Railway station hidden space steel structure disease automatic inspection method

Abstract

The present application belongs to the field of mechanical engineering robot technology, and particularly relates to a railway station hidden space steel structure disease automatic inspection method, comprising the following steps: S1, route autonomous planning; S2, task inspection; S3, periodic timing inspection; S4, accurate automatic obstacle avoidance; S5, intelligent problem identification; S6, automatic data analysis. The present application is aimed at common problems such as steel member corrosion, bolt loosening, weld cracking and metal roof leakage in the hidden space of high-speed rail station house cover steel structure. The intelligent inspection robot needs to have high-precision detection and identification capability, and can comprehensively and carefully inspect the steel structure in a narrow, dense rod piece, complex pipeline and lightless or weak light environment. The present application can improve the roof leakage monitoring and response speed under extreme weather, realize all-weather unmanned intelligent inspection, and reduce labor cost and safety risk.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of mechanical engineering robot technology, and particularly relates to a railway station concealed space steel structure disease automatic inspection method. BACKGROUND

[0002] With the increase of the operation life of railway station steel structure, problems such as steel member corrosion, bolt loosening, weld cracking, metal roof leakage and the like occur successively, which seriously affects the safe operation of railway. At present, the patrol mode mainly by visual inspection cannot carry out daily inspection on the steel structure in the concealed space such as suspended ceiling, the structural members in the concealed space are seriously corroded and damaged due to atmospheric corrosion, train wind vibration, roof leakage and the like, and some structural members are more prone to corrosion and fracture due to thin wall thickness, which seriously threatens the safety of the structure. The metal roof of the station is prone to leakage under extreme weather such as heavy rain and typhoon, but the leakage points cannot be found in time and are not easy to find, which seriously affects the experience of passengers waiting for trains.

[0003] The steel structure in the concealed space of the station is mainly a space grid structure with outer envelope roof panels and suspended ceiling panels. To realize the inspection in the concealed space, the main technical focus is to develop a robot carrier walking in the concealed space and a detection device carried on the robot carrier, and to carry out system integration.

[0004] However, the robot will encounter the following problems when walking and detecting in the space grid: (1) the structural characteristics of the truss determine that the intersection of the web member and the chord member forms a traffic obstacle; (2) the truss is a non-closed space, which is a truss structure connected together according to certain rules, and there is no obvious road for traffic. The node of the grid structure is numerous, the line of sight is seriously blocked, and the effective observation distance is short; (3) the lighting condition in the grid is poor, the scene is dark, and the reflectivity is extremely low, thereby increasing the requirement of the camera on the light adaptation and also increasing the difficulty of the light supplement technology; (4) during the inspection process, the video collection picture is blurred due to the resolution and pixel problems of the camera when shooting the scene far away, which increases the difficulty of AI recognition; (5) in the grid space, there are many members and the environment is complex, how to ensure the comprehensive coverage of the inspection and the accuracy of the positioning; (6) due to the narrowness of the grid structure, how to ensure the passability of the inspection robot. Because of the structural characteristics of the truss itself, it is still a technical problem to be solved to realize the crawling and inspection of the robot in the truss. SUMMARY

[0005] According to the above problems, the present application provides a railway station concealed space steel structure disease automatic inspection method, and the specific scheme is as follows:

[0006] A railway station concealed space steel structure disease automatic inspection method, comprising the following steps:

[0007] S1, route autonomous planning: according to the pre-set inspection route, or according to the inspection task, the planned route inspection is carried out;

[0008] S2, task inspection: according to different situations, the robot is triggered to carry out inspection, and the inspection mode includes remote automatic inspection, manual remote control inspection and special inspection, and various inspection modes support mutual switching, wherein: after switching to the remote automatic inspection mode, the remote inspection is started without intervention, the robot automatically inspects according to the planned route switching picture, and generates an inspection report uploaded to the ground monitoring base station;

[0009] S3, periodic timing inspection: according to the set inspection period, pictures are collected according to the pre-set inspection point, the robot automatically completes the inspection task, and generates an inspection report uploaded to the ground monitoring base station;

[0010] S4, accurate automatic obstacle avoidance: when inspecting, the robot measures the distance through the collector, quickly and accurately identifies the surrounding objects, and realizes rapid and accurate obstacle avoidance;

[0011] S5, intelligent identification of problems: the automatic identification software is deployed at the station end, and the engineering defects and safety hazards, or suspected engineering defects and safety hazards are identified and uploaded to the master station;

[0012] S6, automatic data analysis: the collected inspection data is connected to the comprehensive processing and analysis system, and the trend analysis and judgment of the data are carried out.

[0013] In the above technical solution, the hidden space steel structure of the station is mainly a space grid structure as a skeleton, and a roof panel and a suspended ceiling panel and the like are sealed outside. The three-dimensional point cloud data of the space grid is reconstructed, and the terrain is classified as an obstacle area of the robot according to the extracted geometric information of the terrain, a selection strategy in the passing area is formulated, and the trajectory (route) planning of the robot is carried out.

[0014] In the above technical solution, the robot calculates the target position based on the accurate environment map and the clear inspection route, wherein the navigation positioning algorithm includes the following steps:

[0015] S101, collect environment and robot motion data through laser radar nodes and encoder nodes;

[0016] S102, construct a global map by using a map service node, and realize real-time positioning by using a Monte Carlo positioning node;

[0017] S103, the navigation control node publishes the robot control message according to the current pose message and the target point pose of the robot, and drives the control node to subscribe to the message, parses it into the rotation speed of the dynamic wheel, and transmits it to the servo motor driver to execute motion control;

[0018] S104, time and space synchronization of multi-source sensor data is realized through a coordinate transformation node;

[0019] S105, a robot state publishing node is used for feeding back system running states in real time, and closed loop control is completed.

