Method and device for automatically identifying and grabbing foreign matters and large blocks on coal mine adhesive tape

Abstract

The invention relates to an automatic identifying and grabbing method and device for coal mine adhesive tape foreign matter and large blocks, and belongs to the field of coal mine safety. The method comprises the following steps of: respectively mounting a group of monitoring control devices at the front end, the middle and the rear end of a head rubber belt conveyor after coming out of a mining working surface, so as to realize three-stage accurate capturing and grabbing of foreign matters and large blocks; a high-definition camera, a video acquisition control host and an explosion-proof and intrinsic safety type power supply are mounted in the middle of the middle rubber belt conveyor, so that verification of a three-level grabbing effect is realized; a monitoring control device is installed in the middle of the tail rubber belt conveyor, and it is confirmed that foreign matter and large materials are grabbed by comparing a field foreign matter image with a system foreign matter image. The recognition and grabbing precision is improved, foreign matter and large materials are precisely recognized and grabbed, it is ensured that no foreign matter or large materials finally flow into a coal bunker and a coal leakage hole, and the occurrence of the events that the coal bunker and the coal leakage hole are blocked by the large materials, and an adhesive tape is scratched by hard foreign matter is avoided.

Description

Technical Field

[0001] The invention belongs to the field of coal mine safety and relates to a method and a device for automatically identifying and grabbing foreign matter and bulk objects on a coal mine tape. Background Art

[0002] During the coal mining process, coal blocks, gangue of irregular shapes and sizes, and objects such as anchor rods, anchor cables, and wood blocks used for support will be transported to the ground coal bunker through multiple belt conveyors or coal holes. If foreign objects or large pieces of materials cannot be cleaned up in time on the front belt conveyor, once they enter the coal hole, they will block the coal leakage and affect the transportation efficiency. At the same time, there is a certain risk of bunker collapse when dealing with the coal hole. When transporting at a large angle, large pieces of materials will roll down and injure people. Hard coal gangue, cut anchor rods, anchor cables and other materials will scratch the transport belt during transportation, causing the belt to tear. Large pieces of coal and coal gangue get stuck on the edge of the drum, causing the transport belt to slip, and the friction between the belt and the drum generates heat, resulting in a fire risk. Therefore, it is necessary to clean up various foreign objects or large pieces of materials during the first belt conveyor transportation to minimize or avoid foreign objects from entering the next belt or coal hole, so as to ensure efficient and safe transportation. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a method and device for automatically identifying and grasping foreign objects and large objects on coal mine belts based on AI video, so as to realize the accurate automatic identification and grasping of foreign objects and large objects on belt conveyors, and effectively avoid the situations in which foreign objects and large objects scratch the conveying belt, jam the belt, block coal leaks, and large pieces of materials roll down and injure people, thereby ensuring the safe transportation of coal in mines.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] An automatic identification and grabbing device for foreign matter and large pieces of coal mine tape mainly includes high-definition camera, video acquisition control host, mechanical arm catcher, flameproof and intrinsically safe power supply host and video server and other equipment. Among them, the high-definition camera, video acquisition control host, mechanical arm catcher and flameproof and intrinsically safe power supply host constitute a group of monitoring and control devices. The mechanical arm catcher includes a mechanical arm, a drive device, an electric control, a guide device and an end effector. The video acquisition control host mainly realizes video signal acquisition, data analysis and processing, logical judgment, control output and the like. The video server mainly stores the collected video content for a long time, gives voice alarm prompts when abnormal situations occur, captures images of foreign matter and large pieces of objects, and communicates with the video acquisition control host through the network, provides measurement point parameter definition and data query and other main functions. The high-definition camera collects front-end video data in real time, and uploads the collected video data to the video acquisition control host in real time. The mechanical arm catcher receives control instructions from the video acquisition control host to grab foreign matter and large pieces of materials on the tape. The flameproof and intrinsically safe power supply mainly provides DC intrinsically safe power for the video acquisition control host and the high-definition camera.

