Deburring workstation for inner hole detection and control method thereof

Abstract

The invention relates to the technical field of machining, in particular to a deburring work station for inner hole detection and a control method of the deburring work station. The workstation comprises a machine table, a cutter, an image acquisition device, a robot and a controller. The cutter is used for removing inner hole burrs; the image acquisition device is used for acquiring an inner hole image; the robot is arranged on the machine table and is used for picking up the cutter and the image acquisition device; the controller is electrically connected with the image collecting device and the robot and used for determining the working state of the robot through the obtained inner hole image. According to the deburring work station for inner hole detection and the control method of the deburring work station, automatic deburring of the robot is achieved, manual detection in the prior art is replaced, automatic deburring and burr detection of stainless steel, aluminum alloy and other materials are achieved, the time for transferring workpieces to the work station is shortened, and the detection efficiency and quality are improved.

Description

Technical Field

[0001] This invention relates to the field of machining technology, and in particular to a deburring workstation for internal hole inspection and its control method. Background Technology

[0002] In the field of machining, burrs are easily generated at the intersection of deep holes in ductile materials such as stainless steel and aluminum alloys when machining deep holes with a depth-to-diameter ratio greater than 10. Burrs can affect assembly performance and dimensional accuracy, and in severe cases, can even lead to the failure of the entire product. Therefore, burr removal and inspection are crucial to product quality. Due to the inherent properties of ductile materials, the root of burrs at intersections usually has a certain degree of toughness, causing the deburring tool to bend or fold rather than break the burr during the deburring process, making it difficult to remove it in one go. Therefore, whether deburring is done manually or by machine, it is usually necessary to use an endoscope to inspect the intersection. By visually observing the image returned by the endoscope, it is possible to determine whether burrs exist in the inner hole and their specific location and size. However, this inspection method has several drawbacks, as follows: First, manual inspection is labor-intensive. Since internal bore inspection typically requires operators to maintain a specific posture for extended periods, constantly focusing their eyes on the endoscopic image, it easily leads to operator fatigue, thus affecting inspection efficiency and quality. Moreover, for mass-produced parts, this repetitive inspection work can severely deplete the operator's physical and mental energy.

[0003] Secondly, manual inspection is highly subjective. Different operators may have different standards for judging burrs due to differences in experience, eyesight, and other factors, which can easily lead to inaccurate inspection results. For example, some operators may consider minor burrs as not affecting product performance and ignore them, while others may believe they need to be addressed. This subjective difference can affect the stability of product quality.

[0004] Furthermore, manual inspection is inefficient. Due to the limited speed of manual operation, each internal hole requires a certain amount of time to inspect, which is difficult to meet production demands in a fast-paced environment and can easily lead to production bottlenecks. Additionally, manual inspection requires additional dedicated inspection personnel, increasing production costs.

[0005] Finally, the traditional inspection and deburring processes are independent of each other. After burrs are detected, the parts need to be transferred to deburring equipment for processing. This not only increases the complexity of the production process, but may also cause secondary damage during the transfer of parts. Summary of the Invention

[0006] This invention provides a deburring workstation for internal hole inspection and its control method, which solves the defects of the prior art that uses visual observation of an endoscope to judge the condition of internal holes, such as high labor intensity, inaccurate detection and low work efficiency.

[0007] This invention provides a deburring workstation for internal hole inspection, comprising: Machine tool; A cutting tool, mounted on the machine base, is used to remove burrs from the inner hole; An image acquisition device, mounted on the machine base, is used to acquire images of the inner hole; A robot is mounted on the machine platform and is used to pick up the cutting tool and the image acquisition device. The controller is electrically connected to both the image acquisition device and the robot, and is used to determine the robot's working state based on the acquired internal hole images.

[0008] The deburring workstation for internal hole inspection provided by the present invention further includes: A tool holder is mounted on the machine base, and both the cutting tool and the image acquisition device are mounted on the tool holder.

[0009] According to the deburring workstation for internal hole inspection provided by the present invention, the tool holder is provided with at least two through holes; The cutting tool is provided with a first connecting part, and the image acquisition device is provided with a second connecting part. Both the first connecting part and the second connecting part are located on the through hole, and both the first connecting part and the second connecting part are suitable for being picked up by the robot.