[0020] In the above technical solution, the robot is equipped with a light intelligent linkage, and in the process of patrolling, the robot is equipped with an intelligent follow-up light system and an intelligent light adjusting system, so that clear color images can be seen even in poor light or at night, and the appearance and state of the target to be inspected can be clearly distinguished.

[0021] In the above technical solution, the collector comprises a distance measuring sensor, a front standard definition camera and a rear standard definition camera, the robot can measure the distance, quickly and accurately identify the surrounding objects when patrolling through the collector, and the running speed and position of the robot are checked to plan the inspection route, modify the inspection points and speed to control the running of the robot, the robot form is expanded through the cooperation of the walking mechanism, the swing mechanism and the height adjusting mechanism, rapid and accurate obstacle avoidance is realized, the camera view angle is adjusted through the cooperation of the walking mechanism, the swing mechanism and the height adjusting mechanism, the detection object is shot from all directions without dead angle, and the key target specified by the monitoring system is automatically tracked to realize directional video acquisition.

[0022] In the above technical solution, a model capable of automatically identifying various engineering defects and safety hazards or suspected engineering defects and safety hazards on the steel structure of the high-speed rail station building is constructed, and the construction method of the model is as follows: first, the collected images are preprocessed, then the key features in the images are extracted, and finally the model uses the above features for fault classification and identification.

[0023] The second inventive purpose of the present application is to provide an automatic inspection system for implementing the automatic inspection method of the hidden space steel structure disease of the railway station, which can also be called a spatial grid intelligent inspection robot system.

[0024] The robot system comprises a comprehensive control system, a communication transmission system, a power supply system and an inspection management platform server.

[0025] In the technical scheme, the robot system comprises a control module, a communication transmission module, a positioning and navigation module, a power supply control module, a temperature control module, an obstacle avoidance module and a vision module.

[0026] In the technical scheme, the robot system communicates with the ground monitoring base station through the communication transmission module (wireless transmission system), the comprehensive processing and analysis system accesses the patrol data collected by the robot, completes data interaction, makes decisions according to preset model information and environmental conditions, and realizes high-reliability control operation of the robot. The control module obtains control instructions by communicating with the ground monitoring base station, controls, executes and completes precise positioning, autonomous navigation, obstacle avoidance, charging and temperature control functions, and the system management platform performs data analysis, judges the equipment and environmental state, discovers abnormalities and alarms in time.

[0027] The third object of the present application is to provide a net rack operation robot, comprising a body, further comprising:

[0028] The walking mechanism comprises a set of power limb mechanisms arranged diagonally at the bottom of the body and a set of passive limb mechanisms arranged diagonally at the bottom of the body, the power limb mechanisms are provided with power wheel mechanisms at the ends, and the passive limb mechanisms are provided with passive wheel mechanisms at the ends.

[0029] The swing mechanism comprises two sets of multi-axis mechanical arms arranged diagonally at the top of the body.

[0030] The height adjustment mechanism comprises a folding arm arranged at the middle bottom of each bottom edge of the body.

[0031] The detection and collection device is mounted on the body.

[0032] In the technical scheme, the power limb mechanism comprises a rotating limb one, a rotating motor one, a rotating joint, a rotating motor two, a rotating motor three, a rotating limb two and a shock absorber, and the structure of the passive limb mechanism is the same as that of the power limb mechanism. In the power limb mechanism, the rotating motor one is fixedly connected to the body by bolts, the rotating joint is fixedly connected to the rotating motor one by bolts, the rotating limb one is left-right symmetrical, the rotating motor two is fixedly connected to the end of the rotating limb one by bolts, the rotating motor three is fixedly connected to the front end of the rotating limb one by bolts, the rotating joint is fixedly connected to the rotating motor two by bolts, the end of the rotating limb two is fixedly connected to the rotating motor three by bolts, one end of the shock absorber is hingedly connected to the corresponding lug plate of the rotating limb two through a pin shaft, the other end of the shock absorber is hingedly connected to the corresponding lug plate of the power wheel frame through a pin shaft, and the lug plate at the front end of the rotating limb two is hingedly connected to the corresponding lug plate of the power wheel frame through a pin shaft, so that the power wheel mechanism can have a certain pressure on the chord.

[0033] In the above technical solution, the difference between the passive limb mechanism and the powered limb mechanism is only the wheel mechanism at the end, the passive limb mechanism is a passive wheel mechanism, and the rest is completely the same as the powered limb mechanism.

[0034] In the above technical solution, the folding arm includes a movable limb, a telescopic mechanism, a shock absorber, and a rotating motor four, the telescopic mechanism includes a plurality of long rods and short rods connected in a scissors type, and the folding arm is provided with a powered wheel mechanism at the end.

[0035] In the above technical solution, the powered wheel mechanism includes a powered wheel frame, a powered wheel motor, and a conical rubber wheel.

[0036] In the above technical solution, the passive wheel mechanism includes a magnetic wheel frame and a magnetic wheel, the magnetic wheel is a conical wheel structure composed of a plurality of magnetic wheel monomers sharing an axle and gradually increasing in diameter, and each magnetic wheel monomer is fixed on the axle at equal intervals.