[0006] A set of monitoring and control devices are installed at appropriate positions at the front, middle and rear ends of the first belt conveyor after the mining face to achieve three-level precise capture and grabbing of foreign objects and large blocks;

[0007] A high-definition camera, a video acquisition control host, and a flameproof and intrinsically safe power supply host are installed at a suitable position in the middle of the central belt conveyor to verify the three-level grabbing effect; when there is insufficient grabbing, missed grabbing, or grabbing failure, the ground video server captures the image of foreign objects and large pieces of materials, immediately issues a voice alarm prompt, records the alarm information, and sends the captured image to the video acquisition control host of the rear belt conveyor;

[0008] A set of monitoring and control devices is installed at a suitable position in the middle of the tail belt conveyor. The video acquisition control host receives the foreign body image from the video server and compares it with the foreign body image on site. When the similarity of the shape, size, edge and contour of the foreign body is above 90%, the video acquisition control host issues a control command on the spot, and the robotic arm capturer starts to grab the foreign body. When the foreign body is successfully grabbed, the video acquisition control host uploads the captured video image to the ground video server and compares it with the previous foreign body image. The system realizes the grabbing and confirmation of foreign bodies and large pieces of materials, ensuring that there are no foreign bodies and large pieces of materials in the coal bunker and coal leakage eyes, thereby ensuring the safety, smoothness and efficiency of the mine in the entire transportation link.

[0009] Furthermore, the automatic identification and grasping method of the device mainly includes two parts: accurate identification of hard foreign objects and large objects on the front tape and automatic and rapid capture and grasping of hard foreign objects and large objects at the rear tape. In order to ensure that foreign objects or large objects on the tape can be accurately identified and grasped, a three-step method of "three-level capture method + effect verification method + comparison confirmation method" is adopted, which specifically includes the following steps:

[0010] S1: At the first belt conveyor, a three-level capture method is adopted to capture foreign objects and large pieces of materials on the belt conveyor;

[0011] S2: At the middle belt conveyor, the system identifies and verifies the grabbing effect of the first three levels of foreign objects and large objects;

[0012] S3: At the tail belt conveyor, by comparing the on-site foreign object image with the system foreign object image, it is confirmed that foreign objects and large pieces of materials have been captured.

[0013] Further, step S1 specifically includes: installing a high-definition camera at a suitable position at the front end of the first belt conveyor to detect the coal flow, foreign matter and bulk materials on the belt, using on-site fill light, system brightness adjustment, wide dynamic and image enhancement technology to pick up high-quality dynamic images, and uploading the video signal to the video acquisition control host through the network. The video acquisition control host has a built-in Faster R-CNN's foreign body recognition technology uses key features such as the edge, contour, texture, size and dimensions of an object as the basis for judging large pieces of coal, large pieces of gangue, anchor rods, anchor cables, wood piles and other foreign bodies and large pieces of materials. It establishes models such as random forests or CNN to classify and identify the extracted features. When foreign bodies or large pieces of materials are identified, the video acquisition control host outputs control instructions on the spot. The control port of the video acquisition control host and the electronic control of the robotic arm capturer are connected by a cable. After the robotic arm capturer receives the control instruction, the electronic control drives the motor to the guide device, joints and end effector. The robotic arm reaches out to the foreign bodies or large pieces of materials on the tape, and in accordance with the predetermined direction and action, the foreign bodies or large pieces of materials are grabbed and placed in the lane material basket, realizing the first-level grabbing of foreign bodies. After the grabbing is completed, the robotic arm capturer automatically resets and waits for the next instruction.

