Rail surface damage detection robot based on machine vision

Abstract

The utility model discloses a rail surface damage detection robot based on machine vision, which belongs to the technical field of rail damage detection and comprises a shell and a traveling mechanism, and a camera on the front side of the shell can collect images of a rail head at the top of a rail; a display screen is arranged on the outer surface of the shell, a controller and a battery are arranged in the shell, and the camera, the walking mechanism and the display screen are all connected with the controller and the battery. A rail head of a rail is wrapped with a shell of a semi-surrounding structure, a walking mechanism can run along the rail head, and a camera is used for collecting images of the top face of the rail and transmitting the images to a display screen; the controller controls starting and stopping of the walking mechanism and analyzes the damage condition of the rail. The rail crack detector is simple in structure, low in cost and convenient and rapid to operate, can realize rapid, accurate and low-cost detection of rail cracks, effectively improves the efficiency and accuracy of rail detection, ensures the smoothness and safety of railway transportation, and provides a solid technical guarantee for safe operation of a railway transportation system in China.

Description

Technical Field

[0001] This utility model belongs to the field of track damage detection technology, and specifically relates to a robot for detecting track surface damage based on machine vision. Background Technology

[0002] In my country's railway transportation system, the identification and detection of rail cracks is crucial, directly impacting railway safety and operational efficiency. With the rapid expansion of my country's railway network, especially the rapid development of high-speed and heavy-haul railways, the environmental and load conditions faced by rails are becoming increasingly complex, leading to a year-on-year increase in the incidence of crack defects. Currently, there are numerous types of rail cracks, including vertical cracks, transverse cracks, oblique cracks, and fishtail cracks, among others. These cracks exhibit complex morphologies, ranging from visible macroscopic cracks to microscopic cracks hidden within the material. Under the combined effects of train loads, temperature stress, and environmental corrosion, these cracks can trigger track fractures, seriously threatening railway transportation safety.

[0003] While various methods exist for detecting rail cracks, each has its advantages and disadvantages. Manual inspection relies on the experience of inspectors, and although it can detect some complex crack morphologies, it is inefficient, with each person only covering an average of 3-5 kilometers per day. Furthermore, it is greatly affected by environmental factors such as weather and lighting, with poor results at night and in inclement weather. Electromagnetic detection equipment, while highly sensitive, is prone to signal aliasing when processing complex crack morphologies, resulting in a false alarm rate as high as 15%-20%. Laser scanning technology, while capable of acquiring three-dimensional morphological data with millimeter-level precision, is expensive, with a single system costing over one million yuan, and data processing takes several hours, making it difficult to meet the demands of modern railways for "real-time monitoring and immediate maintenance." Utility Model Content

[0004] To address the above problems, this invention provides a robot for detecting damage to railway tracks based on machine vision.

[0005] To achieve the above objectives, the technical solution adopted by this utility model is as follows:

[0006] A machine vision-based robot for detecting surface damage on railway tracks includes a housing capable of enclosing the rail head and a walking mechanism for driving the housing. A camera is mounted on the front of the housing to capture images of the rail head. A display screen is mounted on the outer surface of the housing. The walking mechanism can travel along the rail head. A controller and a battery are located inside the housing. The camera, walking mechanism, and display screen are all connected to the controller and battery. The battery provides power to the electrical components, and the controller can control the movement of the walking mechanism and store the image data from the camera.

[0007] Furthermore, the walking mechanism includes a motor and a drive wheel. The drive wheel is connected to the motor through a transmission assembly. The drive wheel is located inside the housing and can travel along the top surface of the rail head. The motor is located on the top surface of the housing.

[0008] Furthermore, the housing is a semi-enclosed structure, including a top plate and side plates on both sides, with an opening on the top plate through which the first synchronous belt of the transmission assembly passes.

[0009] Furthermore, the transmission assembly includes a driving pulley, a first driven pulley, a transmission shaft, a second driven pulley, and a third driven pulley. The driving pulley is coaxially fixed with the output shaft of the motor. The driving pulley is connected to the first driven pulley via a first synchronous belt. Both the first and second driven pulleys are mounted on the transmission shaft. The second driven pulley is connected to the third driven pulley via a second synchronous belt. The third driven pulley and the drive wheel are both mounted on the drive shaft. Both ends of the transmission shaft and the drive shaft are connected to the housing via bearing seats.