[0037] The fourth object of the present application is to provide a working method of a net rack working robot, specifically:

[0038] When the robot works in the transverse net rack, all the powered wheel mechanisms and passive wheel mechanisms acting on the transverse chord are on the transverse chord, the transverse folding arm is unfolded and walks on the transverse chord, and the longitudinal folding arm is folded.

[0039] When working in the longitudinal net rack, all the rotating limbs are rotated by 90° to the longitudinal chord, the transverse folding arm is folded, and the longitudinal folding arm is unfolded and walks on the longitudinal chord.

[0040] In combination with all the above technical solutions, the present application has the following advantages and positive effects:

[0041] 1. For the common problems of steel member corrosion, bolt loosening, weld cracking and metal roof leakage in the concealed space of the steel structure of the high-speed rail station building, the intelligent inspection robot needs to have high-precision detection and identification capability, and can conduct comprehensive and detailed inspection on the steel structure in a narrow, densely bar, complex pipeline and environment without light or weak light space. Through real-time monitoring and recording of the disease situation, accurate data support is provided for subsequent maintenance and maintenance to ensure the safety of the structure.

[0042] 2. Improve the response speed of roof leakage monitoring under extreme weather;

[0043] Leaks and poor drainage in the metal roofs of high-speed railway stations have long been a key issue for the industry, with serious leaks causing extremely negative social impacts. Intelligent inspection robots need to possess real-time monitoring and rapid response capabilities. Daily monitoring of roof leaks, gutter flow, and gutter blockages allows for precise repair of leaks and clearing of blockages, providing preventative protection against extreme weather conditions.

[0044] 3. Achieve all-weather, unmanned, intelligent inspection;

[0045] The intelligent inspection robot has the ability to operate online 24 / 7 and without human intervention. It can continuously and stably carry out inspections according to the set inspection path, and can fully meet the large-area inspection needs of the steel structure of the roof of high-speed railway station. Through intelligent inspection process and data analysis, it can improve the reliability and accuracy of inspection and reduce the omissions and misjudgments caused by human factors.

[0046] 4. Reduce labor costs and safety risks;

[0047] By introducing intelligent inspection robots, the safety risks faced by inspection personnel can be reduced while ensuring the quality of inspections, especially in small, complex, and potentially hazardous hidden spaces, thus ensuring the safety and efficiency of inspection work. Attached Figure Description

[0048] The technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that these drawings are designed for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, unless specifically indicated, these drawings are intended only to conceptually illustrate the structural construction described herein and are not necessarily drawn to scale.

[0049] Figure 1 This is a perspective view of the space frame operation robot described in this invention;

[0050] Figure 2 This is a front view of the standby state of the space frame operation robot described in this invention;

[0051] Figure 3 This is a side view of the standby state of the space frame operation robot described in this invention;

[0052] Figure 4 This is a bottom view of the standby state of the space frame operation robot described in this invention;

[0053] Figure 5 This is an exploded view of the overall structure of the space frame operation robot described in this invention;

[0054] Figure 6 for Figure 1Enlarged view of middle A part;

[0055] Figure 7 Exploded view of powered limb of the grid working robot according to the present application;

[0056] Figure 8 Exploded view of passive limb of the grid working robot according to the present application;

[0057] Figure 9 Perspective view of folding limb of the grid working robot according to the present application;

[0058] Figure 10 Exploded view of folding limb of the grid working robot according to the present application;

[0059] Figure 11 Exploded view of powered wheel of the grid working robot according to the present application;

[0060] Figure 12 Exploded view of passive wheel of the grid working robot according to the present application;

[0061] Figure 13 Steering of the grid working robot according to the present application Figure 1 ;

[0062] Figure 14 Steering of the grid working robot according to the present application Figure 2 ;

[0063] Figure 15 Structural block diagram of the robot system according to the present application;

[0064] Figure 16 General structure of the function logic of the intelligent mobile inspection system;

[0065] Figure 17 Business process of the intelligent mobile inspection system;

[0066] Figure 18 Navigation and positioning relationship diagram of the intelligent inspection system;

[0067] Figure 19 Working logic flow chart;

[0068] Figure 20 Abnormality processing program flow chart;

[0069] In the diagram, 1. Rotary limb one, 11. Rotary motor one, 12. Rotary joint, 13. Rotary motor two, 14. Rotary motor three, 15. Rotary limb two, 16. Shock absorber, 2. Movable limb, 21. Long rod one, 22. Short rod one, 23. Long rod two, 24. Long rod three, 25. Long rod four, 26. Short rod two, 27. Shock absorber, 28. Wheel rod, 29. Rotary motor four, 3. Body, 4. Magnetic wheel frame, 41. First bearing, 42. Magnetic wheel and axle, 5. Power wheel frame, 51. Power wheel motor, 52. Conical rubber wheel, 53. Second bearing, 6. Multi-axis robotic arm. Detailed Implementation

[0070] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0071] The present invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0072] Example 1: Robot for Space Frame Operation

[0073] like Figures 1-4 As shown in the figure, this embodiment discloses a space frame operation robot, which includes a walking mechanism, a body 3 and a multi-axis robotic arm 6.

[0074] like Figure 5 As shown, two sets of multi-axis robotic arms 6 are arranged diagonally at the top of the fuselage 3, and are fixedly connected by bolts. Data acquisition and detection equipment can be mounted on the multi-axis robotic arms. At the bottom of the fuselage 3, one set of two powered limb mechanisms is arranged diagonally, and the other set of two passive limb mechanisms is arranged diagonally. The difference between the powered and passive limb mechanisms lies in their end-effector wheel mechanisms; the powered limb mechanisms have powered wheels at their ends, while the passive limb mechanisms have passive wheels at their ends. In addition, a folding arm is arranged in the middle of each bottom edge of the fuselage 3.

