Intelligent Drainage Board Insertion Robot and Control Method Based on Machine Vision

By utilizing machine vision technology and a central control system, automated positioning and shearing of the drainage board insertion equipment have been achieved, solving the problems of large deviations and high safety risks associated with manual positioning, and improving construction efficiency and quality.

CN123109154APending Publication Date: 2026-11-17HUNAN UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611090821.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-22
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing drainage board insertion equipment relies on manual positioning, resulting in large positional deviations, high shearing safety risks, insufficient continuous automation, and low construction efficiency, especially in soft soil foundation construction environments.

Method used

An intelligent drainage board insertion robot based on machine vision is adopted, which combines a tracked frame body, vision perception components, position detection components and a central control system to achieve secondary correction of target points, automatic insertion and automatic cutting. Through visual reference object recognition and pixel-level precise positioning, the accuracy of the insertion point is ensured.

Benefits of technology

It improves the accuracy of plate insertion point identification to the millimeter level, reduces the dispersion of construction quality, realizes seamless automated processes, ensures construction efficiency and safety, and is suitable for various harsh working conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123109154A_ABST
    Figure CN123109154A_ABST
Patent Text Reader

Abstract

This invention relates to the field of construction equipment technology, and discloses an intelligent drainage board insertion robot and its control method based on machine vision. The intelligent drainage board insertion robot includes: a tracked frame body, an insertion actuator, a vision perception component, a position detection component, and a central control system. The insertion actuator is used to lift, lower, and cut the carried pile pipe; the vision perception component acquires images of the surrounding area of ​​the tracked frame body and the descent stroke of the pile pipe; the position detection component acquires the absolute position of the tracked frame body; the central control system receives the target point, controls the tracked frame body to move to the target point based on the absolute position and the target point, moves it a second time based on the surrounding images, and controls the insertion actuator to insert the pile pipe. When the descent stroke of the pile pipe reaches a depth threshold, the insertion actuator is controlled to cut the drainage board. This solves the problems of positioning errors and cutting operation risks associated with manual visual estimation, and improves the continuous operation efficiency and construction safety of drainage board insertion.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of construction equipment technology, and in particular to an intelligent drainage board insertion robot based on machine vision and its control method. Background Technology

[0002] The installation of plastic drainage boards is a common construction procedure in soft soil foundation treatment. It typically involves inserting the drainage boards into the foundation at predetermined points and depths to form drainage channels, and using surcharge preloading or vacuum preloading to improve the consolidation efficiency of the foundation. Traditional board insertion equipment often relies on manual visual inspection, measurement and layout, or semi-automatic control to complete the point positioning, pile pipe pressing, and drainage board cutting.

[0003] In construction environments with soft soil foundations, mud and water cover, or where site markers are easily damaged, manual visual positioning is susceptible to the effects of equipment shaking, changes in lighting, surface mud and water, and differences in personnel experience, leading to increased deviations in the placement of drainage boards. On the other hand, cutting drainage boards usually requires personnel to be close to the construction site or to perform auxiliary operations from a higher position, posing construction safety risks, and manual intervention reduces the efficiency of continuous construction.

[0004] Therefore, it is necessary to provide a drainage board insertion robot that can combine a positioning module, a downward camera, on-site visual markers or stable reference objects, and insertion board depth feedback for closed-loop control, so that it can perform secondary correction of target points, automatic insertion, automatic shearing, and continuous resetting operations in soft soil foundation construction environments. Summary of the Invention

[0005] This invention provides an intelligent drainage board insertion robot and control method based on machine vision to solve the problems of existing drainage board insertion equipment, such as reliance on manual positioning, large point deviation, high shearing safety risk, and insufficient continuous automation.