[0014] Coal mining or excavation is carried out continuously, and the belt conveyor runs continuously. Foreign objects or large pieces of materials may appear continuously. Therefore, there is a situation where the front-end foreign objects or large pieces of materials are caught and failed or missed. A set of monitoring and control devices must be installed at a suitable position in the middle of the first belt conveyor. The recognition and control principles of foreign objects and large pieces of materials are the same as those of the front-end recognition. When the video acquisition control host recognizes that there are foreign objects or large pieces of materials, it outputs control instructions on the spot. After the mechanical arm capturer receives the control instructions, it controls the mechanical arm to reach the foreign objects or large pieces of materials on the belt, grab the foreign objects or large pieces of materials and place them in the roadway material basket, so as to realize the second-level foreign object grabbing;

[0015] After identifying and grabbing foreign objects or large pieces of materials on the belt conveyor twice at the front and middle, a set of monitoring and control devices is installed at a suitable position at the rear end of the first belt conveyor to identify and grab foreign objects or large pieces of materials on the belt from different angles. The identification and control principles of foreign objects and large pieces of materials are the same as those of the front-end identification. When the video acquisition control host identifies the presence of foreign objects or large pieces of materials, it outputs control instructions on the spot. After the robotic arm capturer receives the control instructions, it controls the robotic arm to extend to the foreign objects or large pieces of materials on the belt, grabs the foreign objects or large pieces of materials and places them in the material basket of the aisle, thereby realizing the grabbing of the third-level foreign objects.

[0016] Further, step S2 specifically includes: when the coal flow flows from the first belt conveyor to the middle belt conveyor, the structure, direction and size of the coal flow on the belt change to a certain extent. In order to verify whether the coal flow and foreign objects on the belt are cleanly grabbed, two high-definition cameras, a video acquisition control host and a flameproof and intrinsically safe power supply host are installed at a suitable position in the middle of the middle belt conveyor from different angles. The video acquisition control host and the ground video server adopt network transmission, and the video acquisition control host uploads the collected video signal to the ground video server in real time. It is transmitted to the ground video server. When the coal flow passes through the middle belt conveyor, if there are no foreign objects or large pieces of materials, it means that the first three levels of grasping have achieved the desired effect. If foreign objects or large pieces of materials are identified on the belt conveyor, it means that the first three levels of grasping have not achieved the desired effect. The system immediately issues a voice alarm prompt, captures the image of the foreign object, records the alarm information, and immediately sends the captured image of the foreign object to the video acquisition control host of the rear belt conveyor, waiting for the robotic arm capturer of the rear belt conveyor to grasp it, thereby realizing the front-end grasping effect verification and grasping failure alarm.

[0017] Further, step S3 specifically includes: installing two high-definition cameras, a video acquisition control host, a mechanical arm capturer, and a flameproof and intrinsically safe power supply host at a suitable position in the middle of the tail belt conveyor, identifying foreign objects or large pieces of materials on the belt, and comparing them with the foreign object images received from the ground video server. When the similarity between the foreign objects and large pieces of materials identified by the front-end high-definition camera and the foreign objects and large pieces of materials in the image is more than 90%, the video acquisition control host outputs a control command on the spot, and the mechanical arm capturer receives the control command and controls the foreign objects and large pieces of materials on the belt. The robotic arm reaches out to the foreign objects or large pieces of materials on the belt, grabs them and places them in the material basket in the roadway. The high-definition camera of the rear belt conveyor confirms the image effect captured by the high-definition camera of the middle belt conveyor, and compares it with the foreign object image captured by the high-definition camera of the middle belt conveyor. If the two images are inconsistent, it means that the grabbing is successful, and the voice alarm prompt ends to confirm that the grabbing is successful, thus realizing the grabbing confirmation of foreign objects and large pieces of materials, ensuring that there are no foreign objects and large pieces of materials in the coal bunker and coal leakage eyes, thus ensuring the safety of the rectification and transportation links.

[0018] Furthermore, in order to accurately capture foreign objects and bulk materials, it is also necessary to determine the installation distance between the high-definition camera and the robotic arm capturer. The installation distance is determined according to the running speed of the belt conveyor, the inspection cycle and the action time of the robotic arm capturer.

[0019] S=TV

[0020] T=T 1 +T 2

[0021] Wherein, S represents the installation distance, m; V represents the running speed of the belt conveyor, m / s; T represents the action time, s;

[0022] T 1 Indicates the inspection cycle of the video acquisition control host, ms; T 2 Represents the action time of the robot capturer, s.