[0075] like Figure 6 , 7As shown, in the power limb mechanism, rotary motor 11 is fixedly connected to the body 3 by bolts, rotary joint 12 is fixedly connected to rotary motor 11 by bolts, rotary limb 11 is symmetrical left and right, rotary motor 2 13 is fixedly connected to the end of rotary limb 11 by bolts, rotary motor 3 14 is fixedly connected to the front end of rotary limb 11 by bolts, rotary joint 12 is fixedly connected to rotary motor 2 13 by bolts, the end of rotary limb 2 15 is fixedly connected to rotary motor 3 14 by bolts, one end of shock absorber 16 is hinged to the corresponding lug of rotary limb 2 15 by a pin, the other end of shock absorber 16 is hinged to the corresponding lug of power wheel frame 5 by a pin, and the lug at the front end of rotary limb 2 15 is hinged to the corresponding lug of power wheel frame 5 by a pin, thereby maintaining a certain pressure between the power wheel mechanism and the chord.

[0076] like Figure 8 As shown, the only difference between the passive limb mechanism and the powered limb mechanism is the wheel mechanism at the end. The passive limb mechanism has a passive wheel mechanism at the end, and everything else is exactly the same as the powered limb mechanism.

[0077] like Figure 9 , 10 As shown, movable limb 2 is fixedly connected to rotary motor 4 29 by bolts. Rotary motor 4 29 is fixedly connected to the fixed plate at the bottom of the body 3 by bolts. Long rod 1 21, short rod 1 22, long rod 2 23, long rod 3 24, long rod 4 25, and short rod 2 26 are hinged by pins to form a telescopic mechanism. One end of shock absorber 27 is hinged to long rod 4 25 by a pin, and the other end is hinged to the corresponding ear plate of wheel rod 28 by a pin. The ear plate at the end of wheel rod 28 is hinged to the ear plate at the end of long rod 4 25 by a pin. The power wheel frame 5 is hinged to the corresponding ear plate of wheel rod 28 by a pin. The folding arm is divided into a transverse folding arm and a longitudinal folding arm according to its orientation. When the robot works in the transverse grid, all movable limbs are on the transverse chord. The transverse folding limbs unfold for walking, and the longitudinal folding limbs fold. Figure 13 As shown. When working inside the longitudinal space frame, all movable limbs rotate 90° to the longitudinal chord, the lateral folding arms retract, and the longitudinal folding arms unfold for movement.

[0078] like Figure 11 As shown, the power wheel motor 51 is fixedly connected to the power wheel frame 5 by bolts, the end of the conical rubber wheel 52 is fixedly connected to the power wheel motor 51 by bolts, the second bearing 53 is installed on the outside of the power wheel frame 5, and the front end of the conical rubber wheel 52 is hinged to the second bearing 53 through a wheel axle.

[0079] like Figure 12 As shown, the magnetic wheel frame 4 has first bearings 41 arranged on both sides. The magnetic wheel is hinged to the first bearings 41 on both sides through an axle. The magnetic wheel is a conical wheel structure composed of several magnetic wheel units sharing a common axle and with gradually increasing diameters. Each magnetic wheel unit is arranged at equal intervals and fixed on the axle. Figure 13As shown, when the robot walks between the chord bars, the tapered rubber wheels of the folded arms, the tapered rubber wheels of the powered limbs, and the magnetic wheels of the passive limbs are all on the chord bars. The tapered rubber wheels of the powered limbs are diagonally distributed, and the magnetic wheels of the passive limbs are diagonally distributed, so that the tapered rubber wheels of the powered limbs and the magnetic wheels of the passive limbs are in a relative position, and the large end of the tapered wheel is opposite. The tapered rubber wheels and the magnetic wheels can be tangent to the circumference of the chord bar, can form stable support to make it walk stably between the chord bars, and can avoid the problems of being stuck or derailing. When avoiding obstacles, one of the tapered rubber wheels or magnetic wheels on the chord bar is lifted, and the remaining tapered rubber wheels or magnetic wheels continue to walk on the chord bar, which can realize rapid, accurate and stable obstacle avoidance. In addition, the net rack operating robot can adapt to the change of the width between the chord bars and the change of the distance between the adjacent chord bars through the passive wheel design with magnetism, which facilitates the stable walking of the robot on the chord bar and avoids the problems of being stuck or derailing, and facilitates subsequent inspection operation, providing a good foundation for high-altitude operation.

[0080] Embodiment 2: Inspection method of net rack operating robot

[0081] The net rack operating robot in the above embodiment 1 is mainly applied in the inspection operation of space grid structure, and the inspection process is as follows:

[0082] 1. Robot installation and equipment debugging: when performing inspection operation, first send the inspection robot to the starting point of the track through a basket or an elevator, connect the power supply and control system of the inspection robot, and perform cable wiring and connection; perform initialization setting and debugging of the inspection robot, ensure normal communication with the control system; perform function test of the inspection robot, including test of inspection, monitoring and alarm functions; adjust parameters and settings of the inspection robot to optimize its running effect; perform running test of the inspection robot to simulate actual scene and ensure its normal operation and meet project requirements.

[0083] 2. Inspection robot running process:

[0084] 1) Route self-planning: plan the route according to the pre-set inspection route or according to the inspection task.