[0006] To achieve the above objectives, the present invention employs the following technical solution:

[0007] In a first aspect, the present invention provides an intelligent drainage board insertion robot based on machine vision, comprising: a tracked frame body, an insertion actuator, a vision perception component, a position detection component, and a central control system, wherein a pile pipe is vertically arranged at the center of the tracked frame body, and a channel for accommodating or guiding drainage boards is provided inside the pile pipe. The tracked frame body is used to carry drainage boards and pile pipes for movement; The insert plate actuator is used to lift, cut, and shear the pile pipe carried by the tracked frame body; The visual perception component is used to acquire images of the area around the tracked framed fuselage and the descent stroke of the pile pipe; The position detection component is used to obtain the absolute position of the tracked frame body; The central control system is used to receive the target point position, control the tracked frame body to move to the target point according to the absolute position and the target point position, confirm the target point by moving it a second time according to the surrounding image, and control the insert plate actuator to insert the pile pipe. When the pile pipe descends to a depth threshold, the insert plate actuator is controlled to cut the drainage plate.

[0008] Furthermore, the step of confirming the target point through secondary movement based on surrounding images includes: The central control system identifies the position of a pre-set visual reference in the surrounding image and calculates the plane deviation compensation based on the pixel deviation between the position of the visual reference and the center position of the pile pipe. The central control system controls the tracked movement mechanism to perform compensating movement according to the plane deviation compensation amount, so that the center of the pile pipe coincides with the target point, and completes the secondary movement to confirm the target point.

[0009] Furthermore, the visual reference is set at the target point, and the visual reference includes reflective marks, color block marks, QR code marks, positioning targets, or other ground markings that can be recognized by the camera.

[0010] Furthermore, the tracked frame body includes a base, guide rails, crossbeams, and a track moving mechanism; The base has a discharge port at its center for the movement of the pile pipe, the guide rail column is vertically arranged at the edge of the discharge port, and the crossbeam is arranged at the top of the guide rail column.

[0011] Furthermore, the insert plate actuator includes a lifting drive component, a transmission assembly, a pile tube, and an electrically controlled shearing device; The lifting drive component is located on the upper part of the crossbeam or guide rail column. The lifting drive component drives the pile pipe to rise and fall along the guide rail column via a transmission assembly. The electrically controlled shearing device is located near the discharge port.

[0012] Furthermore, the transmission assembly includes any one of a pulley transmission assembly, a chain transmission assembly, a rack and pinion transmission assembly, or a lead screw transmission assembly; The pile pipe is slidably engaged with the guide rail column via a slider or guide sleeve.

[0013] Furthermore, the central control system includes an image preprocessing module, a target recognition module, a deviation calculation module, a coordinate conversion module, a motion compensation module, and a plug-in control module. The image preprocessing module is used to preprocess the surrounding images; The target recognition module is used to identify visual reference objects based on the preprocessed surrounding images; The deviation calculation module is used to calculate the pixel deviation between the center position of the pile pipe and the position of the identified visual reference object; The coordinate conversion module is used to convert pixel deviation into planar motion compensation amount of the tracked frame body; The motion compensation module is used to control the movement of the tracked frame fuselage according to the planar motion compensation amount.

[0014] Furthermore, the visual perception component includes a camera and a depth detection component. The camera is used to acquire images around the tracked framed body, and the depth detection component is used to acquire the descent stroke of the pile pipe. The depth detection component includes any one or more of the following: encoder, draw rope displacement sensor, proximity switch, laser rangefinder, ultrasonic rangefinder, pressure sensor, or motor current detection unit. The central control system controls the descent process of the pile tube based on the descent stroke of the pile tube of the depth detection component or the running time of the lifting drive component.

[0015] Furthermore, the central control system also includes a shear control module; When the descent stroke of the pile pipe reaches a preset depth threshold, the shearing control module controls the insert plate actuator to shear the drainage plate.

[0016] Furthermore, the electrically controlled shearing device includes a fixed blade holder, a movable blade, a shearing drive component, and a shearing completion detection element; The shearing drive includes one or more of the following: electric push rod, electromagnet, servo motor, hydraulic cylinder or pneumatic cylinder. The shearing completion detection element is used to feed back a shearing completion signal to the central control system. After receiving the shearing completion signal, the central control system controls the tracked frame body to leave the target point.

[0017] Furthermore, the coordinate conversion module converts the pixel deviation into a planar compensation amount for the tracked moving mechanism based on the conversion ratio between pixel coordinates and ground distance.

