Belt conveyor roller detection system and detection method using quadruped robot

Abstract

A kind of belt conveyor roller detection system and detection method using quadruped robot, including belt conveyor and the roller being arranged under the conveyor belt of belt conveyor, end signal transmitting / receiving module is installed on the support of belt conveyor below its roller, the module is provided with WIFI router, and the end signal transmitting / receiving module is connected with field control by optical fiber.The outer side direction of the roller of belt conveyor is provided with quadruped robot belt conveyor roller detection assembly, the quadruped robot belt conveyor roller detection assembly moves along the side of belt conveyor by pre-set program, detects each roller along the way, and the abnormal roller state information and the coordinate position where abnormal roller is located detected by wireless mode are fed back to end signal transmitting / receiving module.The abnormal state of the roller of the present application can effectively patrol and inspect the roller, reduce the burden of artificial inspection, effectively reduce the downtime caused by the failure of roller, improve the transport efficiency of belt conveyor, indirectly improve the enterprise capacity.

Description

Technical Field

[0001] This invention relates to the field of intelligent inspection, and in particular to a system and method for detecting the condition of idlers during belt conveyor inspection using a quadruped robot. Background Technology

[0002] Currently, belt conveyors are frequently used as important mechanical equipment for material transportation in fields such as coal mining, steel, petroleum, and chemical industries. Belt conveyors have advantages such as long conveying distance, high speed, large carrying capacity, and low loss. However, due to their relatively complex structure, numerous parts and components, and potentially harsh transportation environments, belt conveyors often experience malfunctions such as belt misalignment, slippage, breakage, insufficient tension (slack), and abnormal noise.

[0003] As the most important mechanism responsible for the operation of belt drive, the idler rollers of the belt conveyor are also the focus of daily troubleshooting and maintenance. Usually, the number of idler rollers of a belt conveyor is three times or more than its length. Therefore, how to safely and effectively detect the operating status of the idler rollers is an urgent problem to be solved.

[0004] Existing technologies include methods to determine the abnormal state of idlers by detecting abnormal sounds. However, after actual field testing, it was found that the actual application situation in the belt conveyor area is quite complex, and there are many devices in the belt conveyor area, resulting in mixed background noise. It is impossible to accurately extract / determine the noise generated by the belt conveyor idlers, thus making the sound detection solution susceptible to interference.

[0005] Existing technologies also include recording videos of belt conveyor rotation and identifying the movement trajectory of marks on idlers in the video to determine the idler speed and thus detect idler abnormalities. However, this method requires high video quality and clarity, and the operating conditions of belt conveyors, such as dusty environments, directly affect the accuracy of the identification. Summary of the Invention

[0006] In summary, to address the problems of numerous idler rollers, difficult inspection, and harsh inspection environments in existing technologies, this invention provides a belt conveyor idler roller inspection system and method using a quadruped robot. Its vision-based inspection method directly acquires idler roller images, is less affected by the environment, and by setting different markers, the inspection is convenient and efficient. At the same time, it can effectively reduce belt conveyor downtime caused by idler roller failures and alleviate the workload of on-site inspection personnel.

[0007] The present invention provides a belt conveyor idler roller detection system and detection method using a quadruped robot, the specific structure and method steps of which are described below:

[0008] A conveyor belt idler detection system using a quadruped robot includes a conveyor belt and idlers installed under the conveyor belt, characterized in that:

[0009] The belt conveyor is equipped with an end signal transmitting / receiving module mounted on the belt conveyor bracket below its idler rollers. This module is equipped with a WIFI router and is connected to the field control center via optical fiber.

[0010] The end signal transmitting / receiving module is connected to the field control center via optical fiber, enabling it to transmit information at a relatively fast speed. This allows the processed information to be transmitted to the field control center so that it can promptly grasp the fault situation.

[0011] A quadruped robot belt conveyor roller detection component is provided on the outer side of the belt conveyor roller. The quadruped robot belt conveyor roller detection component moves along the side of the belt conveyor through a preset program to detect each roller along the way, and communicates with the end signal transmission / reception module wirelessly to feed back the detected abnormal roller status information and the coordinate position of the abnormal roller to the end signal transmission / reception module.