[0023] The inspection cycle and the action time of the robotic arm capturer can be measured by a professional oscilloscope. The running speed of the belt conveyor can be determined by calculating the belt roller diameter and the roller transmission speed. Since the belt width, roller diameter, motor power and running speed used in each pair of mines are inconsistent, the installation distances are somewhat different.

[0024] The beneficial effects of the present invention are:

[0025] 1) The present invention achieves "three-level grabbing" of foreign objects or bulk materials on the first belt conveyor by installing high-definition cameras, video acquisition control host, robotic arm capturer and other devices at appropriate positions on the front, middle and rear ends of the first belt conveyor, thereby solving the problem of limited robotic arm running speed and insufficient grabbing due to the high running speed of the belt conveyor and the continuous concentration of foreign objects and bulk materials.

[0026] 2) When the coal flows from the first belt conveyor to the middle belt conveyor, the structure and direction of the coal flow on the belt change. The present invention installs two high-definition cameras, a video acquisition control host and other devices from different angles in the middle of the middle belt conveyor to deeply identify foreign objects and verify the effect of the "three-level capture" at the front end, thereby solving the problem of foreign objects being buried under the coal and not being easily identified and missed.

[0027] 3) The present invention completes the acquisition, analysis, processing, and logical judgment of video signals through the on-site video acquisition control host, and realizes "on-site control and forward control", which solves the problems of long system inspection cycle, long control time, and the system's inability to issue control instructions during network failure.

[0028] 4) The present invention captures images of foreign objects identified through system judgment, and sends the captured images to the terminal video acquisition control host. The video acquisition control host receives the foreign object images from the server and compares them with the on-site foreign object images. When the shape, size, edge and outline of the foreign object are consistent, the foreign object is captured and the captured image is uploaded to the server for comparison and confirmation with the system foreign object image. Through two-level recognition of on-site and system software, the recognition and capture accuracy of system foreign objects and bulk materials is improved.

[0029] 5) The present invention improves the accuracy of identification and grasping by adopting a three-step method of "three-level capture method + effect verification method + comparison confirmation method", realizes accurate identification and grasping of foreign objects and large pieces of materials, ensures that there are no foreign objects and large pieces of materials that finally flow into the coal bunker and the coal leakage hole, and ensures the safety of the entire transportation environment.

[0030] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0032] Figure 1A flow chart of the method for automatically identifying and grabbing foreign matter and bulk objects in coal mine tape provided by the present invention;

[0033] Figure 2 This is a schematic diagram of the installation of the automatic identification and grabbing device for foreign matter and large blocks of coal mine tape provided by the present invention. DETAILED DESCRIPTION

[0034] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0035] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0036] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0037] See also Figure 1-2The present invention provides a method and device for automatically grabbing foreign matter and large pieces of coal mine tape based on AI video recognition. The method mainly includes two parts: accurate recognition of hard foreign matter and large pieces on the front tape and automatic and rapid capture and grabbing of hard foreign matter and large pieces at the back end. The device mainly includes: a video server, an explosion-proof and intrinsically safe power supply host, a high-definition camera, a video acquisition control host, a robotic arm capturer and other equipment. Among them, the high-definition camera, the video acquisition control host, and the robotic arm capturer are the core of the system. The robotic arm capturer includes a robotic arm, a drive device, an electric control, a guide device, and an end effector. The video acquisition control host is mainly composed of video signal acquisition, data analysis and processing, logical judgment, control output and other parts. The video server mainly stores the collected video content for a long time, issues voice alarm prompts in case of abnormal situations, captures images of foreign objects and large objects, communicates with the video acquisition control host over the network, provides measurement point parameter definitions, and queries data; the explosion-proof and intrinsically safe power supply mainly provides DC intrinsically safe power for the video acquisition control host and the HD camera; the HD camera collects the front-end video data in real time, and uploads the collected video data to the video acquisition control host in real time; the video acquisition control host realizes network communication, video data acquisition, analysis and processing, and control output with the video server and HD camera; the robotic arm capturer receives control instructions from the video acquisition control host to grab foreign objects and large pieces of material on the tape.