[0085] 2) Remote automatic inspection: after the operation and maintenance personnel start the inspection remotely, the system automatically switches the picture according to the inspection route without intervention.

[0086] 3) Periodic timing inspection: according to the set inspection period, pictures are collected at the predetermined inspection points, and the inspection task is automatically completed, and an inspection report is generated and uploaded to the master station.

[0087] 4) Intelligent lighting linkage: intelligent linkage of light control system during inspection, even if the light is poor or at night, clear color images can be seen, and the appearance and state of the inspection target can be clearly distinguished.

[0088] 5) Precise automatic obstacle avoidance: distance measurement by sensor, fast and accurate identification of surrounding objects, rapid and accurate obstacle avoidance.

[0089] 6) Intelligent problem identification: automatic identification software is deployed at the station to identify steel member corrosion and bolt loss and upload to the main station.

[0090] 7) Automatic data analysis: collected inspection data is connected to the comprehensive processing and analysis system, which can analyze and judge the trend of the data.

[0091] Shooting detection rules: one complete observation task needs to run twice along the path (single run only observes one side), each grid unit stays once, and each shooting task is: 14 node observation photos, 12 side member observation photos, 2 front member observation photos, a total of 28 photos.

[0092] It should be noted that the net frame work robot can also be applied to high-altitude spraying work on roof inclined purlin, and when the high-altitude spraying work is completed, the corresponding spraying equipment needs to be carried on the net frame work robot.

[0093] Example 3: Intelligent inspection robot system

[0094] For the characteristics of small hidden space, no light, dense members, and many obstacles, the net frame work robot in the above example 1 is equipped with an intelligent inspection robot system (also called intelligent mobile inspection system), which is composed of a robot system, an integrated control system (system processing software and peripheral equipment), a detection device, a power carrier communication system, a video server, a ground monitoring base station and other hardware, which is used to realize the walking work of the robot in the space grid, and solve the problems of environmental perception, path planning, carrier adaptability, etc.

[0095] The robot system is mainly composed of a comprehensive control system, a communication transmission system, a power supply system, a patrol management platform server, etc. The control system block diagram of the robot is as shown in Figure 15 .

[0096] The PC machine as the control center of the robot can control the motor, relay and other actuators according to the control instructions of the ground monitoring base station and the data collected by the sensors carried by the robot itself, process and analyze the information collected by the sensors in real time, obtain the accurate position of the robot, temperature, distance of obstacles in front and back, battery capacity and other data. And through communication with the ground monitoring base station, the control instructions are obtained to control, execute and complete the functions of accurate positioning, autonomous navigation, obstacle avoidance, charging, temperature control and the like. And the system management platform is used for data analysis, judgment of equipment and environment state, discovery of abnormalities and timely alarm and the like.

[0097] The monitoring terminal communication: through the TCP / IP communication protocol, the remote monitoring terminal program is connected to obtain the control instructions in real time, and the robot information is fed back.

[0098] The inspection robot communicates with the ground monitoring base station through the wireless transmission system to interact data, makes decisions according to the preset model information and environmental conditions and the like to realize the high reliable control operation of the robot.

[0099] The control mode of PC+distributed controller can realize real-time interconnection with other bus type electrical detection equipment. The function distributed structure is adopted, the working speed and control performance of the controller are obviously improved, and the problem of multi-bus real-time interconnection can be well solved.

[0100] Through the servo motion and synchronous control technology, the effective and accurate control of the walking mechanism, swing mechanism and height adjustment mechanism in the dynamic monitoring process is realized.

[0101] The specific functions of each device are as follows:

[0102] The PC machine is connected with integrated serial port interface card, high-speed digital I / O card, analog I / O card and low-speed digital I / O card through the bus.

[0103] The interface communication layer mainly refers to realizing the interaction between the inspection robot and the ground monitoring base station through the wireless network equipment, realizing the seamless connection of the monitoring information system of various fully mechanized mining equipment, realizing the multi-channel high-definition color video transmission and the real-time transmission of production data.

[0104] The entire robot system communicates and controls in real time with the backend server via a wireless local area network (WLAN) established in the on-site delivery area. The robot itself connects to the WLAN via wireless signals for real-time image and control command transmission. The robot integrates a 5G CPE (fully compatible with 4G networks). The operator must provide a dedicated 5G network card with a fixed IP address for the robot's 5G CPE. The inspection data collected by the robot is connected to the enterprise intranet via a dedicated network line provided by the operator. The enterprise intranet uses the fixed IP address provided by the operator to access the robot's interface. Each of the robot's auxiliary devices is also equipped with a 5G CPE, similarly assigned a fixed IP address by the operator. The robot and its auxiliary devices communicate and coordinate through the fixed IP addresses assigned by the operator.

[0105] The entire robot system at the inspection station uses a single IP address for external communication. An NVR (Network Video Recorder) and workstation computers are installed in the monitoring center for robot control and video storage. The system needs to allocate IP addresses for the NVR and workstations.

[0106] Environmental perception and autonomous navigation are the foundation for mobile robots to achieve true autonomy. By reconstructing surfaces from 3D point cloud data of a spatial grid and classifying the terrain into obstacle areas and passable areas based on extracted geometric information, a selection strategy is formulated within the passable areas to plan the robot's trajectory.