[0012] The idler roller has an idler roller marking module at its end to indicate the idler roller's operating status. Specifically, the idler roller marking module has a vertical strip-shaped groove on the end face of the idler roller shaft. The length, width and depth of the strip-shaped groove are 30mm×10mm×10mm. When the idler roller rotates, the strip-shaped groove forms a white block image on the idler roller. This white block image is collected and processed by the four-legged robot belt conveyor idler roller detection component, and then identified. The identification result is used to determine whether the idler roller is operating normally.

[0013] According to the present invention, a conveyor belt idler detection system using a quadruped robot is characterized in that the quadruped robot conveyor belt idler detection component includes a quadruped robot body, a multi-line laser radar, an image acquisition device, and a signal processing device. The quadruped robot body is a quadruped walking robot with a preset walking path. The top of the quadruped robot body is a platform. A multi-line laser radar is installed at a platform position towards the head of the quadruped robot body to perform 360-degree laser line scanning. An image acquisition device is installed at a platform position towards the tail of the quadruped robot body. A signal processing device is installed at a platform position in the middle of the platform.

[0014] The quadruped robot used in this invention possesses industrial-grade protection and the ability to overcome obstacles, climb stairs, and crawl, adapting to uneven surfaces caused by material accumulation in conveyor belt corridors, as well as harsh environments with high dust and water levels. A multi-line LiDAR system with 360-degree surround laser scanning is used to create a scene map and scan the surrounding environment in real time, enabling the quadruped robot to autonomously locate and navigate. The scene map established by this multi-line LiDAR uses a planar coordinate system with the geometric center of the conveyor belt head as the origin. The coordinates (x, y) represent the real-time position of the quadruped robot's conveyor belt roller detection component, where x is the lateral distance from the quadruped robot's conveyor belt roller detection component to the conveyor belt, +x represents the left side of the conveyor belt from head to tail, and -x represents the right side of the conveyor belt from head to tail. y is the longitudinal distance from the quadruped robot to the conveyor belt head.

[0015] In summary, the quadruped robot walks along the corridor beside the conveyor belt, performing inspection work around the conveyor belt. During inspection, the quadruped robot can be set to travel a single distance. After traveling a certain distance, it stops to collect and inspect images of the idler rollers. The travel distance can be calculated based on the scene map created by the multi-line LiDAR, the arrangement of the idler rollers, and the field of view of the image acquisition device.

[0016] According to the present invention, a conveyor belt idler detection system using a quadruped robot is characterized in that the image acquisition direction of the image acquisition device of the quadruped robot conveyor belt idler detection component faces the shaft end direction of the idler of the conveyor belt, and the image acquisition device selects the image acquisition range according to the distance between the quadruped robot conveyor belt idler detection component and the idler, taking the integer number of idlers that can be seen within its field of view as the standard.

[0017] The image acquisition device focuses on the end of the idler roller shaft to acquire images and transmits the acquired images to the signal processing device. The field of view width can be regarded as a moving window relative to the idler roller. During each inspection, the idler roller is centered within the field of view width (window). After the inspection is completed, the window moves to the subsequent idler rollers until the next set of idler rollers to be tested. The distance moved each time is the single travel distance of the four-legged robot belt conveyor idler roller inspection component during inspection.

[0018] According to the present invention, a conveyor belt idler detection system using a quadruped robot is characterized in that the image acquisition device of the quadruped robot conveyor belt idler detection component is preferably an RGB camera.

[0019] An RGB camera or other type of vision sensor is sufficient to acquire the required image information of the idler rollers.

[0020] According to the present invention, a belt conveyor idler detection system using a quadruped robot is characterized in that the signal processing device (A4) of the quadruped robot belt conveyor idler detection component is equipped with an OpenCV library, which can extract the color information of the idler shaft end image acquired by the image acquisition device, and identify whether the area of ​​the white color block region is greater than a preset threshold. If it is greater than the threshold, the idler is operating normally; otherwise, it is considered that the idler is not operating normally. The status and position of the idler are recorded and sent to the end signal transmission / reception module, and then transmitted to the field control center through the end signal transmission / reception module.

