CN117517327A - Rail transit locomotive cleaning defect detection equipment based on deep learning - Google Patents
Rail transit locomotive cleaning defect detection equipment based on deep learning Download PDFInfo
- Publication number
- CN117517327A CN117517327A CN202410022021.7A CN202410022021A CN117517327A CN 117517327 A CN117517327 A CN 117517327A CN 202410022021 A CN202410022021 A CN 202410022021A CN 117517327 A CN117517327 A CN 117517327A
- Authority
- CN
- China
- Prior art keywords
- bearing
- film
- rail transit
- component
- deep learning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 68
- 238000004140 cleaning Methods 0.000 title claims abstract description 35
- 230000003137 locomotive effect Effects 0.000 title claims abstract description 31
- 230000007547 defect Effects 0.000 title claims abstract description 26
- 238000013135 deep learning Methods 0.000 title claims abstract description 24
- 238000012546 transfer Methods 0.000 claims abstract description 43
- 239000012188 paraffin wax Substances 0.000 claims abstract description 32
- 230000000007 visual effect Effects 0.000 claims abstract description 14
- 238000001816 cooling Methods 0.000 claims abstract description 11
- 238000010438 heat treatment Methods 0.000 claims abstract description 10
- 238000005086 pumping Methods 0.000 claims abstract description 10
- 239000010408 film Substances 0.000 claims description 59
- 238000003780 insertion Methods 0.000 claims description 40
- 230000037431 insertion Effects 0.000 claims description 40
- 239000013039 cover film Substances 0.000 claims description 25
- 239000012528 membrane Substances 0.000 claims description 22
- 239000002390 adhesive tape Substances 0.000 claims description 18
- 238000004804 winding Methods 0.000 claims description 18
- 238000007689 inspection Methods 0.000 claims description 15
- 238000001179 sorption measurement Methods 0.000 claims description 11
- 238000000034 method Methods 0.000 claims description 10
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 6
- 230000000149 penetrating effect Effects 0.000 claims description 5
- 238000005485 electric heating Methods 0.000 claims description 4
- 239000007788 liquid Substances 0.000 claims description 4
- 230000002195 synergetic effect Effects 0.000 claims description 4
- 230000002457 bidirectional effect Effects 0.000 claims description 3
- 239000013013 elastic material Substances 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 239000004033 plastic Substances 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000012423 maintenance Methods 0.000 description 5
- 230000000712 assembly Effects 0.000 description 4
- 238000000429 assembly Methods 0.000 description 4
- 238000011179 visual inspection Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 3
- CWYNVVGOOAEACU-UHFFFAOYSA-N Fe2+ Chemical compound [Fe+2] CWYNVVGOOAEACU-UHFFFAOYSA-N 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 229940057995 liquid paraffin Drugs 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 239000000853 adhesive Substances 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000003670 easy-to-clean Effects 0.000 description 1
- 239000003302 ferromagnetic material Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000005291 magnetic effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000007711 solidification Methods 0.000 description 1
- 230000008023 solidification Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/08—Railway vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N35/00—Automatic analysis not limited to methods or materials provided for in any single one of groups G01N1/00 - G01N33/00; Handling materials therefor
- G01N35/0099—Automatic analysis not limited to methods or materials provided for in any single one of groups G01N1/00 - G01N33/00; Handling materials therefor comprising robots or similar manipulators
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Robotics (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application discloses a defect detection device after cleaning of a rail transit locomotive based on deep learning, which relates to the technical field of detection devices and comprises a detection workshop, a visual detection robot for a multi-axis mechanical arm provided with a 2D camera and a transfer trolley; the transfer trolley comprises a bearing bottom plate, a bearing frame body, a rotary expansion bracket, a first bearing component, a second bearing component and a paraffin pumping component; the number of the bearing frames is two, and the two bearing frames are symmetrically arranged and are fixed on the side wall of the bearing bottom plate; the rotary telescopic frame is rotationally connected to the bearing frame body; the first bearing component and the second bearing component are identical in structure and symmetrically arranged; the first bearing plate is fixed on the rotary expansion bracket, and the first bearing block is fixed on the first bearing plate; the first top film is an elastic rubber film, and the edge of the first top film is fixed on the edge of the first bearing block; the first bearing block is provided with a heating component and a cooling component; the technical effects of high detection efficiency, unified standard and high safety of rail transit locomotive components are achieved.
Description
Technical Field
The invention relates to the technical field of detection equipment, in particular to equipment for detecting defects of a rail transit locomotive after cleaning based on deep learning.
