Multi-modal detection and precise repair robot self-adaptive to multiple working conditions of pipeline operation and use method thereof

Abstract

The invention relates to the technical field of pipeline detection and repair, in particular to a multi-modal detection and precise repair robot self-adaptive to multiple working conditions of pipeline operation and a use method of the multi-modal detection and precise repair robot. The robot comprises an upper vehicle body, a lower vehicle body, a lifting frame, a blade type crawler walking mechanism, a pan-tilt camera, a radar, a sonar and a modular cleaning / spraying assembly. Trapezoid blades are arranged on the surface of the blade type crawler belt; the blade type crawler belt is mechanically meshed with a pipe bottom to advance under a dry pipe or low-water-level working condition; a buoyancy module is additionally arranged under a high-water-level working condition, and the blades paddling to realize floating propulsion; and under the full-pipe working condition, bottom-submerged crawling is achieved, and full-working-condition self-adaptive walking is achieved. The multi-sensor fusion detection system automatically switches sensor combinations according to working conditions, and full-section defect detection and three-dimensional accurate positioning of a main pipe, a half pipe and a full pipe are achieved. The modular operation system rapidly switches the cleaning and spraying repair assembly through the quick-change interface, and the whole process of detection, cleaning and repair is completed. The problems that an existing robot is single in function and poor in working condition adaptability are solved.

Description

Technical Field

[0001] This invention relates to the field of municipal engineering and pipeline inspection and repair technology, specifically to an adaptive multimodal inspection and precision repair robot for pipeline operation under multiple working conditions and its application method. Background Technology

[0002] Currently, CCTV (closed-circuit television) pipeline inspection robots are the mainstream technology used in municipal drainage pipeline inspection and maintenance. These robots typically employ wheeled or conventional tracked locomotives and are equipped with pan-tilt cameras, enabling them to enter the pipeline interior for video image inspection. Their advantages include relatively simple and convenient operation, a certain degree of obstacle-crossing capability and endurance, and suitability for inspection scenarios with favorable working conditions, such as pipeline acceptance and post-pretreatment.

[0003] However, existing technologies have the following drawbacks: First, existing robots have poor adaptability to different working conditions. For dry pipe conditions, mainstream CCTV inspection robots use wheeled or conventional tracked walking mechanisms, but they are prone to slipping when there is a small amount of silt or slippery material at the bottom of the pipe, resulting in insufficient traction and limited cable dragging distance. For semi-pipe conditions, conventional CCTV robots, due to the submersion of their walking mechanisms, only achieve IP67 temporary waterproofing for their drive motors and transmission systems, making them unable to operate in semi-aquatic environments for extended periods. Furthermore, existing semi-pipe robots can only perform video inspections and cannot clean the pipe walls above the water surface, let alone effectively detect the underwater portion. The pipe walls at the water surface junction are prone to corrosion and leakage due to long-term alternating wet and dry conditions, and existing equipment cannot perform comprehensive diagnosis and precise repair of this area. For full-pipe operations, current technologies mainly rely on sonar equipment. Unpowered sonar needs to be hoisted and placed from the wellhead, cannot move autonomously, and can only float on the water surface to detect the underwater pipe wall, completely unable to detect the pipe wall above the water surface. When powered sonar equipment is propelled in a full pipe, the water flow resistance is high, and the resistance that the robot needs to overcome increases exponentially with the length of the towed cable. The insufficient power of conventional motors limits the working distance, and the presence of turbulent currents in a full pipe makes it difficult for the robot to maintain a stable posture, resulting in inaccurate sonar data. At the same time, the pipe wall above the water surface in a full pipe is completely inaccessible, making it impossible to detect and repair corrosion and cracks in the upper part of the pipe, which is precisely the weak point in the pipe structure.

[0004] Secondly, existing robots have overly limited functionality, only capable of visual inspection inside pipelines, and cannot perform pre-cleaning or immediate repair of detected defects. The separation of inspection, defect location, surface cleaning, and repair processes necessitates multiple pipeline entries, resulting in long operation cycles and high costs. In existing technologies, after inspection, the robot must exit the pipeline, requiring separate dredging equipment for pre-treatment before re-entering for repair work; this repeated entry and exit contributes to long operation cycles and high costs.

[0005] Furthermore, conventional CCTV pipeline inspection robots are only equipped with visual components such as gimbal cameras, which are insufficient for inspecting pipelines in half- or full-water conditions, or can only inspect pipelines above the water surface. Lacking a multi-sensor fusion mechanism, they cannot generate a comprehensive and accurate 3D pipeline data model, making it difficult to accurately locate defects in 3D.

[0006] Therefore, there is an urgent need for a pipeline maintenance robot that can adapt to various complex working conditions such as dry pipes, half-pipes, and full pipes, integrate detection, cleaning, and repair functions, and achieve accurate diagnosis of the entire cross-section through multi-sensor fusion. Summary of the Invention

[0007] To address the above problems, this invention proposes a multifunctional pipeline inspection robot and its usage method. It employs a blade-type tracked walking mechanism, integrating a multi-sensor fusion detection system and a modular cleaning / spraying operation system. The robot adaptively switches its walking mode and sensor combinations based on operating conditions such as dry pipe, low water level, high water level, and full pipe, achieving integrated inspection, cleaning, and repair operations.

[0008] To achieve the above objectives, the present invention specifically adopts the following solution: A method for using a multi-functional pipeline inspection robot includes the following steps: The robot enters the pipeline and uses a sensor detection system to identify the initial operating conditions inside the pipeline, including dry pipe condition, low water level condition, high water level condition, or full pipe condition. Based on the identified working conditions, select the walking mode of the blade track walking mechanism and configure the sensor combination of the multi-sensor detection system. The robot moves according to the selected walking pattern, uses the configured combination of sensors to detect the pipeline, obtain pipeline defect information and accurately locate it; Based on the test results, cleaning or spraying repair work will be carried out on the locations that require cleaning or repair. After completing the task, the work area is re-inspected and a work report is generated.