[0021] According to the present invention, a belt conveyor roller detection system using a quadruped robot is characterized in that the roller marking module is provided with a concave groove on the roller surface near the detection position of the quadruped robot belt conveyor roller detection component, the length, width and depth of the concave groove are 20mm×20mm×100mm, and the concave groove is coated with a colored paint.

[0022] According to the present invention, a belt conveyor idler roller detection system using a quadruped robot is characterized in that the concave groove of the idler roller marking module is coated with a colored paint, preferably a white paint, to create a color difference with the shaft end and the roller skin.

[0023] The roller marking module can use either strip grooves or concave grooves. Strip grooves form white block images on the roller when it rotates, while concave grooves use white paint to further improve the recognizability of the white blocks.

[0024] A method for detecting conveyor belt idlers using a quadruped robot, based on the aforementioned conveyor belt idler detection system using a quadruped robot, comprises the following specific steps:

[0025] 1) Set the walking direction and position of the quadruped robot body in the quadruped robot conveyor roller detection component. Specifically, the quadruped robot body walks clockwise around the conveyor belt, starting from the left side of the conveyor belt from the head to the tail, with the starting point of walking located in the middle of the first and second rollers. Confirm that the image acquisition device centers these two rollers within its field of view and detects the rollers. During this process, the multi-line LiDAR performs a 360-degree surround laser line scan to create a scene map and scan the surrounding environment in real time.

[0026] 2) After the quadruped robot body reaches the detection position in step 1), it transmits a working signal to the image acquisition device, so that it starts to acquire the shaft end images of the first and second idler rollers it faces. The acquired images are transmitted to the signal processing device, which extracts the color information of the acquired idler roller shaft end images and identifies whether the area of ​​the white color block region is greater than a preset threshold. If it is greater than the threshold, it is determined that the idler roller is operating normally.

[0027] 3) In step 2), if the signal processing device identifies that the area of ​​the white patch is less than the preset threshold, it is considered that the idler roller is not operating normally. The signal processing device records the coordinate position of the idler roller in the scene map created by the multi-line lidar and the image of the idler roller obtained by the image acquisition device.

[0028] 4) After the first and second rollers are inspected, the quadruped robot continues to move in the direction of the third and fourth rollers until the third and fourth rollers are centered within the field of view of the image acquisition device. After reaching the position, it stops and repeats steps 2) and 3) to inspect the third and fourth rollers until all subsequent rollers are inspected.

[0029] 5) After all the idlers have been inspected, the inspection work of the four-legged robot belt conveyor idler inspection component ends. The signal processing device sends the coordinate position and image of the abnormal idler to the end signal transmission / reception module (4), and then transmits it to the field control through the end signal transmission / reception module. The field control then notifies the maintenance department to handle the fault after understanding the fault situation.

[0030] The following results were obtained using a belt conveyor idler roller detection system and method employing a quadruped robot according to the present invention.

[0031] Beneficial effects:

[0032] 1. The belt conveyor idler detection system and method using a quadruped robot of the present invention can effectively inspect the abnormal condition of the idler, has strong obstacle crossing ability, and is suitable for harsh working conditions and complex road surfaces in belt conveyor areas.

[0033] 2. The belt conveyor roller detection system and method using a quadruped robot of the present invention can reduce the burden of manual inspection and free up labor.

[0034] 3. The belt conveyor idler detection system and method using a quadruped robot of the present invention can effectively reduce downtime caused by idler failure, improve belt conveyor transportation efficiency, and indirectly improve enterprise productivity. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of a partial structure of a belt conveyor roller detection system and detection method using a quadruped robot according to the present invention.

[0036] Figure 2 This is a schematic diagram of the structure of a belt conveyor idler roller detection system and detection method using a quadruped robot according to the present invention, which includes an idler roller marking module (two types).

[0037] Figure 3This is a flowchart illustrating the process of setting up an idler marking module in a belt conveyor idler detection system and detection method using a quadruped robot, as described in this invention.