Background
During service of rail transit locomotives, it is often necessary to disassemble some of the components for more thorough inspection and maintenance. The disassembled parts comprise disassembled electrical equipment, mechanical parts, wheels, rails and the like; disassembly of these components allows the technician to better inspect their condition and perform the necessary maintenance and replacement; this ensures the safety and reliability of the locomotive and extends its service life.
Deep learning is a branch of machine learning, which aims to enable a computer to learn and understand complex data by simulating the working mode of human brain neurons; the core of deep learning is an artificial neural network, which consists of multiple layers of neurons, wherein each layer processes and abstracts input data to different degrees to finally obtain a prediction result; the deep learning can be used for image segmentation, namely segmentation and recognition of different objects in an image; the method is widely applied to the field of mechanical equipment automation and automatic detection.
In the prior art, the detection of the components of the rail transit locomotive is generally finished by manpower, and cleaning is needed before the detection so as to avoid the defect of dirt blocking and the influence of dirt on the normal operation of equipment; however, the manual detection method has the defects of low efficiency, non-uniform standard (different judgment standards of each person, difficulty in obtaining correct data), high safety risk (easy fatigue and easy accident during long-time manual work), no detection data (difficult statistical analysis and incapability of providing guidance for subsequent maintenance and replacement operation), and is unfavorable for normal maintenance.
Therefore, there is a need for a defect detection device for rail transit locomotives, which can automatically complete the detection of components of the rail transit locomotives, and has the advantages of high detection process efficiency, unified standard and high safety.
Disclosure of Invention
According to the rail transit locomotive cleaning defect detection equipment based on deep learning, the technical problems that in the prior art, rail transit locomotive parts need to be detected manually, the detection process is low in efficiency, the standards are not uniform, the safety is poor, and the data record is unclear are solved, and the technical effects of high detection efficiency, uniform standards and high safety of the rail transit locomotive parts are achieved.
The embodiment of the application provides a defect detection device after cleaning of a rail transit locomotive based on deep learning, which comprises a detection workshop, a visual detection robot for a multi-axis mechanical arm provided with a 2D camera and a transfer trolley;
the transfer trolley comprises a bearing bottom plate, a bearing frame body, a rotary expansion bracket, a first bearing component, a second bearing component and a paraffin pumping component;
the bottom of the bearing bottom plate is provided with a plurality of omnidirectional wheels;
the number of the bearing frames is two, and the two bearing frames are symmetrically arranged and are fixed on the side wall of the bearing bottom plate;
the rotary telescopic frames are bidirectional electric telescopic rods, are in one-to-one correspondence with the bearing frame bodies and are rotatably connected to the bearing frame bodies;
the first bearing component and the second bearing component are identical in structure and symmetrically arranged;
the first bearing plate is fixed at one end of the rotary expansion bracket at a position close to the edge, and the first bearing block is fixed on the first bearing plate;
the first top film is an elastic rubber film, is rectangular in shape, and is fixed on the edge of the surface, far away from the first bearing plate, of the first bearing block, so that a flat rectangular space is formed by the first top film and the first bearing block;
the first bearing block is provided with a heating component and a cooling component;
the paraffin pumping assembly communicates the paraffin storage bin with the two rectangular spaces.
Further, the area of the surface of the first bearing block, which is contacted with the first bearing plate, is more than 0.45 times of the surface area of the first bearing plate.
Further, a warehouse area, a charging station, a calling loading area and a buffer station are arranged in the detection workshop; the warehouse area is used for storing accessories and detecting consumable materials; the charging station is used for charging the transfer trolley; the calling feeding area is used for stopping a transfer trolley carrying a workpiece, and the transfer trolley moves to the vicinity of the visual detection robot at the first time after the detection is completed; the buffer position is the parking stall of transportation dolly.
Further, the rotary expansion bracket comprises a base rod, a first sliding rod and a second sliding rod; the base rod is rotationally connected to the bearing frame body, the first sliding rod and the second sliding rod are hard rod bodies, the length direction of the first sliding rod is the same as that of the base rod, the first sliding rod and the second sliding rod are both positioned on the base rod in a sliding mode, and the first sliding rod and the second sliding rod slide under the synergistic effect of the control unit and the power assembly so that the rotating telescopic frame stretches; the sliding directions of the first sliding rod and the second sliding rod are opposite.
Further, the heating component is an electric heating wire; the cooling component is a combination of a liquid cooling pipeline, a water pump and a cold water storage bin.