[0018] Furthermore, the central control system is also used to perform any one or more of the following operations when the target point cannot be identified, the identification result deviation exceeds a preset threshold, or the difference between the planar motion compensation amount and the displacement amount obtained by the position detection component exceeds a preset allowable threshold: re-acquiring images, reducing the track movement speed, making small-range adjustments to the position, stopping the insertion plate, or issuing an alarm.

[0019] Secondly, the present invention provides a control method for an intelligent drainage board insertion robot based on machine vision, applicable to the intelligent drainage board insertion robot described in any of the above claims, the control method comprising the following steps: S1. Receive the coordinates of the target point and obtain the absolute position of the tracked frame fuselage; S2. Control the tracked movement mechanism to move to the target point based on the target point coordinates and absolute position; S3. Acquire surrounding images using a camera and preprocess the surrounding images; S4. Identify pre-set visual references in the construction area and determine the pixel coordinates of the target point in the image based on the visual references; S5. Calculate the pixel deviation based on the pixel coordinates of the target point and the calibrated pixel coordinates of the center position of the pile pipe, and obtain the plane deviation compensation based on the conversion relationship between the pixel and the ground distance. S6. Control the track movement mechanism to perform secondary alignment according to the plane deviation compensation amount, and re-acquire images for verification after alignment until the target point position deviation is within the preset allowable range. S7. Control the lifting drive to drive the pile pipe downward, insert the drainage board into the foundation, and collect the descent stroke of the pile pipe in real time. S8. When the descent stroke of the pile pipe reaches the preset depth threshold, control the electric shearing device to cut the drainage board. After receiving the shearing completion signal, control the pile pipe to move upward to reset and leave the target point.

[0020] Furthermore, in step S7, the central control system performs segmented control of the pile pipe descent speed based on the depth feedback signal; when pile pipe jamming, abnormal insertion resistance, loss of depth signal, or inconsistency between visual and position verification is detected, the insertion plate action is paused and a reset, alarm, or repositioning operation is performed.

[0021] Furthermore, when a visual reference cannot be identified, or the visual reference is obscured by mud or water, the image brightness is insufficient, the image is blurry, or the identification result deviation exceeds a preset threshold, the central control system performs at least one of the following operations: re-acquire the image, reduce the track movement speed, make a small-range adjustment of the position, pause the insertion plate, or issue an alarm.

[0022] Beneficial effects: The intelligent drainage board insertion robot and control method based on machine vision provided by this invention constructs a two-stage progressive mode of "coarse positioning + fine positioning" by integrating position detection and visual perception technologies. First, the position sensor is used to achieve rapid coarse positioning of the working area, and then machine vision is used to perform pixel-level precise positioning of the target point. This effectively overcomes the shortcomings of single sensors being susceptible to interference from lighting, vibration, or ground undulations, and improves the position recognition accuracy of the insertion point to the millimeter level, providing a reliable benchmark for subsequent insertion actions. By introducing visual reference object recognition, the cumulative positioning drift caused by track slippage, uneven ground or long-term operation is effectively eliminated, ensuring that the insertion position of each drainage board strictly conforms to the design coordinates, thereby greatly reducing the dispersion of construction quality. It integrates automatic insertion, automatic depth control, and automatic cutting functions to achieve seamless connection of the "walking → positioning → insertion → cutting" process. Compared with traditional manual operation, each action parameter can be standardized and set to avoid the differences in speed and quality of manual operation, ensuring stable and controllable work efficiency during large-scale construction. The entire operation requires only a small number of monitoring personnel, and the automated system supports continuous operation for a long time, extending the effective operation time; This invention is not only applicable to conventional soft soil foundations, but also has good adaptability to harsh working conditions such as hydraulic filling of ultra-soft soil, high water content silt layers, and large-area zonal construction. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall structure of the intelligent drainage board insertion robot based on machine vision, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the visual secondary alignment and closed-loop control process according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the conversion relationship between the downward-facing camera field of view, the center of the discharge port, the target insertion plate point, and the pixel deviation in an embodiment of the present invention. Figure 4 This is a schematic diagram of the hardware interface of the central control system according to an embodiment of the present invention.