[0107] Because robot motion is dynamic and interconnected, real-time fusion of terrain information and the robot's own posture information becomes crucial for stable robot control. Therefore, constructing a large closed-loop behavioral control model integrating "terrain understanding and gait" is essential to addressing the challenges of integrating the robot with the spatial grid environment. This involves using vision / LiDAR-based SLAM technology to perceive the environment, adapt to issues such as untimely gait adjustments, and inaccurate foot placement.

[0108] Example 4: Intelligent Inspection Robot System

[0109] 1. Overall Functional Logic Structure of the Intelligent Inspection Robot System: To achieve real-time information perception, quickly detect engineering defects and safety hazards, efficiently respond to safety risks, provide real-time information feedback, and support decision-making, this system comprises four layers: a data layer, a parsing layer, a fusion layer, and an application layer. The data layer allows setting inspection plans or manual control of the robot to reach designated locations; the parsing layer collects monitoring data and identifies defect types and parameters; the fusion layer integrates defect data, power status, and expert evaluation results to generate a preliminary report; and the application layer visualizes the report and supports interactive operation and downloading by users. The specific architecture is as follows: Figure 16 As shown, where:

[0110] A data layer including a planned inspection module and a manual inspection module, configured to set an inspection plan, dynamically display a robot position, support fine operation, and input a point position;

[0111] An analysis layer including a defect identification module and an intelligent monitoring module, configured to identify corrosion, cracks, and water seepage defects, calculate parameters of the defects, and perform lightweight processing on monitoring data and analog reading analysis;

[0112] A fusion layer including a robot power management module, a defect marking module, and an automatic report generation module, configured to monitor robot power in real time, locate defect information, jointly analyze data, and integrate remote expert evaluation results;

[0113] An application layer including a defect visualization display module, a defect tracking module, and a report management module, configured to mark an inspection result in real time, generate a visual report, and provide a preview download function.

[0114] Further, in the embodiment, the planned inspection module of the data layer can include:

[0115] A roof scanning plan setting unit configured to define a periodic inspection path;

[0116] A self-defined plan setting unit configured to support a user to adjust an inspection parameter according to a requirement;

[0117] A robot position dynamic display unit configured to update a robot position in real time in combination with a map interface.

[0118] Further, in the embodiment, the defect identification module of the analysis layer can include:

[0119] A corrosion defect identification unit configured to determine existence of corrosion based on image analysis;

[0120] A crack calculation unit configured to measure a crack length and a crack width;

[0121] A water seepage identification and range calculation unit configured to determine a water seepage area and a diffusion range thereof by using an image segmentation technology.

[0122] Further, in the embodiment, the automatic report generation module of the fusion layer can include:

[0123] A monitoring data joint analysis unit configured to integrate multi-source sensor data to generate a comprehensive analysis result;

[0124] An expert remote evaluation interface configured to support an external expert system to access and feed back an evaluation conclusion.

[0125] Further, in the embodiment, the report management module of the application layer can include:

[0126] Real-time marking unit, allowing users to dynamically mark defect locations during inspection;

[0127] Report preview unit, providing an interactive interface to preview report content;

[0128] Multi-format download unit, supporting PDF and Excel format inspection report export.

[0129] 2. The intelligent inspection robot system can achieve the following functions:

[0130] 1) Motion control function: control the operation of the inspection robot, view the running speed and position of the robot, plan the inspection route, modify the inspection point position and speed; and adjust the camera view angle by controlling the translation, deflection, and pitch motion of the multi-axis robot arm, achieve full-coverage and no-dead-angle shooting of the detection object, and automatically track the key target specified by the monitoring system to achieve directional video acquisition.

[0131] 2) Data acquisition and storage function: supports automatic image focusing, video playback, stopping, screenshot, recording, full-screen display, and other functions;

[0132] 3) The intelligent inspection robot system management platform software adopts B / S architecture design. In addition to controlling the robot to perform related inspection tasks, the management platform software also has functions such as data acquisition, statistics, retrieval, reporting, and intelligent analysis.

[0133] 3. Business process of the intelligent inspection robot system:

[0134] As shown in Figure 17 , when using the system, the inspection methods include automatic inspection and manual inspection. Automatic inspection can be triggered by the robot according to different situations. One is to set a timing task for the robot to automatically inspect at regular intervals. Two is to set warning rules and special working conditions for the robot to complete special inspections. When the monitoring data is warned, automatic inspection is started in time, and key parts are automatically associated for key viewing of the warning parts.

[0135] During the inspection process, various types of acquisition and detection equipment carried can collect comprehensive information of the inspection site, and store and analyze the comprehensive information. Based on image recognition technology, the inspection image data is analyzed and recognized, and the intelligent analysis result is permanently stored in the server, and the analysis result is marked on the original picture to form a defect management library and record the defect parameter information. The final inspection report contains all the above inspection information.

[0136] Routine inspection content: bolt and fastener inspection; steel structure corrosion inspection; weld cracking inspection; web plate butt joint weld inspection; structure deformation and size inspection; metal roof leakage inspection.

[0137] Patrol Results Display: During automated patrols, the robot captures high-definition images of suspected anomalies and promptly displays the latest photos on the platform's overview page. Each photo has a recorded capture location, and the robot can be dispatched immediately to that location for manual verification. Each photo can be zoomed in to full screen for detailed viewing.

[0138] The intelligent analysis uses a typical fault identification algorithm model: by training a large amount of image data, the model can automatically identify various faults on the steel structure of the roof of the high-speed railway station.