Preferably, the first bearing block and the second bearing block are respectively internally provided with an adsorption magnet;
the adsorption magnet is a plate-shaped electromagnet;
after shooting the top surface of the workpiece and resetting the vision detection robot, controlling the first bearing assembly and the second bearing assembly to be close to each other, and then controlling the adsorption magnet in the first bearing block to be electrified so as to adsorb the workpiece; controlling the first top film to expand first, and wrapping the ground and part or all of the side surfaces of the workpiece; thereafter controlling the increase in the amount of paraffin in the second space such that the second cover film covers the top surface of the work piece; controlling the second space and the rotating telescopic frame to rotate 180 degrees after the paraffin in the second space is solidified, and controlling the first bearing component and the second bearing component to be far away from each other; then the mechanical arm is controlled to drive the 2D camera to extend between the first bearing component and the second bearing component; photographing the original bottom surface and the side surface of one or more workpieces to perform visual detection; after which the vision inspection robot is controlled to reset.
Preferably, the surface of the first covering film, which is close to the first bearing block, is densely covered with film bottom inserting columns; the membrane bottom insertion column is cylindrical and can be attracted by the magnet;
a limiting elastic net is arranged between the first top film and the first bearing block;
the limiting elastic net is a rubber net made of elastic materials, the whole limiting elastic net is rectangular, and the edge of the limiting elastic net is fixed at the edge of the first covering top film;
the limiting elastic net penetrates through all the membrane bottom inserting columns and is fixed on the membrane bottom inserting columns and used for assisting the membrane bottom inserting columns to be inserted into the inserting holes;
the spacing between the spacing elastic net and the first covering film is more than 0.6 cm;
the surface, close to the first top film, of the first bearing block is densely provided with insertion holes, and the insertion holes correspond to the film bottom insertion columns one by one and are used for inserting the film bottom insertion columns; the length of the membrane bottom insertion column is shorter than the depth of the insertion hole;
a meshed net body penetrating groove is formed in the surface, close to the first top film, of the first bearing block;
the first bearing block is internally provided with an adsorption magnet, and the adsorption magnet plays a role in adsorbing the insertion column at the bottom of the fixed film.
Preferably, the bottom of the membrane bottom insertion column is hemispherical.
Preferably, the membrane bottom insertion column comprises a foundation column and an iron ring body;
the foundation column is a plastic or rubber column and is fixed on the first covering top film;
the iron ring body is sleeved and fixed on the foundation column and is 1.5 to 2.5 cm away from the fixed end of the foundation column;
the attracting magnet is positioned in the first bearing block and is 1.5 to 2.5 cm away from the surface of the first bearing block, which is close to the first covering film.
Preferably, the detection workshop is also internally provided with an ash sticking cleaning assembly, and the transfer trolley is cleaned once by using the ash sticking cleaning assembly after each use;
the ash sticking cleaning assembly comprises a bearing frame, a winding and unwinding roller and an adhesive tape body;
the bearing frame is fixed on the ground and is of a frame structure and is used for bearing and fixing the winding and unwinding roller;
the number of the winding and unwinding rollers is more than 1.5 times of the length of the transfer trolley, motors are arranged in the winding and unwinding rollers, and the winding and unwinding rollers are transversely arranged and positioned on the bearing frame and used for winding and releasing the adhesive tape body;
the adhesive tape body is a tape body with two surfaces coated with adhesive tapes, and the two ends of the adhesive tape body are respectively wound and positioned on the two winding and unwinding rollers; the width of the adhesive tape body is 0.5 to 0.6 times of the width of the transfer trolley.
One or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:
by utilizing visual detection to replace manual detection and performing statistical analysis on detected historical data through deep learning, places which are least easy to clean and places which are most frequently defective can be obtained step by step so as to guide the subsequent cleaning and maintenance to be performed with pertinence; the technical problems that in the prior art, rail transit locomotive components need to be detected manually, the detection process is low in efficiency, the standards are not uniform, the safety is poor, and the data record is unclear are effectively solved, and further the technical effects of high detection efficiency, uniform standards and high safety of the rail transit locomotive components are achieved.
Drawings
FIG. 1 is a schematic diagram of the overall structure of the deep learning-based post-cleaning defect detection device for rail transit locomotives of the present invention.
Fig. 2 is an external structural schematic view of the transfer trolley.
Fig. 3 is a schematic structural view of the transfer trolley.
Fig. 4 is a schematic view of the turning state of the transfer trolley.
Fig. 5 is a schematic diagram of the flipped state of the workpiece.
Fig. 6 is a schematic view of the structure of the first carrier block.
Fig. 7 is a schematic diagram of the flipped over state of the ferrous workpiece.