[0010] Furthermore, the traveling mechanism also includes a driven wheel, which is connected to the housing via a connecting frame; the driving wheel and the driven wheel are spaced apart and can both travel along the top surface of the rail head.

[0011] Furthermore, the inner bottom of the housing is provided with wheel sets, which are two sets and symmetrically arranged on both sides of the rail head; both wheel sets are connected to the side plates of the housing through support frames.

[0012] Furthermore, the wheel assembly on one side of the rail head is driven by a servo motor, which is located inside the housing and connected to the side plate; a support plate is provided below the wheel assembly on the other side, with one end of the support plate connected to the side plate and the other end extending to the bottom surface of the rail head.

[0013] Furthermore, the end of the support plate is provided with a support wheel, which can abut against the bottom surface of the rail head.

[0014] Furthermore, the controller is equipped with a GPS module and a wireless connection module, which are used to upload the detected track damage location information to the central control room.

[0015] The technological advancements achieved by this invention compared to existing technologies are as follows:

[0016] This invention utilizes a semi-enclosed shell structure to enclose the rail head. Driven by a traveling mechanism, the shell moves along the rail head. A camera on the front of the shell captures images of the top of the rail head, which are then displayed on a screen on the shell surface for direct observation. A controller manages the start and stop of the traveling mechanism and analyzes rail damage. This invention features a simple structure, low cost, and convenient and quick operation, greatly reducing the labor intensity of workers. It enables rapid, accurate, and low-cost detection of rail cracks, effectively improving the efficiency and accuracy of rail inspection, ensuring smooth and safe railway transportation, and providing a solid technical guarantee for the safe operation of my country's railway transportation system. Attached Figure Description

[0017] The accompanying drawings are provided to further understand the present invention and form part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation thereof.

[0018] In the attached diagram:

[0019] Figure 1 A schematic diagram of the structure of a machine vision-based rail surface damage detection robot provided for an embodiment of this utility model;

[0020] Figure 2 for Figure 1 Front view of a machine vision-based rail surface damage detection robot;

[0021] Figure 3 for Figure 1 A schematic diagram of the structure of a machine vision-based rail surface damage detection robot after the top plate has been removed.

[0022] In the picture:

[0023] 00-Railway, 01-Railhead; 1-Shell, 11-Top Plate, 12-Side Plate; 2-Camera; 3-Display Screen; 4-Motor; 5-Drive Wheel; 6-First Synchronous Belt; 7-Driving Pulley; 8-First Driven Pulley; 9-Drive Shaft; 10-Second Driven Pulley; 13-Third Driven Pulley; 14-Bearing Housing; 15-Second Synchronous Belt; 16-Driven Wheel; 17-Wheelset; 18-Servo Motor; 19-Support Plate; 20-Support Wheel; 21-Drive Shaft; 22-Motor Mount. Detailed Implementation

[0024] The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this utility model will be described below with reference to the accompanying drawings.

[0025] like Figure 1 , Figure 2 and Figure 3 As shown, a machine vision-based robot for detecting rail surface damage includes a housing 1 capable of enclosing the rail head 01 of rail 00 and a walking mechanism for driving the housing 1. A camera 2 is mounted on the front of the housing 1 to capture images of the rail head 01. A display screen 3 is mounted on the outer surface of the housing 1. The walking mechanism can travel along the rail head 01 of rail 00. A controller and a battery are housed inside the housing 1. The camera 2, walking mechanism, and display screen 3 are all connected to the controller and battery. The battery provides power to the electrical components. The camera captures images of the rail head and displays them on the display screen for easy observation. The controller controls the movement of the walking mechanism, stores the camera image data, and analyzes the damage to the rail.

