill rod storage device. In this embodiment, the drill rod storage device is preferably located on the left and right sides of the movable base, including a left rod box 2 and a right rod box 5, so that the robotic arm can flexibly select the rod retrieval path according to the target location.

[0061] A six-axis industrial robotic arm is mounted on the upper part of a movable base, serving as the actuator in the automatic drill rod docking process. It is used to perform the gripping, handling, posture adjustment, and feeding operations of the drill rod. It should be noted that this invention does not involve the specific structural form, driving method, or inverse kinematics solution method of the six-axis industrial robotic arm.

[0062] The visual sensing device is used to acquire image information of the target docking area of ​​the drilling rig and the drill rod storage area. The visual sensing device adopts an eye-in-hand installation method and is fixedly installed at the end of the robotic arm, changing in real time with the posture of the end of the robotic arm.

[0063] The target calibration plate is set at the target docking position of the drilling rig to provide a spatial positioning reference for drill pipe docking. The target calibration plate is a structural component with known geometric dimensions and feature point distribution, which remains relatively stable during drilling rig operation.

[0064] A visual sensing device is used to acquire image information of the target docking area of ​​the drilling rig.

[0065] The control module is used to execute the above-mentioned image recognition and spatial positioning method for the tunnel pole-changing robot, including: The coordinate extraction module is used to extract the pixel coordinates of multiple feature points on the triangular calibration plate set at the target docking position of the drilling rig in the image coordinate system. The spatial transformation module is used to transform the feature points into spatial coordinates and calculate the spatial pose of the triangulation plate in the preset spatial reference coordinate system. The attitude determination module is used to determine the spatial axis direction and endpoint spatial position of the drill rod at the target docking position based on the spatial pose of the triangular calibration plate and the fixed spatial geometric relationship between the calibration plate and the drill rod, and to convert the spatial pose of the drill rod to the robot arm base coordinate system to obtain the target docking attitude of the drill rod. The docking module is used to send the drill rod to be replaced into the target position of the drilling rig based on the target docking posture of the drill rod to complete the rod replacement.

[0066] In this embodiment, the drill pipe is a rigid body with a known constant length L. The drill pipe does not deform during the docking process, and its geometric shape and dimensional parameters remain unchanged throughout the entire operation cycle.

[0067] Example 3 This embodiment provides an electronic device, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, they complete the steps of the above-described method for image recognition and spatial positioning of a pole-changing robot in a tunnel.

[0068] Example 4 This embodiment provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, they complete the steps of the above-described method for image recognition and spatial positioning of a pole-changing robot in a tunnel.

[0069] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0070] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0071] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0072] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for image recognition and spatial positioning of a pole-changing robot in a tunnel, characterized in that, include: Acquire image information of the target docking area of ​​the drilling rig; For the image recognition, a triangular calibration plate is set at the target docking position of the drilling rig, and the pixel coordinates of multiple feature points on the triangular calibration plate in the image coordinate system are extracted; The feature points are transformed into spatial coordinates, and the spatial pose of the triangulation plate in the preset spatial reference coordinate system is calculated. Based on the spatial pose of the triangular calibration plate and the pre-established fixed spatial geometric relationship between the calibration plate and the drill rod, the spatial axis direction and endpoint spatial position of the drill rod at the target docking position are determined, and the spatial pose of the drill rod is converted to the robot arm base coordinate system to obtain the target docking posture of the drill rod. Based on the target docking posture of the drill pipe, the drill pipe to be replaced is sent into the target position of the drilling rig to complete the pipe replacement.

2. The image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in claim 1, characterized in that, The identification of the triangular calibration plate set at the target docking position of the drilling rig, and the extraction of the pixel coordinates of multiple feature points on the triangular calibration plate in the image coordinate system, specifically includes: The acquired target area image is converted to the HSV color space, and a mask segmentation is performed using a preset color threshold. Then, morphological opening and closing operations are performed on the mask image to remove noise. The contour of the denoised mask image is extracted, and the coordinates of the three rough vertices of the largest triangle contour are obtained by area filtering and polygon approximation algorithm; at the same time, the rough pixel coordinates of the independent physical markers set inside the calibration plate are extracted by using a blob detection algorithm. The target region image is converted into a grayscale image. Based on the three coarse vertex coordinates and the coarse pixel coordinates of the independent physical identifier, sub-pixel level feature extraction is performed on the grayscale image to obtain the sub-pixel high-precision coordinates of the three vertices and the independent physical identifier. The sub-pixel high-precision coordinates of the three vertices and the independent physical identifier are used together as four coplanar and non-collinear feature points for PnP pose calculation.