Fig. 8 is a schematic diagram of the positional relationship between the work and the second coverlay.
Fig. 9 is a schematic diagram showing the positional relationship between the membrane-bottom insertion column and the insertion hole.
Fig. 10 is a schematic diagram showing the positional relationship between the first cover top film and the film bottom insertion column.
Fig. 11 is a schematic diagram of a process of flipping a non-ferrous workpiece.
Fig. 12 is a schematic diagram of the positional relationship between the ash sticking cleaning assembly and the transfer trolley.
In the figure:
warehouse area 001, charging station 002, vision inspection robot 003, call loading area 004, buffer station 005, transfer cart 006, work piece 007, load floor 100, paraffin storage warehouse 110, load frame 200, rotation drive assembly 210, rotation telescoping rack 300, base bar 310, first slide bar 320, second slide bar 330, first support assembly 400, first load floor 410, first load block 420, heater wire 421, liquid cooling channel 422, first cover top film 430, film bottom insert column 440, iron ring 441, spacing elastic net 442, insert holes 443, net body penetration groove 444, second support assembly 500, second load floor 510, second load floor 520, second cover top film 530, paraffin pumping assembly 600, adsorption magnet 700, load frame 810, take-up drum 820, and tape body 830.
Detailed Description
In order that the invention may be readily understood, a more complete description of the invention will be rendered by reference to the appended drawings; the preferred embodiments of the present invention are illustrated in the drawings, however, the present invention may be embodied in many different forms and is not limited to the embodiments described herein; rather, these embodiments are provided so that this disclosure will be thorough and complete.
It should be noted that the terms "vertical", "horizontal", "upper", "lower", "left", "right", and the like are used herein for illustrative purposes only and do not represent the only embodiment.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention; the term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Example 1
As shown in fig. 1, the rail transit locomotive cleaning defect detection device based on deep learning of the application comprises a detection workshop, a visual detection robot 003 and a transfer trolley 006.
The detection workshop is an action place constructed for detection, and is provided with a warehouse area 001, a charging position 002, a calling feeding area 004 and a buffer position 005; the warehouse area 001 is used for storing accessories and detecting consumable materials; the charging potential 002 is used to charge the transfer trolley 006; the call loading area 004 is used for stopping the transfer trolley 006 carrying the workpiece 007, and the transfer trolley 006 moves to the vicinity of the vision inspection robot 003 at the first time after the on-going inspection is completed; buffer position 005 is the parking stall of transfer dolly 006.
Visual inspection robot 003 is the multiaxis arm that is equipped with 2D camera, establishes in the inspection shop and fixes subaerial.
Further, the visual inspection robot 003 is a 4-axis mechanical arm, a 2D camera is arranged at the end part of the mechanical arm, and the 2D camera is rotatably connected to the end part of the 4-axis mechanical arm and is rotated under the control of the control unit.
Further, the visual inspection robot 003 is a 3-axis mechanical arm, and five 2D cameras are arranged at the end part of the mechanical arm, wherein a lens of one 2D camera faces to the right lower side and is used for shooting the top surface of a workpiece; the other four 2D cameras are located in the circumference of the 2D camera and have equal intervals, and the angles between the lens orientations of the four 2D cameras and the ground are 35 to 55 degrees, so as to shoot the side face of the workpiece 007.
As shown in fig. 2 to 5, the transfer cart 006 is configured to transfer the workpiece 007 and timely turn the workpiece 007 over, thereby facilitating the detection of the workpiece 007, and comprises a load floor 100, a load frame 200, a rotating expansion frame 300, a first support assembly 400, a second support assembly 500, a paraffin pumping assembly 600, a power assembly and a control unit;
the bearing bottom plate 100 is a hard rectangular plate body, the bottom of the bearing bottom plate is provided with a plurality of omnidirectional wheels, and the omnidirectional wheels rotate under the synergistic effect of the control unit and the power assembly so as to realize the movement of the transfer trolley 006; one or two elongated paraffin storage bins 110 are arranged at the top of the bearing bottom plate 100 near the edge, and the paraffin storage bins 110 are used for storing liquid paraffin and are internally provided with heating components.
Further, the length direction of the paraffin storage bin 110 is the same as the length direction of the loading base plate 100.
The bearing frame body 200 is plate-shaped and longitudinally arranged to play a role in bearing the positioning rotating telescopic frame 300; the number of the bearing frame bodies 200 is two, and the bearing frame bodies are symmetrically arranged and are fixed on the side wall of the bearing bottom plate 100; the width direction of the carrier body 200 is the same as the width direction of the carrier base plate 100.