[0024] 1. Tracked frame body; 2. Insertion plate actuator; 3. Vision sensing component; 11. Base; 12. Guide rail column; 13. Crossbeam; 14. Tracked movement mechanism; 15. Discharge port; 21. Lifting drive component; 22. Transmission component; 23. Piling pipe; 24. Electrically controlled shearing device; 31. Camera; 32. Depth detection component; 41. Fixed blade holder; 42. Movable blade; 43. Shearing drive component. Detailed Implementation

[0025] The technical solution of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms "an" or "a" and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms "connected" or "linked" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up," "down," "left," "right," etc., are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship also changes accordingly.

[0027] Please see Figure 1 This application provides an intelligent drainage board insertion robot based on machine vision, including: a tracked frame body 1, an insertion actuator 2, a vision perception component 3, a position detection component, and a central control system. A pile pipe 23 is vertically arranged at the center of the tracked frame body 1, and a channel for accommodating or guiding drainage boards is provided inside the pile pipe 23. Tracked frame hull 1 is used to carry drainage boards and pile pipes 23 for movement; Specifically, the tracked frame hull 1 includes a base 11, guide rail posts 12, crossbeams 13, and track moving mechanism 14, such as... Figure 1 As shown, the base 11 has a discharge port 15 at its center for the pile pipe 23 to pass through, and the drainage plate can be inserted into the soft soil foundation by passing through the discharge port 15 as the pile pipe 23 moves down. The base 11 is used to support the whole machine and move on the soft soil foundation. The guide rail column 12 is vertically set on the base 11, and the crossbeam 13 is connected to the top of the guide rail column 12.

[0028] Insertion plate actuator 2 is used to lift and cut the pile pipe 23 carried by the tracked frame body 1; The insert plate actuator 2 includes a lifting drive component 21, a transmission assembly 22, a pile tube 23, and an electrically controlled shearing device 24. The lifting drive component 21 is located on the upper part of the crossbeam 13 or the guide rail column 12. The lifting drive component 21 drives the pile tube 23 to rise and fall along the guide rail column 12 via the transmission assembly 22. The electrically controlled shearing device 24 is located near the discharge port 15.

[0029] The electrically controlled shearing device 24 includes a fixed blade holder 41, a movable blade 42, a shearing drive component 43, and a shearing position detection element; The shearing drive 43 includes one or more of an electric actuator, an electromagnet, a servo motor, a hydraulic cylinder, or a pneumatic cylinder, without limitation herein; The shearing completion detection element is used to feed back the shearing completion signal to the central control system. After receiving the shearing completion signal, the central control system controls the tracked frame body 1 to leave the target point.

[0030] In this embodiment, the transmission component 22 is selected as the pulley transmission component 22. In other embodiments, any one of the chain transmission component 22, rack and pinion transmission component 22 or lead screw transmission component 22 may also be selected; the pile tube 23 is slidably engaged with the guide column 12 through the slider or guide sleeve.

[0031] Visual perception component 3 is used to acquire images of the area around the tracked framed fuselage 1 and the descent stroke of the pile pipe 23; The visual perception component 3 includes a camera 31 and a depth detection component 32. The camera 31 is used to acquire images around the tracked frame body 1, and the depth detection component 32 is used to acquire the descent stroke of the pile pipe 23. In this embodiment, the camera 31 includes a downward camera and a forward camera. The downward camera is used to acquire real-time images below the discharge port 15, while the forward camera is used to acquire real-time images in the forward direction.

[0032] The depth detection component 32 can be any one or more of the following: encoder, draw rope displacement sensor, proximity switch, laser rangefinder, ultrasonic rangefinder, pressure sensor or motor current detection unit, without any limitation here; A position detection component is used to obtain the absolute position of the tracked framed fuselage 1; In this embodiment, GPS positioning is used to achieve the positioning function of the position detection component. In other embodiments, Beidou positioning, RTK positioning, total station positioning or inertial navigation can also be used to achieve positioning.

[0033] The central control system receives the target point position, controls the tracked frame body to move to the target point based on the absolute position and the target point position, confirms the target point by moving it a second time based on the surrounding image, and controls the insert plate actuator 2 to insert the pile pipe 23. When the pile pipe 23 descends to the depth threshold, the insert plate actuator is controlled to cut the drainage plate.