[0009] Furthermore, under the aforementioned main pipeline operating conditions or low water level operating conditions: The walking mode is the pipe bottom walking mode. The blade-type track walking mechanism is not equipped with a buoyancy module and relies on the trapezoidal blades on the track surface to mechanically engage with the pipe bottom or silt layer to move forward. A gimbal camera and LiDAR are used to acquire image data and 3D point cloud data of the pipe wall surface; The cleaning or spraying repair operation involves using a laser positioning light to aim at the defect location, and then using a gimbal servo motor to drive a high-pressure water gun or spray head for targeted operation.

[0010] Furthermore, the buoyancy provided by the robot's undercarriage allows the robot to partially float, reducing ground pressure; the drive motor speed of the blade-type track walking mechanism is automatically increased to 200-400 rpm, and the track blades simultaneously serve as propellers and spurs, maintaining the robot's propulsion and anti-slip capability.

[0011] Furthermore, under the aforementioned high water level conditions: The walking mode is a floating walking mode. The blade-type track walking mechanism is equipped with a buoyancy module, which enables the robot to float on the water surface. The gimbal camera, lidar and sonar are used to acquire images of the pipe wall above the water surface, point cloud data of the pipe wall above the water surface and data of the underwater pipe wall and sedimentation, respectively. The trapezoidal blades of the blade-type track walking mechanism rotate in the water, acting as a paddle wheel to provide propulsion, and the speed of the drive motor is increased to 400-600 rpm. The cleaning or spraying repair work is carried out on the pipe wall area above the water surface.

[0012] Furthermore, under the aforementioned high water level conditions: When performing cleaning or spraying repair operations in the floating walking mode, the robot's posture is kept stable in the following ways: The gyroscope module is used to monitor the changes in the robot's roll angle, pitch angle, and yaw angle in real time. When a change in attitude is detected due to the backlash from the high-pressure water jet, the forward and reverse rotation of the blade-type track walking mechanism is adjusted to adjust the robot's attitude.

[0013] Furthermore, in the floating walking mode, the robot's lifting frame assembly is lowered to its lowest position, causing the overall center of gravity to drop; when the gyroscope module detects that the roll angle exceeds 5°, the control system automatically adjusts the rotation speed of the two side tracks to correct the attitude.

[0014] Furthermore, under the conditions of full pipe operation or ultra-high water level operation: The walking mode is a bottom crawling mode. The blade-type track walking mechanism is not equipped with a buoyancy module, and the robot moves along the bottom of the pipe. Sonar is used to obtain data on underwater pipe wall structural defects and sediment distribution. The robot only performs detection and data collection tasks and does not perform cleaning or spraying repair work.

[0015] Furthermore, under the conditions of full pipe operation or ultra-high water level operation, when severe siltation is detected, the sonar detector is moved from its installation position at the front of the vehicle body to the installation position of the gimbal camera in order to raise the detection height of the sonar and reduce the interference of bottom silt on the detection.

[0016] Furthermore, it also includes a quick switching process: After detecting a defect that needs to be repaired, the robot retreats to the work preparation position; Operators can replace the cleaning component with the spray repair component via a quick-change interface; The replaced spray repair component is driven by a gimbal servo motor and, combined with the three-dimensional positioning information from a multi-sensor detection system, guides the spray head to the defect area for spray repair.

[0017] Furthermore, it also includes steps involving multiple consecutive operations: When there are multiple defects in the pipeline and the distance between the defects is greater than 50 meters, the robot carries the current working components and moves to each defect point in sequence to perform cleaning or repair operations. When the defect spacing is less than 20 meters, the robot first completes the full-line inspection and cleaning, then returns to the starting point to replace the repair components, and then repairs each defect point in sequence.

[0018] Furthermore, it also includes steps for emergency switching: When a component malfunctions during operation, the control system will automatically sound an alarm and record the current location. The robot returns to the wellhead with the faulty component, where ground personnel replace it with a spare component. After the replacement is completed, the robot automatically navigates back to the original work point based on the recorded coordinate information and continues to perform the unfinished work.

[0019] Furthermore, the cleaning operation adopts pulse spraying or segmented spraying mode. The laser positioning light continuously locks onto the target point during the operation. When the target point deviation is detected to exceed the set threshold, the system automatically pauses spraying and starts the attitude correction program. The operation continues after realignment.

[0020] Furthermore, under dry pipe operation or low water level operation conditions, the cleaning operation further includes: The robot moves to the front of the defect location and stops. The laser positioning light forms an indicator spot on the pipe wall, and the operator confirms the target location through a handheld terminal. The high-pressure water gun and the laser positioning light are arranged in parallel with a small spacing, and the light spot of the laser positioning light is the target point for cleaning and repair. Start the high-pressure water pump and perform fan-shaped sweeping or pinpoint impact cleaning on the defective area; During the cleaning process, the gimbal camera captures the cleaning effect in real time, and the operator can adjust the water gun posture or repeat the cleaning at any time.

[0021] After cleaning, a re-inspection step is also included: The high-pressure water pump is turned off, and the gimbal camera is used to magnify and observe the cleaning area to confirm the type, size, and severity of defects. If further evaluation is required, the lidar is used for a detailed local scan.

[0022] The spraying repair operation further includes: After confirming the defects that need to be repaired, the robot retreats to the work preparation position; Operators switch from cleaning components to spraying repair components by replacing the high-pressure water gun nozzle with a quantitative spray nozzle and the high-pressure water supply pipe with a repair material delivery pipeline. The control system switches to the spraying control interface, where the operator selects the type of repair material and sets the spraying parameters, including spraying width, spraying amount, and number of sprays. The robot moves back to the defect location and uses a multi-sensor fusion positioning system to guide the spray head to align with the cleaned defect area. Once the spraying process is initiated, the spray head moves along a preset path under the drive of the gimbal servo assembly to evenly cover the defects. After the coating is applied and the material has set, the gimbal camera captures the repair results and generates a repair record.

[0023] Correspondingly, the present invention also provides a multifunctional pipeline inspection and repair robot, comprising: The robot body system includes an upper body assembly, a lower body assembly, and a lifting frame assembly connecting the two. A blade-type tracked walking mechanism is installed on both sides of the lower body assembly; A multi-sensor detection system, including a gimbal camera, radar, sonar, and gyroscope module; The modular operating system includes a detachably connected pipe cleaning assembly, which is connected to an external water pump via a high-pressure water supply pipe; The control system is connected to an external power source via a power cable.