[0139] The model first preprocesses the acquired images, performing noise reduction and enhancement, then extracts key features such as color, texture, and shape. Finally, the model uses these features for fault classification and identification. Deep learning algorithms, with their powerful feature extraction and classification capabilities, can accurately identify faults in steel components such as corrosion, weld cracks, missing fasteners, leaks, and blockages. Furthermore, by combining real-world conditions, statistical analysis, and interactive methods such as threat alerts, the model can output a complete monitoring report in one stop, thereby simplifying the comprehensive management of operation and maintenance processes in industrial scenarios.

[0140] The intelligent inspection robot utilizes a multi-axis robotic arm to automate its deployment, significantly improving the coverage of high-definition cameras and effectively avoiding obstruction from obstacles. Additionally, the robot is equipped with:

[0141] 1) The inspection robot is equipped with an intelligent following light system and an intelligent dimming system, which can automatically adjust the brightness according to the ambient light to ensure the shooting effect of the camera in the grid space;

[0142] 2) Select a high-resolution camera and equip it with a 30x optical zoom lens to ensure clearer output video images, thereby improving the accuracy of AI recognition and analysis;

[0143] 3) To ensure comprehensive 3D image acquisition, the poles are divided into multiple region bands according to the camera's acquisition range, such as A01-01. Each band is further subdivided into four acquisition surfaces: east, south, west, and north, such as A01-01-01. To ensure image clarity, the camera focuses and acquires images whenever the inspection robot reaches an inspection point, transmitting the acquired images to the backend. The backend then integrates the images according to their serial numbers and compares them with the 3D model to determine the coverage.

[0144] 4. Autonomous navigation and positioning:

[0145] like Figure 18As shown, the robot calculates the target position based on the accurate environment map and the clear work path through the navigation positioning algorithm to achieve autonomous navigation positioning. The navigation control node is the key of the navigation control service. The navigation control node publishes the robot control message according to the current pose message and the target point pose of the robot. The drive control node subscribes to the message, parses it into the speed of the dynamic wheel, and transmits it to the servo motor drive.

[0146] Positioning technology: correct the cumulative error of the robot, and integrate radio frequency positioning, magnetic guide switch, machine vision positioning calibration and motion control method to improve the reliability of the robot positioning system. The robot uses an optical pulse encoder, and a reset magnet is installed at the first end of the track to clear the displacement coordinates of the encoder. The comprehensive navigation positioning technology based on the integration of multiple navigation positioning technologies is the guarantee for the stable and normal work of the inspection robot, and is also the basis for realizing unmanned inspection.

[0147] 5. System business process:

[0148] When using the system, the robot can be triggered for inspection according to different situations, and the modes mainly include automatic routine inspection, manual remote control inspection, special inspection, etc., and various modes support mutual switching.

[0149] The inspection logic execution process is as shown in Figure 19 , and the abnormality processing program flow is as shown in Figure 20 .

[0150] Application Example 1:

[0151] The railway station hidden space steel structure disease automatic inspection method provided in the above embodiment can also run on a computer device, which comprises at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the method in the above embodiment when executing the computer program.

[0152] Application Example 2:

[0153] The railway station hidden space steel structure disease automatic inspection method provided in the above embodiment can also run on a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program can implement the method in the above embodiment when executed by a processor.

[0154] Application Example 3:

[0155] The railway station hidden space steel structure disease automatic inspection method provided in the above embodiment can also run on an information data processing terminal, and the information data processing terminal is used to provide a user input interface to implement the method in the above embodiment when executed on an electronic device. The information data processing terminal is not limited to a mobile phone, a computer, and a switch.

[0156] Application Example 4:

[0157] The railway station concealed space steel structure disease automatic inspection method provided by the above embodiment can also run on a server, and the server is used to provide a user input interface to implement the method in the above embodiment when executed on an electronic device.

[0158] Application Example 5:

[0159] The railway station concealed space steel structure disease automatic inspection method provided by the above embodiment can also run on a computer program product, and when the computer program product is running on an electronic device, the electronic device can implement the method in the above embodiment when executed.

[0160] The present application can implement all or part of the steps of the above-mentioned embodiment method, which can be completed by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer-readable medium at least includes any entity or device capable of carrying the computer program code to the photographing device / terminal equipment, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, magnetic disk or optical disk, etc.

[0161] The above only describes the preferred embodiments of the present application, and it should be understood that the described embodiments are part of the embodiments of the present application, but not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application. The present application is not limited to the forms disclosed herein, and should not be considered as excluding other embodiments, but can be used in various other combinations, modifications and environments. The modifications and changes made by those skilled in the art without departing from the spirit and scope of the present application shall be within the scope of protection of the appended claims of the present application.