[0038] High-definition cameras, video acquisition control hosts, flameproof and intrinsically safe power hosts, robotic arm catchers and other equipment are installed at the front, middle and rear ends of the first belt conveyor, the middle of the middle belt conveyor, and the middle of the tail belt conveyor. Figure 2 shown.

[0039] The HD camera and the video acquisition control host are connected by RJ45 network port, and the distance between the HD camera and the video acquisition control host cannot exceed 50m; the explosion-proof and intrinsically safe power supply host is connected to the video acquisition control host and the HD camera by multi-core cable. The intrinsically safe power supply host outputs 24V and 12V intrinsically safe DC power supply. 24V is mainly used to power the HD camera, and the distance cannot exceed 150m. 12V is mainly used to power the main board and communication board of the video acquisition control host; the control between the robotic arm capturer and the video acquisition control host is connected by cable. The robotic arm capturer is the core of the system. The robotic arm capturer consists of a robotic arm body, a motor, and an electronic control part. The video acquisition control host sends a control instruction to the capturer electronic control. The electronic control part controls the rotation direction and start and stop of the motor, and the motor drives the robotic arm to close and separate; the capturer motor is powered by 380V AC, and the robotic arm end effector consists of 5-finger grippers on the left and right sides to achieve the grasping of foreign objects or large pieces of materials.

[0040] The installation distance between the high-definition camera and the robotic arm capturer is determined according to the running speed of the belt conveyor, the inspection cycle, and the action time of the arm capturer.

[0041] S=TV

[0042] T=T 1 +T 2

[0043] Where S represents the installation distance, m; V represents the running speed of the belt conveyor, m / s; T represents the action time, s; T 1 Indicates the inspection cycle of the video acquisition control host, ms; T 2 Represents the action time of the robot capturer, s.

[0044] The inspection cycle and the action time of the robotic arm capturer can be measured by a professional oscilloscope. The running speed of the belt conveyor can be determined by calculating the belt roller diameter and the roller transmission speed. Since the belt width, roller diameter, motor power and running speed used in each pair of mines are inconsistent, the installation distances are different.

[0045] The system software is used to set the fill light, wide dynamic range, image enhancement and other parameters of each HD camera on site, and the set parameters are sent down to obtain high-quality video images, and the video image signals are uploaded to the video acquisition control host. According to the key features of the object such as edge, contour, texture, size, and dimension, as the basis for judging large pieces of coal, large pieces of gangue, anchor rods, anchor cables, wood piles and other foreign objects and large pieces of materials, random forest, CNN and other models are established to classify and identify the extracted features. When foreign objects or large pieces of materials are identified, the video acquisition control host outputs control instructions on the spot. The control port of the video acquisition control host and the electronic control of the robotic arm capturer are connected by a cable. After the robotic arm capturer receives the control instruction, the electronic control drives the motor to the guide device, joints and end effector. The robotic arm capturer extends to the foreign objects or large pieces of materials on the belt conveyor, and grabs the foreign objects or large pieces of materials in the predetermined direction and action and places them in the material frame of the roadway to achieve the first-level foreign object grabbing. After the grabbing is completed, the robotic arm capturer automatically resets and waits for the next instruction.

[0046] Coal mining or tunneling is carried out continuously, and the belt conveyor runs continuously. Foreign objects or large pieces of materials may appear continuously. Therefore, there is a possibility of failure or omission in the grabbing of foreign objects or large pieces of materials at the front end. High-definition cameras, video acquisition control hosts, robotic arm capturers and other equipment must be installed at a suitable position in the middle of the first belt conveyor. The recognition and control principles of foreign objects and large pieces of materials are the same as those of the front-end recognition. When the video acquisition control host recognizes the presence of foreign objects or large pieces of materials, it outputs control instructions on the spot. After the robotic arm capturer receives the control instructions, it controls the robotic arm to extend to the foreign objects or large pieces of materials on the belt, grabs the foreign objects or large pieces of materials and places them in the tunnel material frame to achieve the second-level foreign object grabbing.