3. The image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in claim 1, characterized in that, The step of determining the spatial axis direction and endpoint spatial position of the drill rod at the target docking position based on the spatial pose of the triangulation plate and the pre-established fixed spatial geometric relationship between the calibration plate and the drill rod is specifically as follows: The unit direction vector of the drill pipe's spatial axis in the camera coordinate system is: in, This is the rotation matrix of the triangulation plate coordinate system relative to the camera coordinate system; This is the unit vector in the direction of the elevation line of the triangulation plate; In the triangular coordinate system, the coordinates of the endpoints are first calculated using length relationships: in, The coordinates of the drill pipe end point are: The coordinates of the starting end of the drill pipe are: L This refers to the length of the drill pipe; Transform the endpoint coordinates to the camera coordinate system: in, These are the endpoint coordinates of the camera coordinate system.

4. The image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in claim 1, characterized in that, The transformation of the drill rod's spatial pose to the robot arm's base coordinate system is as follows: in, Real-time end-effector pose of the robotic arm; This represents the fixed external parameter relationship between the camera coordinate system and the end effector coordinate system of the robotic arm.

5. The image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in claim 4, characterized in that, The fixed external parameter relationship between the camera coordinate system and the end effector coordinate system of the robotic arm is obtained through hand-eye calibration.

6. The image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in claim 1, characterized in that, The drill pipe target docking posture includes: The coordinates of the drill pipe's starting point in the robot arm's base coordinate system are: The coordinates of the drill pipe end point in the robot arm's base coordinate system are: The direction of the drill pipe axis is: in, To ensure stable output, The end point of the drill pipe, The direction of the drill pipe axis. Let be the transformation matrix from the camera coordinate system to the robot arm's base coordinate system. This is the rotation matrix extracted from the transformation matrix. It is the unit direction vector of the spatial axis direction in the camera coordinate system.

7. The image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in claim 1, characterized in that, The process of sending the drill pipe to be replaced into the target position of the drilling rig based on the target docking posture of the drill pipe to complete the pipe replacement is as follows: Based on the spatial coordinates of the two ends of the drill rod in the base coordinate system of the robotic arm, determine the target grasping posture and feeding direction of the drill rod; The robotic arm first moves to the pre-grabbing position above the drill rod storage device to complete the drill rod clamping action; then the drill rod posture is adjusted to make the drill rod axis consistent with the target docking axis. After the drilling rig clamps the drill rod, the robotic arm releases the drill rod and retracts along a safe path to the initial standby position, completing a full automatic rod changing operation cycle.

8. An image recognition and spatial positioning system for a pole-changing robot in a tunnel, characterized in that, include: A visual sensing device is used to acquire image information of the drilling rig target docking area; The coordinate extraction module is used to extract the pixel coordinates of multiple feature points on the triangular calibration plate set at the target docking position of the drilling rig in the image coordinate system. The spatial transformation module is used to transform the feature points into spatial coordinates and calculate the spatial pose of the triangulation plate in the preset spatial reference coordinate system. The attitude determination module is used to determine the spatial axis direction and endpoint spatial position of the drill rod at the target docking position based on the spatial pose of the triangular calibration plate and the fixed spatial geometric relationship between the calibration plate and the drill rod, and to convert the spatial pose of the drill rod to the robot arm base coordinate system to obtain the target docking attitude of the drill rod. The docking module is used to send the drill rod to be replaced into the target position of the drilling rig based on the target docking posture of the drill rod to complete the rod replacement.

9. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, complete the image recognition and spatial positioning method for a pole-changing robot in a tunnel as described in any one of claims 1-7.

Images