[0038] In the diagram: 1-Belt conveyor, 2-Idler roller, 2a-Idler roller marking module, 3-Belt conveyor bracket, 4-End signal transmitting / receiving module, A-Quadruped robot belt conveyor idler roller detection component, A1-Quadruped robot body, A1a-Equipment platform, A2-Multi-line lidar, A3-Image acquisition device, A4-Signal processing device. Detailed Implementation

[0039] The technical means, creative features, objectives, and effects of the belt conveyor roller detection system and method using a quadruped robot of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0040] Example

[0041] like Figures 1-2 As shown, a belt conveyor idler detection system using a quadruped robot includes a belt conveyor 1 and idlers 2 installed under the belt conveyor belt. An end signal transmitting / receiving module 4 is installed on a belt conveyor bracket 3 below the idlers. The module is equipped with a WIFI router and is connected to the field control center via optical fiber.

[0042] A quadruped robot belt conveyor roller detection component A is provided on the outer side of the idler roller 2 of the belt conveyor 1. The quadruped robot belt conveyor roller detection component moves along the side of the belt conveyor through a preset program to detect each idler roller along the way, and communicates with the end signal transmitting / receiving module 4 wirelessly to feed back the detected abnormal idler roller status information and the coordinate position of the abnormal idler roller to the end signal transmitting / receiving module.

[0043] The idler roller 2 has an idler roller marking module 2a at its end to indicate the operating status of the idler roller. Specifically, the idler roller marking module has a vertical strip-shaped groove on the end face of the shaft end of the idler roller. The length, width and depth of the strip-shaped groove are 30mm×10mm×10mm. When the idler roller rotates, the strip-shaped groove forms a white block image on the idler roller. This white block image is collected and processed by the four-legged robot belt conveyor idler roller detection component A. After processing, it is identified, and the identification result is used to determine whether the idler roller is operating normally.

[0044] The quadruped robot conveyor roller detection component A includes a quadruped robot body A1, a multi-line lidar A2, an image acquisition device A3, and a signal processing device A4. The quadruped robot body is a quadruped walking robot with a preset walking path. The top of the quadruped robot body is a device platform A1a. A multi-line lidar is set at the platform position in the direction of the head of the quadruped robot body to perform 360-degree laser line scanning. An image acquisition device is set at the platform position in the direction of the tail of the quadruped robot body. The signal processing device is set at the platform position in the middle of the device platform.

[0045] In this embodiment, the dimensions of the quadruped robot body are 1000mm (length) × 460mm (width) × 600mm (height), which are based on the assembly height of the conveyor belt roller.

[0046] The image acquisition device A3 of the quadruped robot belt conveyor idler roller detection component A faces the shaft end direction of the idler roller 2 of the belt conveyor 1. The image acquisition device selects the image acquisition range according to the distance between the quadruped robot belt conveyor idler roller detection component and the idler roller, taking the integer number of idler rollers that can be seen within its field of view as the standard.

[0047] In this embodiment, taking two sets of rollers within the field of view as an example, with a roller spacing of 1000mm and a detection distance of 600mm, a field of view angle of 89° can be selected, resulting in a field of view width of 1200mm. The sliding distance of the field of view window is: number of rollers × roller spacing, i.e., 2 x 1000mm = 2000mm, which is also the single travel distance of the quadruped robot body during inspection.

[0048] The image acquisition device A3 of the quadruped robot belt conveyor roller detection component A is preferably an RGB camera.

[0049] The signal processing device A4 of the quadruped robot belt conveyor idler roller detection component A is equipped with the OpenCV library. It can extract the color information of the idler roller shaft end image acquired by the image acquisition device A3 and identify whether the area of ​​the white color block region is greater than a preset threshold. If it is greater than the threshold, the idler roller is operating normally; otherwise, it is considered that the idler roller is not operating normally. The status and position of the idler roller are recorded and sent to the end signal transmission / reception module 4, which then transmits the signal to the field control center.

[0050] The roller marking module 2a can also be a concave groove opened on the roller surface near the detection position of the roller detection component A of the quadruped robot belt conveyor. The length, width and depth of the concave groove are 20mm×20mm×100mm, and the concave groove is coated with colored paint.