[0047] After identifying and grabbing foreign objects or large pieces of materials on the belt conveyor twice at the front and middle, high-definition cameras, video acquisition control host, robotic arm capturer and other equipment are installed at appropriate positions at the rear end of the first belt conveyor to identify and grab foreign objects or large pieces of materials on the belt from different angles. The identification and control principles of foreign objects and large pieces of materials are the same as those of the front-end identification. When the video acquisition control host identifies the presence of foreign objects or large pieces of materials, it outputs control instructions on the spot. After receiving the control instructions, the robotic arm capturer controls the robotic arm to extend to the foreign objects or large pieces of materials on the belt, grabs the foreign objects or large pieces of materials and places them in the aisle material frame, thereby realizing the grabbing of the third-level foreign objects.

[0048] Secondly, the system identifies and verifies the grabbing effect of the first three levels of foreign objects and large pieces. When the coal flow flows from the first belt conveyor to the middle belt conveyor, the structure, direction and size of the coal flow on the belt change to a certain extent. In order to verify whether the coal flow and foreign objects on the belt are grabbed cleanly, two high-definition cameras, video acquisition control host, mechanical arm capture device and other devices are installed at the appropriate position in the middle of the middle belt conveyor from different angles. The video acquisition control host and the ground video server use network transmission. The video acquisition control host uploads the collected video signal to the ground video server in real time. When the coal flow passes through the middle belt conveyor, if there are no foreign objects and large pieces of materials, it means that the first three levels of grabbing have achieved the effect. If foreign objects or large pieces of materials are identified on the belt conveyor, it means that the first three levels of grabbing have not achieved the effect. The system immediately issues a voice alarm prompt, captures the image of the foreign object, records the alarm information, and immediately sends the captured foreign object image to the terminal video acquisition control host, waiting for the terminal capturer to grab, and realize the front-end grabbing effect verification and grab failure alarm.

[0049] Finally, the on-site foreign object image is compared with the system foreign object image to confirm the capture. Two high-definition cameras, a video acquisition control host, a robotic arm capturer and other equipment are installed at a suitable position in the middle of the end belt conveyor to identify foreign objects or large pieces of materials on the belt and compare them with the foreign object images received from the ground server. When the front-end camera identifies foreign objects and large pieces of materials with a similarity of more than 90% with the foreign objects and large pieces of materials in the image, the video acquisition control host outputs a control command on the spot. After receiving the control command, the robotic arm capturer controls the robotic arm to reach out to the foreign objects or large pieces of materials on the belt, grabs the foreign objects or large pieces of materials and places them in the tunnel material frame. The rear-end camera confirms the grabbing effect of the front-end camera, and captures the image of the point through coal flow velocity calculation and uploads it to the system as a foreign object image feedback signal, which is compared with the previous foreign object image. If the two images are inconsistent, it means that the grabbing is successful, and the voice alarm prompt ends to confirm that the grabbing is successful, realizing the grabbing confirmation of foreign objects and large pieces of materials, ensuring that there are no foreign objects and large pieces of materials in the coal bunker and coal leakage eyes at the end, and ensuring the safety of the rectification and transportation links.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. An automatic identification and grabbing device for foreign matter and bulk objects in coal mine tapes, characterized in that: The device comprises a high-definition camera, a video acquisition control host, a mechanical arm catcher, a flameproof and intrinsically safe power supply host and a video server; the high-definition camera, the video acquisition control host, the mechanical arm catcher and the flameproof and intrinsically safe power supply host constitute a group of monitoring and control devices; A set of monitoring and control devices are installed at the front, middle and rear ends of the first belt conveyor after the mining face to achieve three-level precise capture and grabbing of foreign objects and large blocks; A high-definition camera, a video acquisition control host, and a flameproof and intrinsically safe power supply host are installed in the middle of the central belt conveyor to verify the three-level grabbing effect; when there is a lack of grabbing, missed grabbing, or grabbing failure, the ground video server captures the image of foreign objects and large pieces of materials, immediately issues a voice alarm prompt, records the alarm information, and sends the captured image to the video acquisition control host of the rear belt conveyor; A set of monitoring and control devices is installed in the middle position of the tail belt conveyor. The video acquisition control host receives the foreign body image from the video server and compares it with the foreign body image on site. When the similarity of the shape, size, edge and contour of the foreign body is above 90%, the video acquisition control host issues a control command on the spot, and the robotic arm capturer starts to grab the foreign body. When the foreign body is successfully grabbed, the video acquisition control host uploads the captured video image to the ground video server and compares it with the previous foreign body image. The system realizes the grabbing confirmation of foreign bodies and large pieces of materials, ensuring that there are no foreign bodies and large pieces of materials flowing into the coal bunker and the coal leakage eye.