[0024] Furthermore, the lower body assembly includes a lower body shell, a lower body sealing cover, a power output shaft sealing cover, and an air nozzle; the lower body assembly is a sealed chamber that can generate a certain buoyancy. The air nozzle is used to inject gas into the lower vehicle body components to maintain positive pressure; The lower body assembly is equipped with a drive motor system, which drives the blade-type track walking mechanism through bevel gears and a power transmission shaft. The drive motor system, bevel gears, and power transmission shaft are arranged at the rear of the robot.

[0025] Furthermore, the blade-type track walking mechanism includes a track frame and an integrated rubber blade track. The surface of the integrated rubber blade track is provided with trapezoidal blades with an inclined angle, which are used to mechanically engage with the bottom of the pipe or the silt layer for movement in dry pipe or low water level conditions, and to generate water propulsion force by rotating and pushing water in high water level conditions.

[0026] Furthermore, the blade-type tracked walking mechanism is equipped with a detachable buoyancy module; under the high water level condition, the blade-type tracked walking mechanism is equipped with a detachable buoyancy module to enable the robot to float on the water surface. Preferably, the buoyancy module is made of rigid polyurethane foam with a foaming density of 0.12 g / cm³, and after processing and molding, the surface is coated with polyurethane paint to prevent water from adhering to the pores on the surface of the buoyancy module.

[0027] Furthermore, the radar is a dual lidar configuration, with the two radars symmetrically arranged on both sides of the front of the robot; the gyroscope module is integrated inside the lower body assembly.

[0028] Furthermore, the gimbal camera and the sonar have the same physical interface, and in the case of severe siltation, the two can be interchanged in their installation positions to increase the installation height of the sonar.

[0029] Furthermore, the pipeline cleaning assembly includes a high-pressure water gun, a laser light, and a gimbal servo assembly; the gimbal servo assembly drives the high-pressure water gun to achieve pitch and horizontal rotation; the laser light is fixedly connected to the high-pressure water gun and is used to project an indicator light spot on the pipe wall. The pipeline cleaning assembly is connected to the high-pressure water supply pipe via a quick-connect interface.

[0030] Furthermore, the modular operation system also includes a spray repair component, which shares the same quick-change interface with the pipeline cleaning component and can be installed interchangeably.

[0031] Furthermore, the spraying repair component is a replacement part of the pipeline cleaning component, wherein the high-pressure water gun is replaced with a metering spray head, the high-pressure water supply pipe is replaced with a high-pressure material supply pipe, and the rear water supply tank is replaced with a repair material tank.

[0032] Furthermore, the control system is integrated into the sealed compartment of the lower body assembly and is used for drive control, communication, and power management.

[0033] Compared with the prior art, the present invention has the following beneficial technical effects: This invention is based on a blade-type tracked walking mechanism. In dry pipe or low water level conditions, it can mechanically engage with the bottom of the pipe or silt layer, providing strong grip and traction. It can stably tow power cables and high-pressure water supply pipes for long distances, overcoming the problems of traditional wheeled or conventional tracked robots slipping easily and having insufficient traction in silt-filled or wet environments. In high water level conditions, after adding a buoyancy module, the trapezoidal blades can generate water propulsion by rotating in the water, enabling the robot to switch from walking on the bottom of the pipe to floating. This achieves full-condition walking capability, allowing it to walk on dry pipes, float on half-pipes, and submerge on full-pipes. It can directly enter the operation without pre-pumping or dredging the pipe.

[0034] By incorporating a multi-sensor fusion detection system consisting of a gimbal camera, lidar, sonar, and gyroscope modules, and setting up interchangeable interfaces between the gimbal camera and sonar, the system can fuse visual and radar data for defect detection and 3D modeling in dry pipe conditions. In half-pipe conditions, it can simultaneously acquire information about the upper and lower pipe walls above the water surface and perform data fitting. In full-pipe or high-siltation conditions, the sonar can be moved to the gimbal camera position to increase the detection height. This effectively overcomes the limitation of single visual sensors failing in water or turbid water conditions, achieving full-section, high-precision pipeline defect detection and 3D positioning.

[0035] The integrated modular operation system allows the cleaning and spray repair components to share the same installation location and high-pressure pipeline via quick-change interfaces. After a defect is detected, the high-pressure water gun of the cleaning component can be used to clean the defective area under the guidance of a laser positioning light to remove surface deposits and expose the defect itself. After cleaning, the system can be quickly replaced on-site with the spray repair component. Combined with a multi-sensor fusion positioning system, the spray head is guided to accurately spray repair the cleaned defect. This integrates the detection, cleaning, and repair processes into one, allowing the entire process to be completed in a single pipeline entry, significantly improving operational efficiency.

[0036] The system uses a gyroscope module to monitor the robot's posture in real time. When it detects a deviation caused by the backlash of the high-pressure water jet, it automatically adjusts the speed difference between the two tracks to generate a reverse compensation torque. The posture adjustment response time is controlled within an extremely short time. At the same time, the lifting frame assembly is lowered to the lowest position to lower the center of gravity of the whole machine, ensuring the accuracy and stability of cleaning and spraying operations in the floating state. This solves the problem of posture instability and poor operation accuracy caused by backlash force in existing floating robots during operation.