[0010] According to the deburring workstation for internal hole inspection provided by the present invention, the image acquisition device includes: A wireless module, connected to the controller via a signal, is used to transmit images of the inner hole; An endoscope, connected to the wireless module, is used to acquire images of the internal opening.

[0011] According to the deburring workstation for internal hole inspection provided by the present invention, the robot includes: Six-axis robot; An end effector, located at the movable end of the six-axis robot, is used to pick up the cutting tool and the image acquisition device.

[0012] The present invention also provides a control method for a deburring workstation for internal hole detection according to the present invention, comprising: S1. Establish the robot coordinate system and mark the initial position of the robot, the position of the workpiece, and the positions of the cutting tool and the image acquisition device in the robot coordinate system; S2. Based on the robot's initial position, the tool's position, and the workpiece's position in the robot coordinate system, control the robot to pick up the tool and move the tool to the workpiece to perform deburring according to the preset process flow. S3. Based on the robot's position, the image acquisition device's position, and the workpiece's position in the robot coordinate system, control the robot to pick up the image acquisition device and move the image acquisition device to the workpiece to acquire the inner hole image. S4. Process the internal hole image and determine the robot's working status based on the internal hole image.

[0013] According to the control method for the deburring workstation provided by the present invention, step S4 specifically includes: S41. Identify the internal hole image and determine if any defects exist; S42. If there is a defect in the inner hole, control the robot to pick up the tool and repeat the deburring process. If there are no defects in the inner hole, the deburring process is completed and the workpiece is removed.

[0014] According to the control method for the deburring workstation provided by the present invention, step S41 specifically includes: S411. Establish the camera coordinate system and preprocess the inner hole image; S412. Use the target detection model to reason about the preprocessed image to obtain the predicted bounding box and segmentation mask of the defect target; S413. Analyze the results output by the model; S414. Extract the contour information of the defect region from the segmentation mask; S415. For different types of defects, algorithms are used for optimization. S416. Using depth information, the two-dimensional defect contour is mapped to three-dimensional space to obtain its three-dimensional coordinates in the camera coordinate system. S417. Calculate the spatial location and orientation information of the defect based on the three-dimensional coordinates, and use the extrinsic parameter matrix to transform the position information of the defect from the camera coordinate system to the robot coordinate system in order to locate and process the defect.

[0015] According to the control method of the deburring workstation provided by the present invention, step S415 specifically includes: ellipse fitting for hole features, including: S41511. For the extracted hole contour, calculate its major axis and minor axis lengths, and evaluate the shape characteristics of the contour; S41512. Use an ellipse fitting algorithm to fit the contour and obtain the center position, major axis, minor axis and rotation angle of the hole. S41513. Correct the parameters of the fitted ellipse according to the preset axis ratio range.

[0016] According to the control method for a deburring workstation provided by the present invention, step S415 specifically includes: contour refinement for burr defects, including: S41521. Calculate the local curvature of each point on the contour and identify the key feature points and abrupt change points of the contour. S41522. Based on the curvature value, select the key points of the burr area and remove noise and interference points; S41523. Use spline curves to fit the selected key points and generate contour lines.

[0017] This invention provides a deburring workstation for internal hole inspection, comprising: a machine base, a cutting tool, an image acquisition device, a robot, and a controller. The cutting tool is mounted on the machine base for removing burrs from internal holes; the image acquisition device is mounted on the machine base for acquiring images of the internal holes; the robot is mounted on the machine base and is used to pick up the cutting tool and the image acquisition device; the controller is electrically connected to both the image acquisition device and the robot, and is used to determine the robot's working state based on the acquired internal hole images. This invention provides a deburring workstation for internal hole inspection, which achieves automatic deburring by using a robot to pick up the cutting tool and the image acquisition device, acquiring internal hole images using the image acquisition device, and using the controller to determine the robot's working state based on the internal hole images. This replaces manual inspection in the prior art, enabling automatic deburring and burr detection of materials such as stainless steel and aluminum alloys, reducing workpiece transfer time, and improving inspection efficiency and quality.

[0018] This invention provides a control method for a deburring workstation for internal hole inspection. The robot picks up the cutting tool and an image acquisition device, which acquires an image of the internal hole. The controller determines the robot's working state based on the image, thereby achieving automatic deburring by the robot. This replaces manual inspection in the prior art and enables automatic deburring and burr inspection of materials such as stainless steel and aluminum alloys. It reduces the time for workpiece transfer to the workstation and improves inspection efficiency and quality. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the overall structure of a deburring workstation for internal hole detection provided in one embodiment of the present invention.