[0026] As a preferred structure, such as Figure 1 , 2 As shown, the traveling mechanism includes a motor 4 and a drive wheel 5. The drive wheel 5 is connected to the motor 4 via a transmission assembly. The drive wheel 5 is located inside the housing 1 and can travel along the top surface of the rail head 01 of the rail 00. The motor 4 is fixed to the top surface of the housing 1 via a motor mount 22. The housing 1 has a semi-enclosed structure, including a top plate 11 and side plates 12 on both sides, ensuring the balance and operation of the detection trolley on the rail. The top plate 11 has an opening through which the first synchronous belt 6 of the transmission assembly passes to facilitate driving the drive wheel inside the housing.

[0027] In specific embodiments of this utility model, such as Figure 3 As shown, the transmission assembly includes a driving pulley 7, a first driven pulley 8, a transmission shaft 9, a second driven pulley 10, and a third driven pulley 13. The driving pulley 7 is coaxially fixed with the output shaft of the motor 4. The driving pulley 7 is connected to the first driven pulley 8 via a first synchronous belt 6. Both the first driven pulley 8 and the second driven pulley 10 are mounted on the transmission shaft 9. The second driven pulley 10 is connected to the third driven pulley 13 via a second synchronous belt 15. The third driven pulley 13 and the drive wheel 5 are both mounted on the drive shaft 21. Both ends of the transmission shaft 9 and the drive shaft 21 are connected to the housing 1 via bearing seats 14. With this structure, the driving pulley can be driven to rotate by the motor, which in turn drives the transmission shaft to rotate via the first synchronous belt. The transmission shaft then drives the drive wheel to rotate via the second synchronous belt, thereby achieving the purpose of traveling along the top surface of the rail.

[0028] Further optimize the above solution, such as Figure 3As shown, the traveling mechanism also includes a driven wheel 16, which is connected to the top plate 11 of the housing 1 via a connecting frame; the driving wheel 5 and the driven wheel 16 are spaced apart and can both travel along the top surface of the rail head 01 of the rail 11. The driving wheel and the driven wheel can improve the stability of the detection trolley during operation.

[0029] In specific embodiments of this utility model, such as Figure 2 As shown, the inner bottom of the housing 1 is provided with wheel sets 17. There are two sets of wheel sets 17, symmetrically arranged on both sides of the rail head 01. Both wheel sets 17 are connected to the side plates 12 of the housing via support frames. Each wheel set consists of four directional wheels, arranged in pairs, and each pair is connected to the side plate via a wheel frame. The wheel sets on both sides provide balanced support for the housing, and the coordination of the drive wheels and wheel sets further improves the stability of the inspection trolley's operation.

[0030] To further optimize the above structure, the wheel set 17 on one side of the rail head 01 is driven by a servo motor 18, which is located inside the housing 1 and connected to the side plate 12. A support plate 19 is located below the wheel set 17 on the other side, with one end connected to the side plate 12 and the other end extending to the bottom surface of the rail head 01. The servo motor is started and stopped by a controller, which improves the power of the detection trolley during operation. In specific assembly, the motor is mounted on the top plate above the support plate, and the motor and servo motor are installed on the upper and lower sides of the housing respectively. The support plate can correct the attitude of the housing, preventing tilting due to uneven weight distribution on both sides.

[0031] In specific manufacturing, a support wheel 20 is provided at the end of the support plate 19, and the support wheel 20 can abut against the bottom surface of the rail head 01. The support wheel can reduce resistance during the correction process. The support wheel is made of nylon, polyurethane or ABS plastic. With its good wear resistance and low coefficient of friction, the continuous contact between the support wheel and the rail head during the operation of the trolley can reduce friction and extend its service life.

[0032] To further optimize the above solution, the controller is equipped with a GPS module and a wireless connection module, used to upload the location information of detected rail damage to the central control room. During operation along the length of the rail, when the camera detects a crack on the rail, it locates the crack using the GPS module and uploads the information in real time to the central control room or a web interface via the wireless connection module. Additionally, to facilitate wireless starting of the inspection trolley, an NFC module can be installed in the controller, enabling NFC-based starting of the inspection trolley for easier operation.