[0034] Among them, confirming the target point through secondary movement based on surrounding images includes: The central control system identifies the position of a pre-set visual reference in the surrounding image and calculates the plane deviation compensation based on the pixel deviation between the position of the visual reference and the center position of the pile pipe 23. Visual references are set at the target point. These visual references include reflective marks, color blocks, QR code marks, positioning targets, or other ground markings that can be recognized by the camera 31.

[0035] The central control system controls the tracked moving mechanism 14 to perform compensating movement based on the plane deviation compensation amount, so that the center of the pile pipe 23 coincides with the target point, and completes the secondary movement to confirm the target point.

[0036] like Figure 2 As shown, for the complete control process, the first step is to receive the coordinates of the defined target point, obtain the robot's absolute position through the position detection component, and then control the tracked frame body 1 to move to the vicinity of the target point by using the coordinates of the target point and its own absolute position to complete the first coarse positioning. After moving to the vicinity of the target point, the robot obtains an image of the discharge port 15 through the downward-facing camera. The obtained image is first converted to grayscale, denoised by Gaussian filtering, and distorted. The region of interest (ROI) is then cropped near the discharge port 15 to reduce the influence of irrelevant ground areas on the recognition results.

[0037] The region of interest is identified, and visual references are identified within it. The identification process can employ any of the following methods: color thresholding, edge detection, contour extraction, template matching, or Sobel gradient detection. After identifying the visual references in the image, the pixel deviation is calculated based on the position of the visual references and the center position of the pile pipe 23. Then, based on the conversion relationship between the pixel and the ground distance, the pixel deviation is converted into planar compensation values ​​dX and dY, and a low-speed fine-tuning compensation movement is performed to make the center of the pile pipe 23 coincide with the target point.

[0038] After fine-tuning, the central control system again acquires images via the downward-facing camera and recalculates the deviation between the target insertion point and the center of the discharge port 15 or the center of the pile pipe 23. When the deviation is less than the preset allowable error threshold, the central control system determines that the secondary alignment is complete and proceeds to the downward insertion step of the pile pipe 23; when multiple consecutive identification failures occur or the deviation consistently exceeds the preset threshold, the central control system issues an alarm.

[0039] A downward-facing camera is installed near the discharge port 15, and its field of view covers the discharge port 15, the center of the pile pipe 23, the target insertion point, and surrounding visual markers or terrain references. During equipment debugging, the downward-facing camera is first calibrated to determine the fixed pixel coordinates of the center of the discharge port 15 or the center of the pile pipe 23 in the image, and to determine the ground distance corresponding to each pixel in the image.

[0040] Specifically, please see Figure 3 The pixel coordinates of the center of the discharge port 15 or the center of the pile pipe 23 in the real-time image are u0, v0; the pixel coordinates of the target insertion point or the center of the visual marker in the real-time image are u1, v1. Then the pixel deviation of the target insertion point relative to the center of the discharge port 15 or the center of the pile pipe 23 is:

[0041]

[0042] Where Δu represents the pixel deviation in the horizontal direction of the image, and Δv represents the pixel deviation in the vertical direction. The central control system converts the pixel deviation into a ground plane compensation amount based on the calibrated conversion ratio between the pixel distance and the ground distance.

[0043]

[0044] Where kx and ky are the pixel-to-ground distance conversion coefficients in the horizontal and vertical directions of the image, respectively. The positive and negative directions of dX and dY are determined during calibration based on the installation direction of the downward-facing camera and the motion coordinate direction of the track base 11. The central control system generates motion control commands for the track drive motor based on dX and dY, controlling the track moving mechanism 14 to make micro-displacement adjustments towards the target insertion point. After adjustment, the image is acquired again and Δu and Δv are recalculated until the actual deviations corresponding to dX and dY are less than the preset allowable error threshold.

[0045] After the compensation movement is completed, the control plate actuator 2 lowers the pile pipe 23, and the drainage pipe also lowers along with the pile pipe 23. During the descent, the descent process of the pile pipe 23 is controlled according to the descent stroke of the pile pipe 23 by the depth detection component 32 or the running time of the lifting drive component 21. When the descent stroke of the pile pipe 23 reaches the preset depth threshold, the control plate actuator 2's electric shearing device 24 is controlled to shear the drainage plate. After the shearing is completed, a shearing completion signal is fed back, and then the tracked frame body 1 is controlled to leave the target point or move to the next new target point.