[0021] Figure 2This is a schematic diagram of the structure of an image acquisition device provided in one embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of the tool holder, tool, and image acquisition device provided in one embodiment of the present invention.

[0023] Figure 4 This is a flowchart illustrating the control method of the deburring workstation for internal hole detection provided in one embodiment of the present invention.

[0024] Figure 5 This is a flowchart illustrating step S4 provided in one embodiment of the present invention.

[0025] Figure 6 This is a flowchart illustrating step S41 provided in one embodiment of the present invention.

[0026] Figure 7 This is one of the flowcharts illustrating step S415 provided in one embodiment of the present invention.

[0027] Figure 8 This is one of the flowcharts illustrating step S415 provided in one embodiment of the present invention.

[0028] Figure label: 1. Robot; 2. Machine base; 3. Screen; 4. Button box; 5. Workpiece; 6. Tool holder; 7. Electric spindle; 8. Quick-change tool holder; 9. Wireless module; 10. Endoscope; 11. Cutting tool; 12. Image acquisition device. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0030] In the description of this embodiment, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this embodiment and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this embodiment.

[0031] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this embodiment, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0032] In this embodiment, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this embodiment based on the specific circumstances.

[0033] In embodiments of the present invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0034] The following is combined with Figures 1-3 This invention describes a deburring workstation for internal hole inspection. The deburring workstation includes: a machine base 2, a cutting tool 11, an image acquisition device 12, a robot 1, and a controller.

[0035] The tool 11 is mounted on the machine base 2 and is used to remove burrs from the inner hole; the image acquisition device 12 is mounted on the machine base 2 and is used to acquire images of the inner hole; the robot 1 is mounted on the machine base 2 and is used to pick up the tool 11 and the image acquisition device 12; the controller is electrically connected to the image acquisition device 12 and the robot 1 respectively and is used to determine the working status of the robot 1 by acquiring the inner hole images.

[0036] Specifically, the machine base 2 serves as the supporting structure for the entire workstation and is used to place the workpiece 5. It is equipped with a button box 4 containing operation buttons. The cutting tool 11, image acquisition device 12, robot 1, controller, and workpiece 5 are all mounted on the machine base 2. The machine base 2 can adopt a platform structure, with casters and a height-adjustable support rod at its bottom, providing movement, support, and lifting functions. The machine base 2 has internal space to accommodate the controller and wiring.

[0037] Specifically, the cutting tool 11 has a cutting head for removing burrs from the inner hole; the image acquisition device 12 is used to acquire an image of the inner hole, the controller receives the image of the inner hole, and makes a judgment based on the image of the inner hole to determine whether it is necessary to continue removing the burrs from the inner hole. If it is necessary to continue removing the burrs from the inner hole, the robot 1 picks up the cutting tool 11 to further remove the burrs from the inner hole; if it is not necessary to continue removing the burrs, the workpiece 5 is qualified, the workpiece 5 is taken away, and the deburring process is completed.

[0038] This invention provides a deburring workstation for internal hole inspection, comprising: a machine base 2, a cutting tool 11, an image acquisition device 12, a robot 1, and a controller. The cutting tool 11 is mounted on the machine base 2 for removing burrs from internal holes; the image acquisition device 12 is mounted on the machine base 2 for acquiring images of the internal holes; the robot 1 is mounted on the machine base 2 and is used to pick up the cutting tool 11 and the image acquisition device 12; the controller is electrically connected to both the image acquisition device 12 and the robot 1, and is used to determine the working state of the robot 1 based on the acquired internal hole images. This invention provides a deburring workstation for internal hole inspection, where the robot 1 picks up the cutting tool 11 and the image acquisition device 12, the image acquisition device 12 acquires internal hole images, and the controller determines the working state of the robot 1 based on the internal hole images, thereby achieving automatic deburring by the robot 1. This replaces manual inspection in the prior art, realizing automatic deburring and burr inspection of materials such as stainless steel and aluminum alloys, reducing the time for transferring workpieces to different workstations, and improving inspection efficiency and quality.