2. The automatic identification and grabbing device for foreign matter and bulk objects in coal mine tape according to claim 1 is characterized in that: The automatic recognition and grasping method of the device specifically comprises the following steps: S1: At the first belt conveyor, a three-level capture method is adopted to capture foreign objects and large pieces of materials on the belt conveyor; S2: At the middle belt conveyor, the system identifies and verifies the grabbing effect of the first three levels of foreign objects and large objects; S3: At the tail belt conveyor, by comparing the on-site foreign object image with the system foreign object image, it is confirmed that foreign objects and large pieces of materials have been captured.

3. The automatic identification and grabbing device for foreign matter and bulk objects in coal mine tape according to claim 2 is characterized in that: Step S1 specifically includes: installing a high-definition camera at the front end of the first belt conveyor to detect the coal flow, foreign matter and bulk materials on the belt, using on-site fill light, system brightness adjustment, wide dynamic and image enhancement technology to pick up high-quality dynamic images, and uploading the video signal to the video acquisition control host through the network. The video acquisition control host has built-in Faster R-CNN foreign matter recognition technology, and uses the edge, contour, texture, size and dimension of the object as the basis for judging foreign matter and bulk materials, establishes a random forest or CNN model, and classifies and recognizes the extracted features. When foreign matter or bulk materials are identified, the video acquisition control host outputs a control instruction on the spot, and the control port of the video acquisition control host is connected to the electronic control of the manipulator capturer by a cable. After the manipulator capturer receives the control instruction, the electronic control drives the motor to the guide device, the joint and the end effector, and the manipulator extends to the foreign matter or bulk materials on the belt, and in accordance with the predetermined direction and action, the foreign matter or bulk materials are grabbed and placed in the lane material basket, so as to realize the grabbing of the first-level foreign matter. After the grabbing is completed, the manipulator capturer automatically resets and waits for the next instruction; A set of monitoring and control devices is installed in the middle of the first belt conveyor. The recognition and control principles of foreign objects and bulk materials are the same as those of the front-end recognition. When the video acquisition control host recognizes that there are foreign objects or bulk materials, it outputs control instructions on the spot. After the mechanical arm capturer receives the control instructions, it controls the mechanical arm to reach the foreign objects or bulk materials on the belt, grab the foreign objects or bulk materials and place them in the lane material basket, thus realizing the second-level foreign object grabbing. After identifying and grabbing the foreign objects or large pieces of materials on the belt conveyor twice at the front and middle, a set of monitoring and control devices is installed at the rear end of the first belt conveyor to identify and grab the foreign objects or large pieces of materials on the belt from different angles. The identification and control principles of foreign objects and large pieces of materials are the same as those of the front-end identification. When the video acquisition control host identifies the presence of foreign objects or large pieces of materials, it outputs control instructions on the spot. After the robotic arm capturer receives the control instructions, it controls the robotic arm to extend to the foreign objects or large pieces of materials on the belt, grab the foreign objects or large pieces of materials and place them in the material basket of the aisle, thereby realizing the grabbing of the third-level foreign objects.