[0037] The lower body assembly adopts an IP68 thin-walled sealed cabin structure and is equipped with an air nozzle to inject gas to maintain positive pressure. While providing good waterproof performance, it can also generate a certain amount of buoyancy. The rear-mounted layout of the drive motor system and transmission mechanism makes the center of gravity of the whole machine stable and enhances the ability to tow cables and water pipes. Attached Figure Description

[0038] Figure 1 This is a perspective view of a specific embodiment of the present invention; Figure 2 This is another perspective view of a specific embodiment of the present invention; Figure 3 This is a side view of a specific embodiment of the present invention; Figure 4 This is a rear view of a specific embodiment of the present invention; Figure 5 This is a schematic diagram of the walking mechanism according to a specific embodiment of the present invention; Figure 6 This is a schematic diagram of the buoyancy module according to a specific embodiment of the present invention; Figure 7 This is a structural diagram of a buoyancy module according to a specific embodiment of the present invention; wherein 7(a) is a cross-sectional view of the buoyancy module; and 7(b) is an exploded view of the buoyancy module. Figure 8 This is a schematic diagram of the lower body assembly according to a specific embodiment of the present invention; Figure 9 This is a cross-sectional view of the lower body assembly according to a specific embodiment of the present invention; Figure 10 This is a diagram of the upper body assembly and cleaning / repair assembly according to a specific embodiment of the present invention; Figure 10 (a) is the cleaning component; Figure 10 (b) is a repair component; Figure 11 This is a diagram illustrating a scenario for detecting low water levels in a main pipeline. Figure 12 This is a sonar radar fitting interface diagram for half-pipe / high water level detection operations; Figure 13 This is a sonar interface diagram for full-pipe / ultra-high water level detection operations; Figure 14 This is a schematic diagram of sonar data modeling for full-pipe / ultra-high water level. Figure 15 This is a scene diagram of a main pipe / low water level cleaning operation; Figure 16 This is a scene diagram of a main pipeline / low water level repair operation; Figure 17 This is a scene diagram of a half-pipe / high-water-level cleaning operation; Figure 18 This is a scene of a half-pipe / high-water-level spraying operation.

[0039] Numbers in the diagram: 1. Upper body assembly; 2. Lifting frame assembly; 3. Lower body assembly; 3-1. Lower body sealing cover; 3-2. Power output shaft sealing cover; 3-3. Air nozzle; 4. Blade-type track walking mechanism; 5. Gimbal camera; 6. Radar; 7. Sonar; 8. Pipeline cleaning assembly; 8-1. High-pressure water gun; 8-2. Laser light; 8-3. Gimbal servo assembly; 8-4. Metering nozzle (repaired); 9-1. High-pressure water supply pipe; 9-2. High-pressure material supply pipe (repaired); 10. Power supply cable; 11. Lower body shell; 12. Power transmission shaft; 13. Bevel gear; 14. Drive motor system; 15. Gyroscope module; 16. Control system; 17. Buoyancy module. Detailed Implementation

[0040] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0041] Please see Figures 1 to 4This embodiment provides a multifunctional pipeline inspection and repair robot, which mainly includes an upper body assembly 1, a lower body assembly 3, a lifting frame assembly 2, a blade-type tracked walking mechanism 4, a gimbal camera 5, a radar 6, a sonar 7, and a pipeline cleaning assembly 8. The upper body assembly 1 and the lower body assembly 3 are connected by the lifting frame assembly 2.

[0042] Please see Figure 5 and Figure 9 In this embodiment, the lower body assembly 3 is an IP68 thin-walled sealed chamber structure, including a lower body shell 11, a lower body sealing cover 3-1, a power output shaft sealing cover 3-2, and an air nozzle 3-3. A standard mechanical seal design is used between the lower body shell 11 and the lower body sealing cover 3-1, and between the lower body shell 11 and the power output shaft sealing cover 3-2. The O-rings and skeleton oil seals used in the planar seal and rotary seal are all made of fluororubber, which has good high-temperature resistance and corrosion resistance.

[0043] Air nozzle 3-3 is used to inject dry air or helium into the lower body assembly 3 to provide a positive pressure of 0.1 to 0.12 MPa, effectively ensuring the waterproof performance of the body. At the same time, the lower body assembly 3 is a sealed cabin structure that can generate a certain amount of buoyancy, providing the necessary buoyancy under high water level and high siltation conditions to ensure the stability of the body.

[0044] The lower body assembly 3 houses a drive motor system 14. The output of this drive motor system 14 is reversed via a bevel gear 13 and transmitted to the blade-type track walking mechanism 4 via a power transmission shaft 12, thereby driving the robot forward, backward, or turning. To ensure the robot's stability when equipped with multiple sensors and operating components, especially to ensure a stable center of gravity when dragging cables and water pipes, the drive motor system 14 and its transmission mechanism are specifically positioned at the rear of the robot body. This layout, combined with a suitable body weight and a high-power motor, optimizes the overall compactness and power performance of the robot while meeting the space constraints of narrow pipes.

[0045] Please see Figures 7 to 9 The lifting frame assembly 2 is installed between the upper body assembly 1 and the lower body assembly 3. It adopts a scissor fork structure, is driven by an electric push rod, and has a lifting height of 280mm to 500mm. It is equipped with a self-locking device to accommodate municipal pipelines with diameters ranging from 500mm to 2000mm. The gimbal camera 5 and radar 6 are mounted on top of the lifting frame assembly 2.

[0046] The blade-type tracked walking mechanism 4 is installed on both sides of the lower body assembly 3. Its track surface has trapezoidal blades, which are specifically designed for movement in complex environments with high siltation, sand, or obstacles in pipelines. Specifically, in dry pipe or low water level conditions, the mechanism does not have the buoyancy module 17 installed. It relies on the trapezoidal blades to form a mechanical engagement with the bottom of the pipe or the silt layer, providing strong grip and traction. This ensures stable movement of the robot and also provides sufficient traction for towing the power cable 10 and high-pressure water supply pipe 9 connected to the rear of the robot. This allows the robot to directly enter and perform long-distance operations in harsh pipeline conditions without prior pumping or dredging. In high water level conditions, the mechanism can be equipped with the buoyancy module 17, allowing the robot to float on the water surface. In this case, the trapezoidal blades rotate in the water, acting as paddle wheels to provide propulsion and drive the robot forward or backward, realizing the switching between walking on the bottom of the pipe and floating on the water surface.