[0039] In one embodiment of the present invention, the deburring workstation for internal hole detection further includes: a tool holder 6, which is disposed on the machine base 2, and both the cutting tool 11 and the image acquisition device 12 are disposed on the tool holder 6. Preferably, the tool holder 6 is provided with at least two through holes; the cutting tool 11 is provided with a first connecting part, and the image acquisition device 12 is provided with a second connecting part, both the first connecting part and the second connecting part are disposed on the through holes, and both the first connecting part and the second connecting part are suitable for being picked up by the robot 1.

[0040] In one embodiment of the present invention, the first connecting part and the second connecting part adopt the same quick-change tool holder 8, for example, it can be a ring-shaped quick-change tool holder 8. On the one hand, it ensures that the robot 1 picks up and positions accurately, can quickly change tools, and the angle between the tool 11 and the image acquisition device 12 is consistent, resulting in better detection effect. On the other hand, it can be adapted to the various through holes on the tool holder 6. Preferably, the quick-change tool holder 8 of the tool 11 is located at its top, and the quick-change tool holder 8 of the image acquisition device 12 is located at its top.

[0041] In one embodiment of the present invention, the image acquisition device 12 includes a wireless module 9 and an endoscope 10. The wireless module 9 is signal-connected to the controller and is used to transmit images of the internal cavity; the endoscope 10 is connected to the wireless module 9 and is used to acquire images of the internal cavity. Specifically, the endoscope 10 acquires images of the internal cavity and transmits these images to the controller via the wireless module 9. Preferably, the wireless module 9 may be a Bluetooth module, a wireless network module, or a 5G network module, etc.

[0042] In one embodiment of the present invention, the image acquisition device 12 consists of three parts: a second connecting part, a wireless module 9, and an endoscope 10. The second connecting part is a quick-change tool holder 8, and the quick-change tool holder 8, the wireless module 9, and the endoscope 10 are connected in sequence from top to bottom to form an integral structure. When not in use, it is suspended on the tool holder 6 by the quick-change tool holder 8. When in use, the quick-change tool holder 8 is picked up by the electric spindle 7 at the end of the robot 1. Similarly, the tool 11 consists of two parts: a tool head and a first connecting part, and the first connecting part is the quick-change tool holder 8. The quick-change tool holder 8 and the tool head are connected in sequence from top to bottom to form an integral structure. The electric spindle 7 at the end of the robot 1 can also pick up the quick-change tool holder 8 of the tool 11.

[0043] In one embodiment of the present invention, the robot 1 includes a six-axis robot 1 and an end effector. The end effector is located at the movable end of the six-axis robot 1 and is used to pick up a tool 11 and an image acquisition device 12. Specifically, an electric spindle 7 is provided on the end effector, and the quick-change function of the electric spindle 7, in cooperation with a quick-change tool holder 8, quickly picks up the tool 11 and the image acquisition device 12.

[0044] like Figures 4 to 8As shown, the present invention also provides a control method for a deburring workstation for internal hole inspection according to the above embodiments of the present invention. The control method for the deburring workstation for internal hole inspection specifically includes the following steps: S1. Establish the robot coordinate system and mark the initial position of the robot, the position of the workpiece, and the positions of the cutting tool and the image acquisition device in the robot coordinate system; S2. Based on the robot's initial position, the tool's position, and the workpiece's position in the robot coordinate system, control the robot to pick up the tool and move the tool to the workpiece to perform deburring according to the preset process flow. S3. Based on the robot's position, the image acquisition device's position, and the workpiece's position in the robot coordinate system, control the robot to pick up the image acquisition device and move the image acquisition device to the workpiece to acquire the inner hole image. S4. Process the internal hole image and determine the robot's working status based on the internal hole image.

[0045] Specifically, step S1 involves determining the robot's coordinate system, its initial position, the workpiece's position, and the positions of the cutting tool and image acquisition device. If a tool holder is provided, the positions of the cutting tool and image acquisition device in the robot's coordinate system can also be determined based on the tool holder, allowing the robot to accurately position the workpiece, cutting tool, and image acquisition device for picking up the cutting tool and image acquisition device and performing internal hole deburring.