[0033] The application process of this utility model is as follows:

[0034] After the motor starts, the camera will activate, and the "hugging" inspection trolley will continuously acquire video streams of the rail surface as it moves along the track. The controller will continuously analyze the data, obtaining information on the type of rail damage, and then return the results to the lower-level computer's display screen. The number following the corresponding damage type on the display screen will increase to indicate the damage. Simultaneously, the GPS coordinates of the damage point will be uploaded to the central control room or a web interface for subsequent manual inspection.

[0035] In summary, this utility model has the advantages of simple and compact structure and high degree of automation. As the inspection trolley runs along the rails, it uses a camera to collect images of the rail head. When damage is detected, the images are uploaded in real time to the central control room or a web interface via a GPS module and a wireless connection module, facilitating real-time monitoring of rail damage and subsequent maintenance. Using a machine vision-based inspection trolley to detect rail damage reduces labor costs and enables rapid, accurate, and low-cost detection of rail cracks, effectively improving the efficiency and accuracy of rail inspection. It minimizes resource consumption, achieves efficient inspection, ensures smooth and safe railway transportation, and provides a solid technical guarantee for the safe operation of my country's railway transportation system.

[0036] Finally, it should be noted that the above description is merely a preferred embodiment of this utility model and is not intended to limit the utility model. Although the utility model has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this utility model should be included within the scope of protection of the claims of this utility model.

Claims

1. A machine vision-based rail surface defect detection robot, characterized by: The shell can wrap the rail head, the front side of the shell is provided with a camera for collecting images of the rail head, the outer surface of the shell is provided with a display screen, the walking mechanism can travel along the rail head, the inside of the shell is provided with a controller and a battery, and the camera, the walking mechanism and the display screen are connected with the controller and the battery.

2. The machine vision-based rail surface damage detection robot according to claim 1, wherein: The walking mechanism comprises a motor and a driving wheel, the driving wheel is connected with the motor through a transmission assembly, the driving wheel is arranged inside the shell and can travel along the top surface of the rail head, and the motor is arranged on the top surface of the shell.

3. The machine vision-based rail surface damage detection robot according to claim 2, wherein: The shell is a semi-enclosed structure comprising a top plate and side plates on both sides of the top plate, the top plate is provided with an opening, and the first synchronous belt of the transmission assembly is arranged through the opening.

4. The machine vision-based rail surface damage detection robot according to claim 3, wherein: The transmission assembly comprises a driving pulley, a first driven pulley, a transmission shaft, a second driven pulley and a third driven pulley, the driving pulley is coaxially fixed with the output shaft of the motor, the driving pulley is connected with the first driven pulley through the first synchronous belt, the first driven pulley and the second driven pulley are arranged on the transmission shaft, the second driven pulley is connected with the third driven pulley through the second synchronous belt, the third driven pulley and the driving wheel are arranged on the driving shaft, and the two ends of the transmission shaft and the driving shaft are connected with the shell through bearing seats.

5. The machine vision-based rail surface damage detection robot according to claim 2, wherein: The walking mechanism further comprises a driven wheel, the driven wheel is connected with the shell through a connecting frame, the driving wheel and the driven wheel are arranged in front of and behind each other and can travel along the top surface of the rail head.

6. The machine vision-based rail surface damage detection robot according to claim 3, wherein: The inside bottom of the shell is provided with a wheel set, the wheel set is two sets and is symmetrically arranged on both sides of the rail head, and the two wheel sets are connected with the side plates of the shell through support frames.

7. The machine vision-based rail surface damage detection robot according to claim 6, wherein: The wheel set on one side of the rail head is driven by a rudder, the rudder is arranged inside the shell and connected with the side plate, and the lower side of the wheel set on the other side is provided with a support plate, one end of the support plate is connected with the side plate, and the other end extends to the bottom surface of the rail head.

8. The machine vision-based rail surface damage detection robot according to claim 7, wherein: The end of the support plate is provided with a support wheel, and the support wheel can abut against the bottom surface of the rail head.

9. A machine vision-based rail surface damage detection robot according to any one of claims 1-8, characterized by: The controller is provided with a GPS module and a wireless connection module for uploading the detected rail damage position information to a central control room.

Images