The rotary telescopic frames 300 are bidirectional electric telescopic rods, are in one-to-one correspondence with the bearing frame bodies 200, and are rotatably connected to the bearing frame bodies 200 at the non-rotary parts in the middle; a rotation driving assembly 210 is also positioned on the bearing frame 200, and the rotation driving assembly 210 is preferably a motor; the rotating telescopic frame 300 is driven by the rotating driving assembly 210 to rotate, and the axial direction of the rotating shaft is the same as the length direction of the carrying floor 100.
Further, the rotating telescopic frame 300 includes a base rod 310, a first sliding rod 320 and a second sliding rod 330; the base rod 310 is rotatably connected to the bearing frame 200, the first sliding rod 320 and the second sliding rod 330 are hard rod bodies, the length direction is the same as the length direction of the base rod 310, and the first sliding rod and the second sliding rod are both positioned on the base rod 310 in a sliding manner, and slide under the synergistic effect of the control unit and the power assembly to enable the rotating telescopic frame 300 to stretch; the sliding directions of both the first sliding bar 320 and the second sliding bar 330 are opposite.
The first support assembly 400 and the second support assembly 500 are both plate-shaped, have the same structure, are symmetrically arranged and are respectively fixed at two ends of the rotating telescopic frame 300, and are used for bearing and placing the workpiece 007;
the first support assembly 400 includes a first carrier plate 410, a first carrier block 420, and a first cover film 430;
the first carrying plate 410 is a hard rectangular plate body, and positions close to edges of the first carrying plate are respectively fixed at one ends of the two rotating telescopic frames 300, and are transversely arranged to perform the function of carrying and supporting;
the first bearing block 420 is a rectangular plate-shaped block, and is fixed on the first bearing plate 410, for bearing and fixing the first cover film 430; the area of the surface of the first bearing block 420 contacted with the first bearing plate 410 is more than 0.45 times of the surface area of the first bearing plate 410;
the first top film 430 is an elastic rubber film, and has a rectangular shape, and the edge is fixed on the edge of the surface of the first bearing block 420 away from the first bearing plate 410, and forms a flat rectangular space together with the first bearing block 420;
as shown in fig. 6, a heating component and a cooling component for controlling the temperature and the state of paraffin in the rectangular space are disposed on the first bearing block 420 at a position close to the first cover film 430; the heating component is a heating wire 421, and the heating wire 421 is preferably an electric heating wire; the cooling component is preferably a combination of a liquid cooling pipeline 422, a water pump and a cold water storage bin (a refrigerating component is arranged in the bin).
The first carrier plate 410 and the second carrier plate 510 have the same structure, the first carrier block 420 and the second carrier block 520 have the same structure, and the first cover top film 430 and the second cover top film 530 have the same structure, which is not described herein.
The paraffin pumping assembly 600 is a combination of a pump and a hose, and electric heating wires for preventing paraffin from solidifying are wound on the pump and the hose; the paraffin pumping assembly 600 communicates the paraffin storage bin 110 with two rectangular spaces for controlling the amount of paraffin in the rectangular spaces; for convenience of description, a space formed by the first carrier block 420 and the first cover film 430 is defined herein as a first space, and a space formed by the second carrier block 520 and the second cover film 530 is defined herein as a second space.
The power component is used for providing power for the operation of each part of the transfer trolley 006, and the control unit plays a role in controlling the coordinated operation of each part of the transfer trolley 006, which is the prior art and is not described herein.
Preferably, the control unit is a combination of a programmable logic controller and a remote control unit.
The deep learning-based rail transit locomotive cleaning defect detection device of the embodiment of the application needs to be trained firstly before being used (in a training stage, a large amount of image data containing defects, dirt and normal workpieces need to be used for training a neural network model, defects comprise cracks, rust and the like, the image data are marked, namely a defect part, a dirt part and a normal part in an image are marked manually, then the marked data are used for training a deep learning model, and a visual detection and deep learning algorithm is the prior art and is not repeated here.
The steps are as follows in turn when in use:
1. cleaning (avoiding the influence of dirt on detection defects);
2. the transfer trolley 006 is controlled to approach the vision inspection robot 003, and then the mechanical arm is controlled to drive the 2D camera to extend between the first support assembly 400 and the second support assembly 500; photographing the top surface and the side surface of one or more workpieces to perform visual detection; thereafter controlling the vision inspection robot 003 to reset;
3. the paraffin amount in the first space and the second space is smaller than 30 milliliters under normal state; as shown in fig. 5, the first and second support assemblies 400,500 are controlled to be adjacent to each other, and then the amount of paraffin in the second space is controlled to be increased so that the second cover film 530 covers the work piece 007 on the transfer cart 006; thereafter controlling the solidification of the paraffin in the second space;
4. controlling the rotating gantry 300 to rotate 180 degrees and the first and second support assemblies 400 and 500 to move away from each other; the control arm then drives the 2D camera to extend between the first support assembly 400 and the second support assembly 500; photographing the top surface (original bottom surface) and the side surface of one or more workpieces to perform visual detection; after that, the vision inspection robot 003 is controlled to reset.