[0047] Please see Figure 10 Radar 6 is preferably configured with dual lidar, with the two radars symmetrically arranged at a 45° angle on both sides of the front of the robot. By scanning the pipe wall from different angles, it can reduce the blind spots that occur when scanning with a single radar, and obtain more accurate detection data such as pipe deformation, scaling, and sediment volume through data fitting. Sonar 7 is located at the front of the lower body assembly 3. The gimbal camera 5, radar 6, sonar 7, and gyroscope module 15 together constitute the robot's multi-sensor fusion detection system. In particular, the gimbal camera 5 and sonar 7 have the same physical interface. In the case of severe sediment buildup, sonar 7 can be disassembled and replaced with the interface where the gimbal camera 5 was originally installed, thereby raising the installation height of the sonar and effectively reducing the interference of bottom sediment on sound wave transmission and reception, improving the clarity and accuracy of underwater detection. Gyroscope module 15 is integrated inside the lower body assembly 3 to record the robot's posture and fuse it with radar and sonar data to achieve precise positioning inside the pipe without satellite signals.

[0048] A pipe cleaning assembly 8 is installed at the upper front of the robot. This assembly includes a high-pressure water gun 8-1, a laser light 8-2, and a gimbal servo assembly 8-3. The high-pressure water gun 8-1 is connected to a water pump truck on the ground via a high-pressure water supply pipe 9. The water pump truck provides a water supply pressure of approximately 10 MPa, and the high-pressure water gun 8-1 outputs a terminal cleaning pressure of 1 to 2 MPa. The gimbal servo assembly 8-3 can drive the high-pressure water gun 8-1 to achieve a pitch angle of -10° to +45° and a horizontal rotation angle of -45° to +45°. When the multi-sensor system detects a suspected defect in the pipe wall, the operator can control the laser light 8-2 to emit a laser beam, forming an indicator spot on the pipe wall to accurately locate the defect. Subsequently, the gimbal servo assembly 8-3 is controlled to adjust the spray angle of the high-pressure water gun 8-1, aligning it with the laser indicator point, and firing a high-pressure water stream to wash away dirt and deposits at the defective area. The pre-cleaning process removes surface coverings, fully exposing pipe defects and facilitating a more accurate subsequent assessment of the true level, type, and extent of the defects.

[0049] The pipeline cleaning component 8 adopts a modular design. When defects that need to be repaired are detected and cleaned, the pipeline cleaning component 8 is replaced with a spray repair component. The high-pressure water gun 8-1 is replaced with a quantitative spray head 8-4, the high-pressure water supply pipe 9-1 is replaced with a high-pressure material supply pipe (9-2), and the rear water supply tank is replaced with a repair material tank.

[0050] Installed in the same location, the spray repair assembly includes a metering spray head 8-4, a high-pressure supply pipe 9-2, a material tank on the ground, and a micro air pump. The material tank has a capacity of 2 to 5L and can be quickly replaced according to the type of repair material. The metering spray head is equipped with a fan-shaped or circular adjustable nozzle, with a spray width adjustable from 30 to 150mm. The spray volume is controlled by the micro air pump pressure of 0.2 to 0.6MPa and the opening of the solenoid valve. The spray repair assembly can also be precisely positioned by the existing gimbal servo assembly 8-3. Combining the three-dimensional spatial positioning information provided by the gimbal camera 5, radar 6, sonar 7, and inertial navigation sensor fusion, the robot can control the nozzle of the spray repair assembly to accurately move above the corrosion point or micro-damage of the pipeline for targeted and metered spray repair, achieving in-situ and precise repair of pipeline defects, forming an integrated multi-functional operation of detection, cleaning, and repair.

[0051] The working process of the present invention will be described in detail below in conjunction with different working conditions.

[0052] Please see Figure 11In dry pipe or low water level conditions, the blade-type tracked walking mechanism 4, without the buoyancy module 17, moves by mechanically engaging the trapezoidal blades on the track surface with the bottom of the pipe or the silt layer. The trapezoidal blades have an inclination angle of 10°, a height of 25mm, a blade spacing of 50mm, and a track width of 100mm. The track material is high-wear-resistant EPDM rubber or polyurethane composite material, with a Shore hardness controlled between 65 and 75A, and is internally reinforced with high-modulus aramid fiber cords.

[0053] In dry pipe travel mode, the output speed of the drive motor system 14 is controlled between 100 and 300 rpm, corresponding to a travel speed of 0.1 to 0.5 m / s. When silt or obstacles are detected, the control system automatically increases the motor torque, with a maximum torque of not less than 50 Nm, to enhance traction and ensure stable forward movement of the towed power supply cable 10 and high-pressure water supply pipe 9.

[0054] In low-water-level walking mode, the buoyancy structure of the lower body component 3 causes the robot to partially float, reducing ground pressure. The buoyancy generated by the entire robot in the water can reach 50% to 60% of its own weight. At this time, the drive motor speed is automatically increased to 200 to 400 rpm, and the track blades act as both propellers and spurs. On the one hand, they generate a backward thrust on the water through rotation to assist propulsion; on the other hand, they maintain engagement with the sediment layer at the bottom of the pipe to prevent sideslip and maintain the robot's propulsion and anti-slip capability. Mode switching is achieved by real-time monitoring of the water level using water level and pressure sensors at the bottom of the vehicle. When the water level exceeds a preset threshold of 100 mm and remains above it for more than 3 seconds, the system automatically switches from dry pipe mode to low-water-level walking mode; conversely, when the water level drops below the threshold, the system switches back to dry pipe mode.

[0055] Please see Figure 11 , Figure 15 and Figure 16 When inspecting pipes under conditions of dry pipe or low water level, the gimbal camera 5 continuously captures video of the pipe wall to identify surface defects such as cracks and breaks; the dual lidar 6 scans the pipe wall symmetrically at 45° to acquire point cloud data of the pipe cross-section, which is used to calculate the deformation rate, scale thickness, and 3D modeling. By integrating the images from the gimbal camera, the lidar point cloud, and the pose data from the gyroscope module 15, accurate 3D positioning of defects is achieved, with a positioning error of no more than 50mm.

[0056] After detecting a defect, the robot moves to a position 0.5 to 1 meter in front of the defect and stops. The laser positioning light 8-2 forms an indicator spot on the pipe wall, and the operator confirms the target location via a handheld terminal. The gimbal servo assembly 8-3 automatically adjusts the pitch and rotation angles of the high-pressure water gun 8-1 based on laser echo ranging, aligning it with the target. The high-pressure water pump is activated at a working pressure of 1 to 2 MPa, performing fan-shaped sweeping or pinpoint impact cleaning on the defect area. The cleaning time is adjustable from 5 to 30 seconds until surface deposits are removed and the defect is exposed. During the cleaning process, the gimbal camera 5 captures the cleaning effect in real time, allowing the operator to adjust the water gun's attitude or repeat the cleaning process at any time.