[0051] The concave groove of the roller marking module 2a is coated with colored paint. In this embodiment, the colored paint is white, which is intended to create a color difference with the shaft end and roller skin.

[0052] like Figure 3 As shown, a method for detecting conveyor belt idlers using a quadruped robot, based on the aforementioned conveyor belt idler detection system using a quadruped robot, comprises the following specific steps:

[0053] 1) Set the walking direction and position of the quadruped robot body A1 of the quadruped robot conveyor roller detection component A. Specifically, the quadruped robot body walks clockwise around the conveyor 1, starting from the left side of the conveyor from the head to the tail, with the starting point of walking located in the middle of the first and second rollers 2 (in this embodiment, the distance between the first and second rollers is 1000mm). Confirm that the image acquisition device A3 centers the two rollers within its field of view and detects the rollers. During this process, the multi-line lidar A2 performs a 360-degree surround laser line scan to create a scene map and scan the surrounding environment in real time.

[0054] 2) After the quadruped robot body reaches the detection position in step 1), it transmits a working signal to the image acquisition device, so that it starts to acquire the shaft end images of the first and second idler rollers it faces. The acquired images are transmitted to the signal processing device (A4), which extracts the color information of the acquired idler roller shaft end images and identifies whether the area of ​​the white color block region is greater than a preset threshold. If it is greater than the threshold, it is determined that the idler roller is operating normally.

[0055] 3) In step 2), if the signal processing device identifies that the area of ​​the white patch is less than the preset threshold, it is considered that the idler roller is not operating normally. The signal processing device records the coordinate position of the idler roller in the scene map created by the multi-line lidar and the image of the idler roller obtained by the image acquisition device.

[0056] 4) After the first and second rollers are inspected, the quadruped robot continues to move in the direction of the third and fourth rollers until the third and fourth rollers are centered within the field of view of the image acquisition device (in this embodiment, the distance traveled in one go is 2000mm, that is, it stops exactly in the middle of the third and fourth rollers, so that it is centered within the field of view of the image acquisition device). After reaching the position, it stops and repeats steps 2) and 3) to inspect the third and fourth rollers until all subsequent rollers are inspected.

[0057] 5) After all the idlers have been inspected, the inspection work of the four-legged robot belt conveyor idler inspection component ends. The signal processing device sends the coordinate position and image of the abnormal idler to the end signal transmission / reception module (4), and then transmits it to the field control through the end signal transmission / reception module. The field control then notifies the maintenance department to handle the fault after understanding the fault situation.

[0058] The present invention provides a belt conveyor idler detection system and method using a quadruped robot, which can effectively inspect the abnormal condition of the idlers, has strong obstacle-crossing ability, and is suitable for harsh working conditions and complex road surfaces in belt conveyor areas. The present invention can reduce the burden of manual inspection and free up labor. The present invention can effectively reduce downtime caused by idler failure, improve belt conveyor transportation efficiency, and indirectly improve enterprise productivity.

[0059] However, those skilled in the art should recognize that the above embodiments are only used to illustrate this application and are not intended to limit this application. Any changes or modifications to the above embodiments within the spirit and essence of this application will fall within the scope of the claims of this application.