[0046] Specifically, the functional modules of the central control system include an image preprocessing module, a target recognition module, a deviation calculation module, a coordinate conversion module, a motion compensation module, a plug-in control module, and a shearing control module. The image preprocessing module is used to preprocess the surrounding image. The preprocessing here includes grayscale conversion, Gaussian filtering for noise reduction, distortion correction, brightness adjustment, and region of interest (ROI) extraction.

[0047] The target recognition module is used to identify visual references based on the preprocessed surrounding images; The deviation calculation module is used to calculate the pixel deviation between the center position of the pile pipe 23 and the position of the identified visual reference object; The coordinate conversion module is used to convert pixel deviation into planar motion compensation amount of tracked frame body 1. The coordinate conversion module converts pixel deviation into planar compensation amount of tracked moving mechanism 14 according to the conversion ratio between pixel coordinates and ground distance. The motion compensation module is used to control the movement of the tracked frame fuselage 1 based on the planar motion compensation amount; The shearing control module is used to control the insert plate actuator to shear the drainage plate when the descent stroke of the pile pipe 23 reaches a preset depth threshold.

[0048] In this embodiment, when the central control system fails to identify the target point, the identification result deviation exceeds a preset threshold, or the difference between the planar motion compensation amount and the displacement amount obtained by the position detection component exceeds a preset allowable threshold, it performs any one or more of the following operations: re-acquire images, reduce track movement speed, make small-range position adjustments, stop inserting plates, or issue an alarm.

[0049] The central control system can use embedded controllers, PLCs, industrial control boards, or microcontroller control boards.

[0050] like Figure 4 As shown, the central control system receives GPS coordinate information via a UART interface; controls the lifting motor and track drive motor via PWM, CAN, or RS485 interfaces to achieve speed adjustment and direction control; reads feedback signals from depth / position sensors via an ADC / GPIO interface; and controls the shearing device's operation via relays, drivers, or power switches. Depending on the specific engineering requirements, the central control system can also be connected to a wireless communication module, a human-machine interface, a data storage module, and a remote monitoring module.

[0051] Through the above structure and method, the present invention can integrate manually input points, positioning modules, downward cameras, on-site visual markers, and pile pipe 23 depth feedback into the same closed-loop control process, realizing automatic alignment, automatic insertion, automatic shearing, and automatic reset of drainage board insertion operations, which is suitable for continuous construction scenarios of drainage boards in soft soil foundations.

[0052] This application also provides a control method for an intelligent drainage board insertion robot based on machine vision, applied to the aforementioned intelligent drainage board insertion robot. The control method includes the following steps: S1. Receive the coordinates of the target point and obtain the absolute position of the tracked frame fuselage 1; S2. Control the tracked moving mechanism 14 to move to the target point according to the target point coordinates and absolute position; S3. Acquire surrounding images through camera 31 and preprocess the surrounding images; S4. Identify pre-set visual references in the construction area and determine the pixel coordinates of the target point in the image based on the visual references; When a visual reference cannot be identified, or the visual reference is obscured by mud or water, the image brightness is insufficient, the image is blurry, or the identification result deviation exceeds a preset threshold, the central control system performs at least one of the following operations: re-acquire the image, reduce the track movement speed, make a small-range adjustment to the position, pause the insertion plate, or issue an alarm. S5. Calculate the pixel deviation based on the pixel coordinates of the target point and the calibrated pixel coordinates of the center position of the pile pipe 23, and obtain the plane deviation compensation based on the conversion relationship between the pixel and the ground distance. S6. Control the track moving mechanism 14 to perform secondary alignment according to the plane deviation compensation amount, and re-acquire images after alignment for verification until the target point position deviation is within the preset allowable range. S7. Control the lifting drive component 21 to drive the pile pipe 23 downward, insert the drainage board into the foundation, and collect the downward stroke of the pile pipe 23 in real time. The central control system performs segmented control of the descent speed of the pile tube 23 based on the depth feedback signal. When it detects that the pile tube 23 is stuck, the insertion resistance is abnormal, the depth signal is lost, or the visual / position verification is inconsistent, it pauses the insertion plate action and performs a reset, alarm, or repositioning operation. S8. When the descent stroke of the pile pipe 23 reaches the preset depth threshold, the electrically controlled shearing device 24 is controlled to cut the drainage board. After receiving the shearing completion signal, the pile pipe 23 is controlled to move upward to reset and leave the target point.