[0057] After cleaning is completed, the high-pressure water pump is turned off, and the gimbal camera 5 magnifies the cleaned area to confirm the type, size and severity of defects; if further evaluation is needed, the radar 6 can be called to perform a detailed scan of the local area.

[0058] After confirming the defects requiring repair, the robot retreats to the work preparation position. The operator switches from the cleaning component to the spraying repair component by replacing the high-pressure water gun 8-1 nozzle with a quantitative spraying head and the high-pressure water supply pipe 9 with the repair material delivery pipeline. The manual operation time does not exceed 10 seconds. The control system switches to the spraying control interface, where the operator selects the repair material type and sets the spraying parameters, including spray width, spray volume, and number of sprays.

[0059] The robot retraces its steps to the defect location and uses a multi-sensor fusion positioning system to guide the spray head towards the cleaned defect area. Spraying is initiated, and the spray head moves along a preset path under the drive of the gimbal servo motor 8-3, evenly covering the defect. The spraying pressure is 0.2 to 0.6 MPa, and the spraying width is adjustable from 30 to 150 mm. After spraying, the robot waits 3 to 10 minutes for the material to initially set. The gimbal camera 5 then captures the repair results, generating a repair record.

[0060] Please see Figure 12 , Figure 17 and Figure 18 In half-pipe or high-water-level conditions, when the water level completely submerges the lower vehicle body and the water depth is greater than the vehicle body height, the system assists in identifying it as a high-water-level condition. On-site operators added a buoyancy module 17 to the blade-type tracked walking mechanism 4. The buoyancy provided by the tracked walking mechanism and the lower vehicle body 3 meets the robot's floating driving requirements.

[0061] At this point, the robot is in floating mode, suspended in the water. The blade-type tracks detach from the bottom of the tube and rotate entirely in the water. The trapezoidal blades rotate in the water, acting as paddle wheels to provide propulsion. The drive motor speed is increased to 400-600 rpm to ensure sufficient propulsion. Steering is achieved by adjusting the speed difference between the two tracks. To prevent the robot from tipping over or becoming unstable while floating, the lifting frame assembly 2 automatically lowers to its lowest position, with a height not exceeding 280 mm, to lower the overall center of gravity. Simultaneously, the control algorithm monitors the roll angle data from the gyroscope module 15 in real time. When the roll angle exceeds 5°, it automatically adjusts the speed of the two tracks to correct the robot's attitude, ensuring stable floating operation.

[0062] For data acquisition, the gimbal camera 5 detects the condition of the pipe wall above the water surface, the radar 6 scans the above-water portion of the pipe wall, and the sonar 7 detects the underwater portion of the pipe wall and siltation. The control system performs real-time fitting of the radar and sonar data to reconstruct the complete shape of the entire pipe cross-section. By integrating the detection data above and below the water surface with inertial navigation pose information, the three-dimensional coordinates of defects are accurately located.

[0063] When performing cleaning or spraying repair operations in floating walking mode, cleaning and repair are limited to the pipe wall area above the water surface. The gyroscope module 15 monitors the robot's roll, pitch, and yaw angles in real time. When a change in attitude is detected due to the backlash from the high-pressure water jet, the control system automatically adjusts the speed difference between the two tracks to generate a counter-compensation torque, with the attitude adjustment response time controlled within 100ms. Simultaneously, the water jet uses pulsed or segmented spraying modes to avoid cumulative deviation caused by continuous high-pressure spraying. The laser positioning light 8-2 continuously locks onto the target point during operation. If the target point deviation exceeds a set threshold (e.g., ±20mm), the system automatically pauses spraying and initiates an attitude correction program, resuming operation only after re-alignment. During operation, the power cable 10 and high-pressure water supply pipe 9 can be appropriately tightened via the ground station to provide auxiliary stability and reduce robot swaying using cable tension.

[0064] Please see Figure 13 and Figure 14 In full-pipe or ultra-high water level conditions, when the water level in the pipe exceeds two-thirds of the pipe diameter, the space above the water surface cannot support the robot's floating operation, and the walking mode is a bottom-crawling mode. Without the buoyancy module 17 installed, the buoyancy of the undercarriage is insufficient to lift the robot. The track blades, through the combined action of propulsion paddles and serrations, enable the robot to crawl on the bottom. The drive motor speed is automatically adjusted according to the resistance to ensure stable forward movement.

[0065] In this operating condition, the optical sensor's effectiveness is limited. The robot primarily relies on sonar 7 to emit sound waves and receive echoes to assess the thickness and distribution of underwater sediment, identifying structural defects such as pipe ruptures and deformations. Under conditions of high sediment buildup, sonar 7 can be mounted at the interface of the gimbal camera 5, raising its installation height and reducing interference from bottom sediment. By fusing sonar detection data with inertial navigation pose information, precise three-dimensional localization of underwater defects can be achieved. Under full-pipe conditions, the robot only performs detection and data acquisition tasks, without performing cleaning or repair operations.

[0066] For scenarios with multiple defects within the same pipeline, a multi-segment continuous operation mode can be adopted. When the distance between defects is greater than 50 meters, the robot carries the current working component and moves sequentially to each defect point to perform cleaning or repair operations. When the distance between defects is less than 20 meters, the robot first completes the full-line inspection and cleaning, then returns to the starting point to replace the repair component, and then repairs each defect point sequentially.

[0067] When a component malfunctions during operation, such as a clogged nozzle or an empty material tank, the control system automatically alarms and records the current position. The robot then carries the faulty component back to the wellhead, where ground personnel quickly replace it with a spare. Based on the recorded coordinates, the robot automatically navigates back to the original work point and continues to perform the unfinished tasks without needing to re-detect and reposition itself.