[0046] Specifically, step S2 involves controlling the robot to pick up the tool and move it to the workpiece location based on the positions of the robot, the tool, and the workpiece. The controller has a preset deburring control program for the internal hole, and the robot performs the deburring process according to this program. After completing the deburring process, the robot returns the tool to the tool holder.

[0047] Specifically, step S3 involves controlling the robot to pick up the image acquisition device based on the positions of the robot, the image acquisition device, and the workpiece, and moving it to the workpiece location. The robot then acquires an image of the inner hole using the image acquisition device and transmits it to the controller via a wireless module. After acquiring the inner hole image, the robot returns the image acquisition device to the tool holder.

[0048] Specifically, step S4 involves the controller analyzing the internal hole image to determine if defects exist and whether the deburring process needs to be repeated. If the deburring process needs to be repeated, steps S2 and S3 are repeated; otherwise, the deburring process is completed, and the workpiece is removed. By using a robot to pick up the cutting tool and an image acquisition device to acquire an internal hole image, and using the controller to determine the robot's working status based on the image, automatic robot deburring is achieved. This replaces manual inspection in existing technologies, enabling automatic deburring and burr detection for materials such as stainless steel and aluminum alloys. It reduces workpiece transfer time and improves inspection efficiency and quality. This invention provides a control method for a deburring workstation for internal hole inspection. The robot picks up the cutting tool and an image acquisition device, which acquires an image of the internal hole. The controller determines the robot's working state based on the image, thereby achieving automatic deburring by the robot. This replaces manual inspection in the prior art and enables automatic deburring and burr inspection of materials such as stainless steel and aluminum alloys. It reduces the time for workpiece transfer to the workstation and improves inspection efficiency and quality.

[0049] In one embodiment of the present invention, S4 includes: S41, recognizing the image of the inner hole and determining whether a defect exists; S42, if a defect exists in the inner hole, controlling the robot to pick up the cutting tool and repeat the deburring process; if no defect exists in the inner hole, completing the deburring process and removing the workpiece. Specifically, the controller uses image processing and recognition technology to determine whether a defect exists in the inner hole. Inner hole defects include, but are not limited to, determining whether burrs remain, and also analyzing the image to determine if there are scratches, jamming, or other defects inside the hole. If a defect exists, controlling the robot to pick up the cutting tool and repeat the deburring process; if there are scratches or jamming, repairing and removing the jamming in a timely manner. If no defect exists, completing the deburring process and removing the workpiece.

[0050] In one embodiment of the present invention, step S41 includes the following processes: image acquisition and preprocessing, target detection and instance segmentation, post-processing and optimization, pose estimation and coordinate transformation, and result output and application. Specifically: 1. Image acquisition and preprocessing include: S410, Data Acquisition: Subscribes to color and depth images published by the camera via ROS; S411. Establish the camera coordinate system and preprocess the inner hole image. Image preprocessing: Correct, denoise, and normalize the acquired image, and adjust the image size and format to meet the input requirements of the neural network model.

[0051] 2. Object detection and instance segmentation: S412. Model Inference: Use the target detection model (preferably YOLOv8 model) to infer the preprocessed image and obtain the predicted bounding box and segmentation mask of the defective target.

[0052] S413. Result Decoding: Analyze the results output by the model; including: target category, confidence level, location, and segmentation information.

[0053] 3. Post-processing and optimization: S414. Contour Extraction: Extract the contour information of the defect region from the segmentation mask.

[0054] S415 Feature Optimization: Specific algorithms are used to optimize different types of defects; for example, ellipse fitting is performed on holes, and contour refinement is performed on burrs.

[0055] S416, 3D mapping: Using depth information, the two-dimensional defect contour is mapped to three-dimensional space to obtain its three-dimensional coordinates in the camera coordinate system.

[0056] 4. Pose estimation and coordinate transformation: S417, Pose Calculation: Calculate the spatial position and orientation information of the defect based on the three-dimensional coordinates; Coordinate Transformation: Use the extrinsic parameter matrix to transform the position information of the defect from the camera coordinate system to the robot coordinate system, ensuring that the robot can accurately locate and process the defect.

[0057] 5. Results Output and Application: Visualization: The detection results are displayed in real time on screen 3, including the location, type, and segmentation effect of the defects.

[0058] Data interface: Sends the processed data to the robot control system (i.e., the controller) to guide the robot to perform corresponding operations.