[0053] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A machine vision-based intelligent drain board inserting robot, characterized in that, include: The tracked frame body (1), the insert plate actuator (2), the vision perception component (3), the position detection component and the central control system, wherein a pile pipe (23) is vertically arranged at the center of the tracked frame body (1), and a channel for accommodating or guiding the drainage plate is provided inside the pile pipe (23); The tracked frame body (1) is used to carry drainage boards and pile pipes (23) for movement; The insert plate actuator (2) is used to lift and cut the pile pipe (23) carried by the tracked frame body (1). The visual perception component (3) is used to acquire images of the surrounding area of ​​the tracked framed fuselage (1) and the descent stroke of the pile pipe (23); The position detection component is used to obtain the absolute position of the tracked frame fuselage (1); The central control system is used to receive the target point position, control the tracked frame body to move to the target point according to the absolute position and the target point position, confirm the target point by moving it a second time according to the surrounding image, and control the insert plate actuator (2) to insert the pile pipe (23). When the pile pipe (23) descends to the depth threshold, the insert plate actuator is controlled to cut the drainage plate.

2. The machine vision-based intelligent SPT inserting robot according to claim 1, wherein, The step of confirming the target point by secondary movement based on surrounding images includes: The central control system identifies the position of a pre-set visual reference in the surrounding image and calculates the plane deviation compensation based on the pixel deviation between the position of the visual reference and the center position of the pile pipe (23). The central control system controls the tracked moving mechanism (14) to perform compensation movement according to the plane deviation compensation amount, so that the center of the pile pipe (23) coincides with the target point, and completes the secondary movement to confirm the target point.

3. The machine vision-based intelligent SPT inserting robot according to claim 2, wherein, The visual reference is set at the target point, and the visual reference includes reflective marks, color block marks, QR code marks, positioning targets or other ground markings that can be recognized by the camera (31).

4. The machine vision-based intelligent SPT inserting robot according to claim 1, wherein, The tracked frame body (1) includes a base (11), guide rail posts (12), crossbeams (13), and track moving mechanism (14). The base (11) has a discharge port (15) at its center for the movement of the pile pipe (23), the guide rail column (12) is vertically arranged at the edge of the discharge port (15), and the crossbeam (13) is arranged at the top of the guide rail column (12).

5. The machine vision-based intelligent SPT inserting robot according to claim 4, wherein, The insert actuator (2) includes a lifting drive (21), a transmission assembly (22), a pile tube (23), and an electrically controlled shearing device (24). The lifting drive component (21) is located on the upper part of the crossbeam (13) or the guide rail column (12). The lifting drive component (21) drives the pile pipe (23) to rise and fall along the guide rail column (12) via the transmission assembly (22). The electrically controlled shearing device (24) is located near the discharge port (15).

6. The machine vision-based intelligent SPT inserting robot according to claim 5, wherein, The transmission assembly (22) includes any one of the following: pulley transmission assembly (22), chain transmission assembly (22), rack and pinion transmission assembly (22), or lead screw transmission assembly (22); The pile pipe (23) slides with the guide rail column (12) through a slider or guide sleeve.

7. The machine vision-based intelligent SPT inserting robot of claim 2, wherein, The central control system includes an image preprocessing module, a target recognition module, a deviation calculation module, a coordinate conversion module, a motion compensation module, and a plug-in control module. The image preprocessing module is used to preprocess the surrounding images; The target recognition module is used to identify visual reference objects based on the preprocessed surrounding images; The deviation calculation module is used to calculate the pixel deviation between the center position of the pile pipe (23) and the position of the identified visual reference object; The coordinate conversion module is used to convert pixel deviation into planar motion compensation amount of the tracked framed fuselage (1); The motion compensation module is used to control the movement of the tracked frame fuselage (1) according to the planar motion compensation amount.