The technical scheme in the embodiment of the application at least has the following technical effects or advantages:
the technical problems that in the prior art, rail transit locomotive components need to be detected manually, the detection process is low in efficiency, the standards are not uniform, the safety is poor, and the data record is unclear are solved, and the technical effects of high detection efficiency, uniform standards and high safety of the rail transit locomotive components are realized; through the setting of the work piece flip structure of transfer trolley 006, can be convenient high-efficient carry out comparatively thorough multiaspect to the work piece and detect.
Example two
Considering that the manner of covering the workpiece 007 with the second cover film 530 in the above embodiment can effectively and quickly fix the workpiece 007, after the workpiece is turned over, the raised second cover film 530 and the shadow formed by the same may have a certain adverse effect on the visual detection result, and may affect the detection accuracy to a certain extent; moreover, when the second top film 530 is inflated and pressed down, the workpiece 007 is easy to topple over, and therefore, the partial surface of part of the workpiece 007 is easy to be undetected (missed detection caused by displacement);
in view of the above problems, in the embodiments of the present application, the attracting magnet 700 is added in the first bearing block 420 and the second bearing block 520 based on the above embodiments; the method comprises the following steps:
as shown in fig. 7 and 8, the attracting magnet 700 is a plate-shaped electromagnet.
After photographing the top surface of the workpiece 007 and resetting the vision inspection robot 003, controlling the first and second supporting members 400 and 500 to be close to each other, and then controlling the suction magnets 700 in the first supporting block 420 to be energized to suction the workpiece 007; controlling the first cover film 430 to first expand, wrapping the ground and part or all of the sides of the workpiece 007; thereafter controlling the increase in the amount of paraffin in the second space so that the second cover film 530 covers the top surface of the work 007; controlling the rotation of the rotating expansion bracket 300 by 180 degrees after the paraffin in the second space and the second space is solidified, and controlling the first and second supporting members 400 and 500 to be away from each other; then, the mechanical arm is controlled to drive the 2D camera to extend between the first bearing assembly 400 and the second bearing assembly 500, and the top surface (original bottom surface) and the side surface of one or more workpieces 007 are photographed so as to perform visual inspection; after that, the vision inspection robot 003 is controlled to reset.
Example III
Considering that the attracting magnet 700 in the above embodiment can attract only the workpiece 007 of ferromagnetic material, there is a limit in positioning the workpiece 007 that cannot be attracted by the magnet; in view of the foregoing, embodiments of the present application provide for an optimized improvement in the construction of the first and second support assemblies 400 and 500, specifically:
as shown in fig. 9 to 11, the surface of the first top film 430 close to the first bearing block 420 is densely covered with film bottom insertion columns 440; the membrane bottom insertion column 440 is cylindrical and can be attracted by a magnet;
a spacing elastic net 442 is further disposed between the first top film 430 and the first bearing block 420; the limiting elastic net 442 is a rubber net made of elastic material, is rectangular overall, and has an edge fixed at the edge of the first cover top film 430; the limiting elastic net 442 penetrates through all the membrane bottom inserting columns 440 and is fixed on the membrane bottom inserting columns 440, and is used for assisting the membrane bottom inserting columns 440 to be inserted into the inserting holes 443; the spacing elastic net 442 is spaced from the first cover film 430 by more than 0.6 cm;
the first carrier block 420 has a surface close to the first top film 430, and the surface is densely covered with insertion holes 443, and the insertion holes 443 are in one-to-one correspondence with the film bottom insertion columns 440 for insertion of the film bottom insertion columns 440; the length of the film bottom insertion posts 440 is shorter than the depth of the insertion holes 443;
the surface of the first bearing block 420 near the first top film 430 is provided with a net-shaped net body penetrating groove 444, and the net body penetrating groove 444 is used for penetrating the limiting elastic net 442, so that the first top film 430 can be tightly attached to the first bearing block 420;
the first carrier block 420 is internally provided with an attracting magnet 700, and the attracting magnet 700 plays a role of attracting the fixed film bottom inserting column 440.