[0059] In one embodiment of the present invention, S415 specifically includes: ellipse fitting for the hole features, including: S41511, Major axis ratio calculation: For the extracted hole profile, calculate its major axis and minor axis lengths, and evaluate the shape characteristics of the profile; S41512, Ellipse Fitting: Using an ellipse fitting algorithm, the contour is accurately fitted to obtain the center position, major axis, minor axis and rotation angle of the hole. S41513. Reasonableness Adjustment: Based on the preset axis ratio range, the parameters of the fitted ellipse are corrected to avoid fitting errors caused by outliers.

[0060] In one embodiment of the present invention, S415 specifically includes: contour refinement for burr defects, including: S41521, Local Curvature Calculation: Calculate the local curvature of each point on the contour, and identify key feature points and abrupt change points of the contour; S41522, Key Point Screening: Screen key points of burr areas based on curvature values, and remove noise and interference points; S41523, Spline Interpolation: Using spline curves to fit the selected key points to generate smooth and accurate contour lines, improving the ability to describe burr boundaries.

[0061] In one embodiment of the present invention, S416 specifically includes: 3D pose estimation, which combines depth information to achieve three-dimensional localization of the defect target, including: camera calibration, two-dimensional to three-dimensional mapping and coordinate system transformation.

[0062] 1. Camera calibration: S4161, Intrinsic parameter calibration: Obtain the camera's intrinsic parameter matrix, including focal length (f_x, f_y) and principal point coordinates (c_x, c_y).

[0063] S4162, Depth Correction: Corrects the depth image based on the camera's depth scaling factor to ensure the accuracy of the depth values.

[0064] 2. Two-dimensional to three-dimensional mapping S4163, Pixel Coordinates to Camera Coordinates: For each contour point (u, v) in the image, calculate its three-dimensional coordinates in the camera coordinate system using the following formula: Z = depth(u, v) × scale X = (u - c_x) × Z / f_x Y = (v - c_y) × Z / f_y S4164, Point Cloud Generation: Map all contour points to three-dimensional space to generate a three-dimensional point cloud of the defect area.

[0065] 3. Coordinate system transformation S4165, External parameter calibration: Obtain the transition matrix T_{camera}^{base} from the camera coordinate system to the robot coordinate system.

[0066] S4166, Coordinate Transformation: Using a transition matrix, transform the point cloud from the camera coordinate system to the robot coordinate system: P_{base} = T_{camera}^{base} × P_{camera} S4167. Pose Information Acquisition: Based on the converted point cloud, calculate the center position and orientation information of the defect area for the robot to locate and operate.

[0067] The present invention also provides a system for controlling a deburring workstation for implementing the internal hole detection of the above embodiments. This system adopts a layered design approach, with its overall architecture divided into three layers: a sensing layer, a processing layer, and an application layer. 1. Perception Layer: Responsible for image data acquisition and preprocessing. It uses a stereo camera to acquire color images and depth information, and leverages ROS's topic mechanism to enable real-time subscription and publishing of image data. The main functions of the perception layer include image correction, denoising, scaling, and normalization, providing high-quality input data for subsequent algorithmic processing.

[0068] 2. Processing Layer: As the core of the system, the processing layer includes a YOLOv8 object detection and instance segmentation model accelerated by TensorRT, as well as a 3D pose estimation algorithm. The processing layer mainly performs the following tasks: A. Target Detection and Instance Segmentation: The input image is processed using an improved YOLOv8 network to obtain defect areas on the surface of the part, including holes and burrs.

[0069] B. Post-processing and optimization: Specific post-processing algorithms, such as ellipse fitting and contour refinement, are used to improve the detection accuracy, taking into account the special characteristics of industrial parts.

[0070] C. 3D Pose Estimation: Combining depth information, the detected defect area is mapped to 3D space, its three-dimensional position in the camera coordinate system is calculated, and then transformed to the robot coordinate system for the robot to perform precise operations.

[0071] 3. Application Layer: Provides a human-computer interaction interface and result output interface, supports the visualization of detection results, and interfaces with robot control systems to achieve automated defect handling and quality control.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A deburring workstation for internal hole inspection, characterized in that, include: Machine (2); A cutting tool (11) is provided on the machine base (2) for removing burrs from the inner hole; An image acquisition device (12) is installed on the machine base (2) and is used to acquire images of the inner hole; A robot (1) is mounted on the machine base (2) and is used to pick up the cutting tool (11) and the image acquisition device (12). The controller is electrically connected to the image acquisition device (12) and the robot (1) respectively, and is used to determine the working state of the robot (1) by acquiring the inner hole image.