Claims

1. A railway station concealed space steel structure disease automatic inspection method, characterized in that: Comprise the following steps: ​ S1, route autonomous planning: according to the pre-set route or according to the inspection task, planning route inspection; S2, task inspection: according to different situations, the robot is triggered to carry out inspection, and the inspection mode includes remote automatic inspection, manual remote control inspection and special inspection, and various inspection modes support mutual switching, wherein: after switching to the remote automatic inspection mode, the remote start inspection is not required, the robot automatically inspects according to the planned route, and generates an inspection report and uploads it to the ground monitoring base station; S3, periodic timing inspection: according to the set inspection period, pictures are collected at the predetermined inspection points, and the robot automatically completes the inspection task and generates an inspection report and uploads it to the ground monitoring base station; S4, accurate automatic obstacle avoidance: when inspecting, the robot measures the distance through the collector, quickly and accurately identifies the surrounding objects, and realizes rapid and accurate obstacle avoidance; S5, intelligent identification of problems: the automatic identification software is deployed at the station to identify engineering defects and safety hazards, or suspected engineering defects and safety hazards, and upload them to the ground monitoring base station; S6, automatic data analysis: the collected inspection data is connected to the comprehensive processing and analysis system to analyze the trend of the data; The robot is a net frame operation robot, which comprises a body and further comprises: A walking mechanism comprising a set of power limb mechanisms arranged diagonally at the bottom of the body, a set of passive limb mechanisms arranged diagonally at the bottom of the body, the power limb mechanism being provided with a power wheel mechanism at the end, and the passive limb mechanism being provided with a passive wheel mechanism at the end; the power wheel mechanism comprises a power wheel frame, a power wheel motor and a conical rubber wheel, the passive wheel mechanism comprises a magnetic wheel frame and a magnetic wheel, the magnetic wheel is a conical wheel structure composed of a plurality of magnetic wheel monomers sharing an axle and gradually increasing in diameter, and each magnetic wheel monomer is arranged and fixed on the axle at equal intervals; A swing mechanism comprising two sets of multi-axis mechanical arms arranged diagonally at the top of the body; A height adjustment mechanism comprising a folding arm arranged at the middle bottom of each bottom side of the body; A detection and collection device mounted on the body; Wherein: the robot walks and works in the spatial grid of the steel structure of the railway station hidden space in a narrow, dark, dense rod, complex pipeline and dark or weak light environment, the robot realizes form expansion through the cooperation of the walking mechanism, the swing mechanism and the height adjustment mechanism, and adapts to walking between the chords of the horizontal and vertical net frames, when the robot walks between the chords, the conical rubber wheel of the power limb mechanism and the magnetic wheel of the passive limb mechanism are in a relative position, and the large end of the conical wheel is opposite, the conical rubber wheel and the magnetic wheel can be tangent to the circumference of the chord, forming stable support to avoid being stuck or derailing.

2. The automatic inspection method for steel structure defects in concealed spaces of railway passenger stations according to claim 1, characterized in that: The robot calculates the target position based on the accurate environment map and the clear inspection route through the navigation positioning algorithm, wherein the navigation positioning algorithm comprises the following steps: S101, collect environment and robot motion data through laser radar nodes and encoder nodes; S102, construct a global map using a map service node, and realize real-time positioning through a Monte Carlo positioning node; S103, the navigation control node publishes a robot control message according to the current pose message of the robot and the target point pose, and drives the control node to subscribe to the message, parses the rotational speed of the driving wheel, and transmits it to the servo motor driver to execute motion control; S104, time and space synchronization of multi-source sensor data is realized through a coordinate transformation node; S105, the robot state publishing node feeds back the system running state in real time to complete closed-loop control.

3. The method for automatically inspecting diseases of a railway station concealed space steel structure according to claim 1, characterized in that: When the robot patrols, the distance is measured by the collector, the surrounding objects are quickly and accurately identified, the running speed and position of the robot are checked, the patrol route is planned, the patrol points and speed are modified to control the running of the robot, the robot form is expanded through the cooperation of the walking mechanism, the swing mechanism and the height adjustment mechanism, the obstacles are quickly and accurately avoided, the camera view is adjusted through the cooperation of the walking mechanism, the swing mechanism and the height adjustment mechanism, the detection object is shot from all directions without dead angle, and the specified key target is automatically tracked to realize directional video acquisition.

4. The method for automatically inspecting diseases of a railway station concealed space steel structure according to claim 1, characterized in that: In step S5, a model capable of automatically identifying various engineering defects and safety hazards, or suspected engineering defects and safety hazards, on the steel structure of the high-speed rail station building roof is constructed, and the model construction method is as follows: first, the collected images are preprocessed, then the key features in the images are extracted, and finally the model uses the above features for fault classification and identification.

5. The method for automatically inspecting diseases of a railway station concealed space steel structure according to claim 1, characterized in that: The automatic inspection system for implementing the automatic inspection method of the hidden space steel structure disease of the railway passenger station comprises a robot system, an integrated control system, a detection device, a power carrier communication system, a video server and a ground monitoring base station.

6. The method for automatically inspecting diseases of a railway station concealed space steel structure according to claim 5, characterized in that: The robot system comprises a control module, a communication transmission module, a positioning and navigation module, a power supply control module, a temperature control module, an obstacle avoidance module and a vision module. The robot system communicates with the ground monitoring base station through the communication transmission module to interact data, makes decisions according to the preset model information and environmental conditions, and realizes high-reliability control operation of the robot; the control module communicates with the ground monitoring base station to obtain control instructions, and controls, executes and completes accurate positioning, autonomous navigation, obstacle avoidance, charging and temperature control functions, and simultaneously analyzes data, judges the state of the equipment and the environment, discovers abnormalities and alarms in time.

7. The method for automatically inspecting diseases of a railway station concealed space steel structure according to claim 1, characterized in that: The power limb mechanism comprises a rotating limb one, a rotating motor one, a rotating joint, a rotating motor two, a rotating motor three, a rotating limb two and a shock absorber, and the structure of the passive limb mechanism is the same as that of the power limb mechanism.

8. The method according to claim 7, wherein the method further comprises the steps of: determining the location of the hidden space of the railway station; and determining the location of the hidden space of the railway station. The folding arm comprises a movable limb, an extension mechanism, a shock absorber and a rotating motor four, the extension mechanism comprises a plurality of long rods and short rods connected in a scissors type, and the folding arm is provided with a driving wheel mechanism at the end.

Images