The structure of the second support assembly 500 is identical to that of the first support assembly 400 and will not be described in detail herein.
Preferably, the bottom of the membrane bottom insertion column 440 is tapered for ease of insertion.
Preferably, the bottom of the membrane bottom insertion column 440 is hemispherical for easy insertion.
When in use, the utility model is characterized in that: after photographing the top surface of the work piece 007 and resetting the vision inspection robot 003, controlling the first and second holding assemblies 400 and 500 to be adjacent to each other, and then controlling the paraffin pumping assembly 600 to introduce liquid paraffin into the first space such that the portion of the first cover film 430 not contacting the work piece 007 bulges (the portion contacting the work piece 007 does not bulge under the gravity of the work piece 007); then controlling the adsorption magnet 700 in the first bearing block 420 to be electrified, and adsorbing the non-raised film bottom insertion column 440 to fix the film bottom insertion column; then the first top film 430 is controlled to continue to expand, wrapping the ground and part or all of the sides of the workpiece 007; thereafter controlling the increase in the amount of paraffin in the second space so that the second cover film 530 covers the top surface of the work 007; controlling the rotation of the rotating expansion bracket 300 by 180 degrees after the paraffin in the second space and the second space is solidified, and controlling the first and second supporting members 400 and 500 to be away from each other; the control arm then drives the 2D camera to extend between the first support assembly 400 and the second support assembly 500; photographing the top surface (original bottom surface) and the side surface of one or more workpieces to perform visual detection; after that, the vision inspection robot 003 is controlled to reset.
Further, the membrane bottom insertion column 440 includes a base column and an iron ring 441; the foundation column is a plastic or rubber column and is fixed on the first top film 430; the iron ring body 441 is sleeved and fixed on the foundation column and is 1.5 to 2.5 cm away from the fixed end of the foundation column; the attracting magnet 700 is positioned within the first bearing block 420 and is 1.5 to 2.5 cm from the face of the first bearing block 420 that is close to the first cover film 430; after the membrane bottom insertion column 440 is moved up, the magnetic force of the attracting magnet 700 is less bound thereto.
Example IV
In order to facilitate cleaning after the first cover top film 430 and the second cover top film 530 are stained, an ash sticking cleaning assembly is further arranged in the detection workshop, and the transfer trolley 006 is cleaned once by using the ash sticking cleaning assembly after each use; the method comprises the following steps:
as shown in fig. 12, the ash adhesion cleaning assembly includes a carrier 810, a receiving roller 820 and an adhesive tape body 830;
the bearing frame 810 is fixed on the ground and is in a frame structure and is used for bearing and fixing the retractable roller 820;
the number of the winding and unwinding rollers 820 is two, the distance between the winding and unwinding rollers is more than 1.5 times of the length of the transferring trolley 006, motors are arranged in the winding and unwinding rollers, and the winding and unwinding rollers are transversely arranged and positioned on the bearing frame 810 and are used for winding and unwinding the adhesive tape 830;
the adhesive tape 830 is a tape with two sides coated with self-adhesive, and two ends are respectively wound and positioned on the two winding and unwinding rollers 820; the width of the tape body 830 is 0.5 to 0.6 times the width of the transfer cart 006.
After the transfer trolley 006 is used, the transfer trolley moves to the vicinity of the ash adhesion cleaning component and enables the first bearing component 400 and the second bearing component 500 to be respectively positioned right below and right above the adhesive tape body 830, and finally the first top cover film 430 and the second top cover film 530 are tightly attached to the adhesive tape body 830 by controlling the first bearing component 400 and the second bearing component 500 to be close to each other and controlling the first top cover film 430 and the second top cover film 530 to expand; subsequently, the first cover film 430 and the second cover film 530 are controlled to shrink so as to realize adhesion and ash removal; after that, the transfer trolley 006 is controlled to move, and the adhesive tape body 830 is used for cleaning the other parts of the first cover film 430 and the second cover film 530; when the ash is removed from the other transfer trolley 006, the retraction roller 820 is controlled to rotate, and the exposed adhesive tape 830 is replaced.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The defect detection equipment after cleaning of the rail transit locomotive based on deep learning comprises a detection workshop, a visual detection robot (003) which is a multi-axis mechanical arm provided with a 2D camera and a transfer trolley (006); the method is characterized in that:
the transfer trolley (006) comprises a bearing bottom plate (100), a bearing frame body (200), a rotary telescopic frame (300), a first bearing assembly (400), a second bearing assembly (500) and a paraffin pumping assembly (600);
a plurality of omnidirectional wheels are arranged at the bottom of the bearing bottom plate (100);
the number of the bearing frames (200) is two, and the two bearing frames are symmetrically arranged and are fixed on the side wall of the bearing bottom plate (100);
the rotary telescopic frames (300) are bidirectional electric telescopic rods, are in one-to-one correspondence with the bearing frame bodies (200), and are rotatably connected to the bearing frame bodies (200);
the first bearing component (400) and the second bearing component (500) are identical in structure and symmetrically arranged;
the first support assembly (400) comprises a first carrier plate (410), a first carrier block (420) and a first cover film (430);
the first bearing plate (410) is fixed at one end of the rotary expansion bracket (300) near the edge, and the first bearing block (420) is fixed on the first bearing plate (410);
the first top film (430) is an elastic rubber film, is rectangular in shape, and is fixed on the edge of the surface, far away from the first bearing plate (410), of the first bearing block (420), so that a flat rectangular space is formed by the first top film and the first bearing block (420);
the first bearing block (420) is provided with a heating component and a cooling component;
the paraffin pumping assembly (600) communicates a paraffin storage bin (110) with two rectangular spaces.