2. The deburring workstation for internal hole inspection according to claim 1, characterized in that, Also includes: The tool holder (6) is located on the machine base (2), and the tool (11) and the image acquisition device (12) are both located on the tool holder (6).

3. The deburring workstation for internal hole inspection according to claim 2, characterized in that, The tool holder (6) is provided with at least two through holes; The cutting tool (11) is provided with a first connecting part, and the image acquisition device (12) is provided with a second connecting part. Both the first connecting part and the second connecting part are located on the through hole, and both the first connecting part and the second connecting part are suitable for being picked up by the robot (1).

4. The deburring workstation for internal hole inspection according to claim 1, characterized in that, The image acquisition device (12) includes: The wireless module (9) is connected to the controller signal and is used to transmit the internal hole image; An endoscope (10) is connected to the wireless module (9) for acquiring images of the internal orifice.

5. The deburring workstation for internal hole inspection according to any one of claims 1 to 4, characterized in that, The robot (1) includes: Six-axis robot (1); An end effector, located at the movable end of the six-axis robot (1), is used to pick up the cutting tool (11) and the image acquisition device (12).

6. A control method for a deburring workstation for internal hole inspection according to any one of claims 1 to 5, characterized in that, include: S1. Establish the robot coordinate system and mark the initial position of the robot (1), the position of the workpiece (5), the position of the tool (11), and the position of the image acquisition device (12) in the robot coordinate system; S2. Based on the initial position of the robot (1), the position of the tool (11), and the position of the workpiece (5) in the robot coordinate system, control the robot (1) to pick up the tool (11) and move the tool (11) to the workpiece (5) to perform deburring according to the preset process flow. S3. Based on the position of the robot (1), the image acquisition device (12) and the workpiece (5) in the robot coordinate system, control the robot (1) to pick up the image acquisition device (12) and move the image acquisition device (12) to the workpiece (5) to acquire the inner hole image; S4. Process the inner hole image and determine the working state of robot (1) based on the inner hole image.

7. The control method for the deburring workstation for internal hole inspection according to claim 6, characterized in that, S4 specifically includes: S41. Identify the internal hole image and determine if any defects exist; S42. If there is a defect in the inner hole, control the robot (1) to pick up the tool (11) and repeat the deburring process; If there are no defects in the inner hole, the deburring process is completed and the workpiece is removed (5).

8. The control method for the deburring workstation for internal hole inspection according to claim 6, characterized in that, S41 specifically includes: S411. Establish the camera coordinate system and preprocess the inner hole image; S412. Use the target detection model to reason about the preprocessed image to obtain the predicted bounding box and segmentation mask of the defect target; S413. Analyze the results output by the model; S414. Extract the contour information of the defect region from the segmentation mask; S415. For different types of defects, algorithms are used for optimization. S416. Using depth information, the two-dimensional defect contour is mapped to three-dimensional space to obtain its three-dimensional coordinates in the camera coordinate system. S417. Calculate the spatial location and orientation information of the defect based on the three-dimensional coordinates, and use the extrinsic parameter matrix to transform the position information of the defect from the camera coordinate system to the robot coordinate system in order to locate and process the defect.

9. The control method for the deburring workstation for internal hole inspection according to claim 8, characterized in that, S415 specifically includes: ellipse fitting for hole features, including: S41511. For the extracted hole contour, calculate its major axis and minor axis lengths, and evaluate the shape characteristics of the contour; S41512. Use an ellipse fitting algorithm to fit the contour and obtain the center position, major axis, minor axis and rotation angle of the hole. S41513. Correct the parameters of the fitted ellipse according to the preset axis ratio range.

10. The control method for the deburring workstation for internal hole inspection according to claim 8, characterized in that, Specifically, S415 includes: contour refinement for burr defects, including: S41521. Calculate the local curvature of each point on the contour and identify the key feature points and abrupt change points of the contour. S41522. Based on the curvature value, select the key points of the burr area and remove noise and interference points; S41523. Use spline curves to fit the selected key points and generate contour lines.

Images