8. The intelligent drainage board insertion robot based on machine vision according to claim 5, characterized in that, The visual perception component (3) includes a camera (31) and a depth detection component (32). The camera (31) is used to acquire images around the tracked framed fuselage (1), and the depth detection component (32) is used to acquire the descent stroke of the pile pipe (23). The depth detection component (32) includes any one or more of the following: encoder, draw rope displacement sensor, proximity switch, laser rangefinder, ultrasonic rangefinder, pressure sensor or motor current detection unit; The central control system controls the descent process of the pile tube (23) based on the descent stroke of the pile tube (23) of the depth detection component (32) or the running time of the lifting drive component (21).

9. The intelligent drainage board insertion robot based on machine vision according to claim 8, characterized in that, The central control system also includes a shear control module; When the descent stroke of the pile pipe (23) reaches the preset depth threshold, the shearing control module controls the insert plate actuator to shear the drainage plate.

10. The intelligent drainage board insertion robot based on machine vision according to claim 5, characterized in that, The electrically controlled shearing device (24) includes a fixed blade holder (41), a movable blade (42), a shearing drive (43), and a shearing position detection element; The shearing drive (43) includes one or more of the following: electric push rod, electromagnet, servo motor, hydraulic cylinder or pneumatic cylinder; The shearing completion detection element is used to feed back a shearing completion signal to the central control system. After receiving the shearing completion signal, the central control system controls the tracked frame body (1) to leave the target point.

11. The intelligent drainage board insertion robot based on machine vision according to claim 7, characterized in that, The coordinate conversion module converts the pixel deviation into the planar compensation amount of the tracked moving mechanism (14) according to the conversion ratio between the pixel coordinates and the ground distance.

12. The intelligent drainage board insertion robot based on machine vision according to claim 6, characterized in that, The central control system is also used to perform any one or more of the following operations when the target point cannot be identified, the identification result deviation exceeds a preset threshold, or the difference between the planar motion compensation amount and the displacement amount obtained by the position detection component exceeds a preset allowable threshold: re-acquiring images, reducing the track movement speed, making small-range adjustments to the position, stopping the insertion plate, or issuing an alarm.

13. A control method for an intelligent drainage board insertion robot based on machine vision, characterized in that, The control method applied to the intelligent drainage board insertion robot according to any one of claims 1 to 12 includes the following steps: S1. Receive the coordinates of the target point and obtain the absolute position of the tracked frame fuselage (1); S2. Control the tracked movement mechanism (14) to move to the target point according to the target point coordinates and absolute position; S3. Acquire surrounding images through camera (31) and preprocess the surrounding images; S4. Identify pre-set visual references in the construction area and determine the pixel coordinates of the target point in the image based on the visual references; S5. Calculate the pixel deviation based on the pixel coordinates of the target point and the calibrated pixel coordinates of the center position of the pile pipe (23), and obtain the plane deviation compensation based on the conversion relationship between the pixel and the ground distance. S6. Control the track moving mechanism (14) to perform secondary alignment according to the plane deviation compensation amount, and re-acquire images for verification after alignment until the target point position deviation is within the preset allowable range. S7. Control the lifting drive (21) to drive the pile pipe (23) downward, insert the drainage board into the foundation, and collect the downward stroke of the pile pipe (23) in real time; S8. When the descent stroke of the pile pipe (23) reaches the preset depth threshold, control the electric shearing device (24) to cut the drainage board. After receiving the shearing completion signal, control the pile pipe (23) to move upward to reset and leave the target point.

14. The control method according to claim 13, characterized in that, In step S7, the central control system performs segmented control of the downward speed of the pile tube (23) based on the depth feedback signal; when the pile tube (23) is detected to be stuck, the insertion resistance is abnormal, the depth signal is lost or the visual / position verification is inconsistent, the insertion plate action is paused and a reset, alarm or repositioning operation is performed.

15. The control method according to claim 13, characterized in that, When a visual reference object cannot be identified, or the visual reference object is obscured by mud or water, the image brightness is insufficient, the image is blurry, or the identification result deviation exceeds a preset threshold, the central control system performs at least one of the following operations: re-acquire the image, reduce the track movement speed, make a small-range adjustment to the position, pause the insertion plate, or issue an alarm.