[0068] All test data, cleaning records, before and after repair images, and positioning information are automatically stored. After the work is completed, a comprehensive test and repair report containing the complete work process is generated with one click.

[0069] Specific example Please see Figure 12 , Figure 17 and Figure 18 The invention will be further illustrated below with a specific working example (DN600 half-water pipeline, which belongs to the half-pipe / high water level working condition).

[0070] When performing inspection and repair work on a DN600 semi-water pipeline, the height of the lifting frame assembly 2 is pre-adjusted according to the pipe diameter before the robot enters the pipeline. After entering the pipeline, the front-mounted water level sensor, pressure sensor, and gimbal camera vision recognition system identify that it is in a semi-pipe condition. At this time, the water level is about 30% of the pipe diameter, and the system automatically selects the detection mode that uses the camera as the primary sensor and sonar as a secondary sensor.

[0071] The robot moves at a speed of 0.15 m / s, with dual radars (6) continuously scanning to construct the pipe outline, a gimbal camera (5) capturing images of the inner wall, and sonar (7) detecting the underwater portion. When the system detects a material detachment at an interface and deformation of a pipe section, it accurately records the three-dimensional coordinates of the defect by fusing images from the gimbal camera, radar point clouds, sonar data, and pose information from the gyroscope module (15), and outputs a fused cross-sectional view of the sonar and radar data, as well as a 3D model.

[0072] After the robot moves to the vicinity of the defect, the operation module switches to the cleaning component. The laser lamp 8-2 emits a light spot to accurately locate the crack area, and the gimbal servo component 8-3 adjusts the attitude of the high-pressure water gun 8-1 to perform point cleaning on the crack area under a pressure of 1.5MPa until the attached material at the crack is removed and the main structure is exposed.

[0073] After cleaning, the operator quickly replaced the work module with the spray repair component on site, selected the fast-setting epoxy repair material, and the spray head moved along the preset path under the drive of the gimbal servo motor to spray the crack evenly with a coating thickness of about 3mm.

[0074] After the spraying is completed and the material has set, the repaired area is re-inspected using a gimbal camera 5 and a radar 6 to confirm that the repair effect meets the standards.

[0075] The entire testing, cleaning, and repair process is completed in a single tube entry, and a complete report containing before-and-after repair comparison data is automatically generated.

Claims

1. A method for using a multimodal detection and precision repair robot for adaptive pipeline operation under multiple working conditions, characterized in that, Includes the following steps: The robot enters the pipeline and uses a sensor detection system to identify the initial operating conditions inside the pipeline, including dry pipe condition, low water level condition, high water level condition, or full pipe condition. Based on the identified working conditions, select the walking mode of the blade track walking mechanism and configure the sensor combination of the multi-sensor detection system. The robot moves according to the selected walking pattern, uses the configured combination of sensors to detect the pipeline, obtain pipeline defect information and accurately locate it; Based on the test results, cleaning or spraying repair work will be carried out on the locations that require cleaning or repair. After completing the task, the work area is re-inspected and a work report is generated.

2. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 1, characterized in that, Under the aforementioned main pipeline operating conditions or low water level conditions: The walking mode is the pipe bottom walking mode. The blade-type track walking mechanism is not equipped with a buoyancy module and relies on the trapezoidal blades on the track surface to mechanically engage with the pipe bottom or silt layer to move forward. A gimbal camera and LiDAR are used to acquire image data and 3D point cloud data of the pipe wall surface; The cleaning or spraying repair operation involves using a laser positioning light to aim at the defect location, and then using a gimbal servo motor to drive a high-pressure water gun or spray head for targeted operation.

3. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 2, characterized in that, The buoyancy provided by the robot's undercarriage allows the robot to partially float, reducing ground pressure; the drive motor of the blade-type track walking mechanism automatically increases its speed, and the track blades simultaneously serve as propellers and spurs, maintaining the robot's propulsion and anti-slip capability.

4. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 1, characterized in that, Under the aforementioned high water level conditions: The walking mode is a floating walking mode. The blade-type track walking mechanism is equipped with a buoyancy module, which enables the robot to float on the water surface. The gimbal camera, lidar and sonar are used to acquire images of the pipe wall above the water surface, point cloud data of the pipe wall above the water surface and data of the underwater pipe wall and sedimentation, respectively. The trapezoidal blades of the blade-type track walking mechanism rotate in the water, acting as a paddle wheel to provide water propulsion and drive the motor speed to a certain speed. The cleaning or spraying repair work is carried out on the pipe wall area above the water surface.

5. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 4, characterized in that, Under the aforementioned high water level conditions: When performing cleaning or spraying repair operations in the floating walking mode, the robot's posture is kept stable in the following ways: The gyroscope module is used to monitor the changes in the robot's roll angle, pitch angle, and yaw angle in real time. When a change in attitude is detected due to the backlash from the high-pressure water jet, the forward and reverse rotation of the blade-type track walking mechanism is adjusted to adjust the robot's attitude.

6. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 4, characterized in that, In the floating walking mode, the robot's lifting frame assembly is lowered to its lowest position, causing the overall center of gravity to drop. When the gyroscope module detects that the roll angle exceeds a certain angle, the control system automatically adjusts the rotation speed of the two side tracks to correct the attitude.

7. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 4, characterized in that, Under the conditions of full pipe operation or ultra-high water level operation: The walking mode is a bottom crawling mode. The blade-type track walking mechanism is not equipped with a buoyancy module, and the robot moves along the bottom of the pipe. Sonar is used to obtain data on underwater pipe wall structural defects and sediment distribution. The robot only performs detection and data collection tasks and does not perform cleaning or spraying repair work.

8. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 7, characterized in that, Under full-pipe or ultra-high water level conditions, when severe siltation is detected, the sonar detector is moved from its mounting position at the front of the vehicle body to the mounting position of the gimbal camera to raise the detection height of the sonar and reduce the interference of bottom silt on the detection.

9. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 1, characterized in that, It also includes a quick switching step: After detecting a defect that needs to be repaired, the robot retreats to the work preparation position; Operators can replace the cleaning component with the spray repair component via a quick-change interface; The replaced spray repair component is driven by a gimbal servo motor and, combined with the three-dimensional positioning information from a multi-sensor detection system, guides the spray head to the defect area for spray repair.

10. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 1, characterized in that, It also includes steps involving multiple consecutive operations: When there are multiple defects in the pipeline and the distance between the defects is greater than 50 meters, the robot carries the current working components and moves to each defect point in sequence to perform cleaning or repair operations. When the defect spacing is less than 20 meters, the robot first completes the full-line inspection and cleaning, then returns to the starting point to replace the repair components, and then repairs each defect point in sequence.

11. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 1, characterized in that, It also includes steps for emergency switching: When a component malfunctions during operation, the control system will automatically sound an alarm and record the current location. The robot returns to the wellhead with the faulty component, where ground personnel replace it with a spare component. After the replacement is completed, the robot automatically navigates back to the original work point based on the recorded coordinate information and continues to perform the unfinished work.

12. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 1, characterized in that, The cleaning operation adopts pulse spraying or segmented spraying mode. The laser positioning light continuously locks onto the target point during the operation. When the target point deviation is detected to exceed the set threshold, the system automatically pauses spraying and starts the attitude correction program. The operation continues after realignment.

13. The method of using the adaptive pipeline operation multi-mode detection and precise repair robot according to claim 2 or 3, characterized in that, In dry pipe operation or low water level operation, the cleaning operation further includes: The robot moves to the front of the defect and stops. The laser positioning light forms an indicator spot on the pipe wall. The operator confirms the target point position through a handheld terminal. The high-pressure water gun is arranged parallel to the laser positioning light with a small spacing. The laser positioning light spot is the target point for cleaning and repair. Start the high-pressure water pump and perform fan-shaped sweeping or pinpoint impact cleaning on the defective area; During the cleaning process, the gimbal camera captures the cleaning effect in real time, and the operator can adjust the water gun posture or repeat the cleaning at any time.

14. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 1 or 13, characterized in that, After cleaning, a re-inspection step is also included: The high-pressure water pump is turned off, and the gimbal camera is used to magnify and observe the cleaning area to confirm the type, size, and severity of defects. If further evaluation is required, the lidar is used for a detailed local scan.

15. The method of using the adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 1 or 14, characterized in that, The spraying repair operation further includes: After confirming the defects that need to be repaired, the robot retreats to the work preparation position; Operators switch from cleaning components to spraying repair components by replacing the high-pressure water gun nozzle with a quantitative spray nozzle and the high-pressure water supply pipe with a repair material delivery pipeline. The control system switches to the spraying control interface, where the operator selects the type of repair material and sets the spraying parameters, including spraying width, spraying amount, and number of sprays. The robot moves back to the defect location and uses a multi-sensor fusion positioning system to guide the spray head to align with the cleaned defect area. Once the spraying process is initiated, the spray head moves along a preset path under the drive of the gimbal servo assembly to evenly cover the defects. After the coating is applied and the material has set, the gimbal camera captures the repair results and generates a repair record.

16. A multimodal detection and precision repair robot for adaptive pipeline operation under multiple working conditions, characterized in that, include: The robot body system includes an upper body assembly (1), a lower body assembly (3), and a lifting frame assembly (2) connecting the two. A blade-type track walking mechanism (4) is installed on both sides of the lower body assembly (3); The multi-sensor detection system includes a gimbal camera (5), radar (6), sonar (7), and a gyroscope module (15); The modular operating system includes a detachably connected pipe cleaning assembly (8), which is connected to an external water pump via a high-pressure water supply pipe (9-1); The control system (16) is connected to an external power source via a power supply cable (10).

17. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The lower body assembly (3) includes a lower body shell (11), a lower body sealing cover (3-1), a power output shaft sealing cover (3-2), and an air nozzle (3-3); the lower body assembly (3) is a sealed chamber that can generate a certain buoyancy. The air nozzle (3-3) is used to inject gas into the lower body assembly (3) to maintain positive pressure; The lower body assembly (3) is equipped with a drive motor system (14), which drives the blade track walking mechanism (4) through a bevel gear (13) and a power transmission shaft (12). The drive motor system (14), bevel gear (13) and power transmission shaft (12) are arranged at the rear of the robot.

18. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The blade-type track walking mechanism (4) includes a track frame and an integrated rubber blade track. The surface of the integrated rubber blade track is provided with trapezoidal blades with an inclined angle, which are used to mechanically engage with the bottom of the pipe or the silt layer in the case of dry pipe or low water level, and generate water propulsion force by rotating and pushing water in the case of high water level.

19. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The blade-type track walking mechanism (4) is equipped with a detachable buoyancy module (17); under the high water level condition, the blade-type track walking mechanism (4) is equipped with a detachable buoyancy module (17) to enable the robot to float on the water surface.

20. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The radar (6) is a dual lidar configuration, with the two radars symmetrically arranged on both sides of the front of the robot; the gyroscope module (15) is integrated inside the lower body assembly (3).

21. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The gimbal camera (5) and the sonar (7) have the same physical interface. Under severe siltation conditions, the two can interchange their installation positions to raise the installation height of the sonar (7).

22. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The pipe cleaning assembly (8) includes a high-pressure water gun (8-1), a laser light (8-2), and a gimbal servo assembly (8-3); the gimbal servo assembly (8-3) drives the high-pressure water gun (8-1) to achieve pitch and horizontal rotation; the laser light (8-2) is fixedly connected to the high-pressure water gun (8-1) and is used to project an indicator light spot on the pipe wall; The pipe cleaning assembly (8) is connected to the high-pressure water supply pipe (9-1) via a quick-connect interface.

23. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 16, characterized in that, The modular operation system also includes a spray repair component, which shares the same quick-change interface with the pipeline cleaning component (8) and can be installed interchangeably.

24. The adaptive pipeline operation multi-condition multimodal detection and precise repair robot according to claim 23, characterized in that, The spray repair component is a replacement part of the pipeline cleaning component (8), wherein the high-pressure water gun (8-1) is replaced with a quantitative spray head (8-4), the high-pressure water supply pipe (9-1) is replaced with a high-pressure material supply pipe (9-2), and the rear water supply tank is replaced with a repair material tank.

Images