4. The automatic identification and grabbing device for foreign matter and bulk objects in coal mine tape according to claim 2 is characterized in that: Step S2 specifically includes: installing two high-definition cameras, a video acquisition control host and a flameproof and intrinsically safe power supply host at different angles in the middle position of the middle belt conveyor, using network transmission between the video acquisition control host and the ground video server, and the video acquisition control host uploading the collected video signal to the ground video server in real time. When the coal flow passes through the middle belt conveyor, if there are no foreign objects or large pieces of materials, it means that the first three levels of grasping have achieved the effect. If foreign objects or large pieces of materials are identified on the belt conveyor, it means that the first three levels of grasping have not achieved the effect. The system immediately issues a voice alarm prompt, captures the image of the foreign object, records the alarm information, and immediately sends the captured foreign object image to the video acquisition control host of the rear belt conveyor, waiting for the mechanical arm capturer of the rear belt conveyor to grasp it, so as to realize the front-end grasping effect verification and grasping failure alarm.

5. The automatic identification and grabbing device for foreign matter and bulk objects in coal mine tape according to claim 2 is characterized in that: Step S3 specifically includes: installing two high-definition cameras, a video acquisition control host, a mechanical arm catcher and a flameproof and intrinsically safe power supply host at the middle position of the rear belt conveyor, identifying foreign objects or large pieces of materials on the belt, and comparing them with the foreign object images received from the ground video server. When the front-end high-definition camera identifies foreign objects and large pieces of materials and the image foreign objects and large pieces of materials with a similarity of more than 90%, the video acquisition control host outputs a control instruction on the spot. After the mechanical arm catcher receives the control instruction, it controls the mechanical arm to extend to the foreign objects or large pieces of materials on the belt, grabs the foreign objects or large pieces of materials and places them in the lane material basket. The high-definition camera of the rear belt conveyor confirms the image effect captured by the high-definition camera of the middle belt conveyor, and compares them with the foreign object images captured by the high-definition camera of the middle belt conveyor. If the two images are inconsistent, it means that the grabbing is successful, and the voice alarm prompt is ended to confirm that the grabbing is successful, and the grabbing confirmation of foreign objects and large pieces of materials is realized to ensure that there are no foreign objects and large pieces of materials in the coal bunker and the coal leakage eye at the end.

6. The automatic identification and grabbing device for foreign matter and bulky objects in coal mine tape according to any one of claims 1 to 5, characterized in that: In order to accurately capture foreign objects and large pieces of materials, it is necessary to determine the installation distance between the high-definition camera and the robotic arm capturer. The installation distance is determined according to the running speed of the belt conveyor, the inspection cycle and the action time of the robotic arm capturer. S=TV T=T1+T2 Among them, S represents the installation distance; V represents the running speed of the belt conveyor; T represents the action time; T1 represents the inspection cycle of the video acquisition control host; T2 represents the action time of the robotic arm capturer.

7. The automatic identification and grabbing device for foreign matter and bulk objects in coal mine tape according to claim 6 is characterized in that: The inspection cycle and the action time of the robot arm catcher are measured by a professional oscilloscope, and the running speed of the belt conveyor is determined by calculating the belt roller diameter and the roller transmission speed.

8. The automatic identification and grabbing device for foreign matter and bulky objects in coal mine tape according to any one of claims 1 to 5, characterized in that: The video acquisition control host is used to realize video signal acquisition, data analysis and processing, logical judgment and control output; The high-definition camera collects front-end video data in real time and uploads the collected video data to the video acquisition control host in real time; the robotic arm capturer receives control instructions from the video acquisition control host to grab foreign objects and large pieces of materials on the tape; the video server saves the collected video content for a long time, and in case of abnormal conditions, it issues voice alarm prompts, captures images of foreign objects and large pieces, and communicates with the video acquisition control host over the network, providing measurement point parameter definition and data query functions; the explosion-proof and intrinsically safe power supply provides DC intrinsically safe power for the video acquisition control host and the high-definition camera.

9. The automatic identification and grabbing device for foreign matter and bulky objects in coal mine tape according to any one of claims 1 to 5, characterized in that: The robot arm capture device includes a robot arm, a driving device, an electric control, a guiding device and an end effector.

Images