Claims

1. A belt conveyor idler detection system using a quadruped robot, comprising a belt conveyor (1) and idlers (2) disposed under the belt conveyor belt, characterized in that: The belt conveyor (1) is equipped with an end signal transmitting / receiving module (4) on the belt conveyor bracket (3) below its idler roller (2). The module is equipped with a WIFI router and the end signal transmitting / receiving module is connected to the field control center via optical fiber. The belt conveyor (1) is provided with a quadruped robot belt conveyor roller detection component (A) on the outer side of the idler roller (2). The quadruped robot belt conveyor roller detection component moves along the side of the belt conveyor through a preset program to detect each idler roller along the way, and communicates with the end signal transmitting / receiving module (4) wirelessly to feed back the detected abnormal idler roller status information and the coordinate position of the abnormal idler roller to the end signal transmitting / receiving module. The idler roller (2) has an idler roller marking module (2a) at its end to indicate the operating status of the idler roller. Specifically, the idler roller marking module is a vertical strip groove on the end face of the shaft end of the idler roller. The length, width and depth of the strip groove are 30mm×10mm×10mm. When the idler roller rotates, the strip groove forms a white block image on the idler roller. The white block image is collected and processed by the quadruped robot belt conveyor idler roller detection component (A). After processing, it is identified, and the identification result is used to determine whether the idler roller is operating normally. The quadruped robot conveyor roller detection component (A) includes a quadruped robot body (A1), a multi-line lidar (A2), an image acquisition device (A3), and a signal processing device (A4). The quadruped robot body is a quadruped walking robot with a preset walking path. The top of the quadruped robot body is a device platform (A1a). A multi-line lidar is set at the platform position in the direction of the head of the quadruped robot body to perform 360-degree laser line scanning. An image acquisition device is set at the platform position in the direction of the tail of the quadruped robot body. A signal processing device is set at the platform position in the middle of the device platform. The image acquisition direction of the image acquisition device (A3) of the quadruped robot belt conveyor roller detection component (A) is facing the shaft end direction of the roller (2) of the belt conveyor (1). The image acquisition device selects the image acquisition range according to the distance between the quadruped robot belt conveyor roller detection component and the roller, and takes the number of rollers that can be seen within its field of view as the standard. The signal processing device (A4) of the quadruped robot belt conveyor roller detection component (A) is equipped with the OpenCV library, which can extract the color information of the roller shaft end image collected by the image acquisition device (A3) and identify whether the area of ​​the white color block region is greater than a preset threshold. If it is greater than the threshold, the roller is judged to be operating normally; otherwise, it is considered to be operating abnormally. The status and position of the roller are recorded and sent to the end signal transmission / reception module (4), and then transmitted to the field control center through the end signal transmission / reception module.

2. The belt conveyor roller detection system using a quadruped robot as described in claim 1, characterized in that, The image acquisition device (A3) of the quadruped robot belt conveyor roller detection component (A) is an RGB camera.

3. A method for detecting conveyor belt idlers using a quadruped robot, based on any of the preceding claims, comprising the following specific steps: 1) Set the walking direction and position of the quadruped robot body (A1) of the quadruped robot conveyor roller detection component (A). Specifically, the quadruped robot body walks clockwise around the conveyor (1). Starting from the left side of the conveyor from the head to the tail, the starting point of the walk is located in the middle of the first and second rollers. Confirm that the image acquisition device (A3) centers the two rollers within its field of view and detects the rollers. During this process, the multi-line laser radar (A2) performs a 360-degree surround laser line scan to create a scene map and scan the surrounding environment in real time. 2) After the quadruped robot body reaches the detection position in step 1), it transmits a working signal to the image acquisition device, so that it begins to acquire the shaft end images of the first and second idler rollers it faces. The acquired images are transmitted to the signal processing device (A4), which extracts the color information of the acquired idler roller shaft end images and identifies whether the area of ​​the white color block region is greater than a preset threshold. If it is greater than the threshold, it is determined that the idler roller is operating normally. 3) In step 2), if the signal processing device identifies that the area of ​​the white patch is less than the preset threshold, it is considered that the idler roller is not operating normally. The signal processing device records the coordinate position of the idler roller in the scene map created by the multi-line lidar and the image of the idler roller obtained by the image acquisition device. 4) After the first and second rollers are inspected, the quadruped robot continues to move in the direction of the third and fourth rollers until the third and fourth rollers are centered within the field of view of the image acquisition device. After reaching the position, it stops and repeats steps 2) and 3) to inspect the third and fourth rollers until all subsequent rollers are inspected. 5) After all the idlers have been inspected, the inspection work of the four-legged robot belt conveyor idler inspection component ends. The signal processing device sends the coordinate position and image of the abnormal idler to the end signal transmission / reception module (4), and then transmits it to the field control through the end signal transmission / reception module. The field control then notifies the maintenance department to handle the fault after understanding the fault situation.

Images