2. The deep learning based post-cleaning defect detection device for rail transit locomotives according to claim 1, wherein: the area of the surface of the first bearing block (420) contacted with the first bearing plate (410) is more than 0.45 times of the surface area of the first bearing plate (410).
3. The deep learning based post-cleaning defect detection device for rail transit locomotives according to claim 1, wherein: a warehouse area (001), a charging position (002), a calling feeding area (004) and a buffer position (005) are arranged in the detection workshop; the warehouse area (001) is used for storing accessories and detecting consumable materials; the charging potential (002) is used for charging the transfer trolley (006); the calling loading area (004) is used for stopping the transfer trolley (006) carrying the workpiece (007), and the transfer trolley (006) calling the loading area (004) moves to the vicinity of the visual detection robot (003) at the first time after the detection is completed; the buffer position (005) is the parking stall of transfer dolly (006).
4. The deep learning based post-cleaning defect detection device for rail transit locomotives according to claim 1, wherein: the rotary telescopic frame (300) comprises a base rod (310), a first sliding rod (320) and a second sliding rod (330); the base rod (310) is rotationally connected to the bearing frame body (200), the first sliding rod (320) and the second sliding rod (330) are hard rod bodies, the length direction is the same as the length direction of the base rod (310), the base rod (310) is slidingly positioned, and the base rod and the second sliding rod slide under the synergistic effect of the control unit and the power assembly to enable the rotary telescopic frame (300) to stretch; the sliding directions of the first sliding rod (320) and the second sliding rod (330) are opposite.
5. The deep learning based post-cleaning defect detection device for rail transit locomotives according to claim 1, wherein: the heating component is an electric heating wire; the cooling component is a combination of a liquid cooling pipeline (422), a water pump and a cold water storage bin.
6. The deep learning based post-cleaning defect detection device for rail transit locomotives according to any of claims 1 to 5, wherein: the first bearing block (420) and the second bearing block (520) are internally provided with adsorption magnets (700);
the adsorption magnet (700) is a plate-shaped electromagnet;
after photographing the top surface of the workpiece (007) and resetting the vision inspection robot (003), controlling the first and second support members (400, 500) to be close to each other, and then controlling the attracting magnet (700) in the first support block (420) to be electrified to attract the workpiece (007); controlling the first cover top film (430) to expand first to wrap the ground and part or all of the side surfaces of the workpiece (007); thereafter controlling the increase in the amount of paraffin in the second space such that the second cover film (530) covers the top surface of the work piece (007); controlling the rotation of the rotating telescopic frame (300) by 180 degrees after the paraffin in the second space and the paraffin in the second space is solidified, and controlling the first bearing component (400) and the second bearing component (500) to be far away from each other; then the mechanical arm is controlled to drive the 2D camera to extend into the space between the first bearing component (400) and the second bearing component (500); photographing the original bottom surface and the side surface of one or more workpieces to perform visual detection; thereafter, the vision inspection robot (003) is controlled to reset.
7. The deep learning based post-cleaning defect detection device for rail transit locomotives according to any of claims 1 to 5, wherein: film bottom inserting columns (440) are densely distributed on the surface, close to the first bearing block (420), of the first top film (430); the membrane bottom insertion column (440) is columnar and can be attracted by a magnet;
a limiting elastic net (442) is arranged between the first top film (430) and the first bearing block (420);
the limiting elastic net (442) is a rubber net made of elastic materials, is rectangular overall and is fixed at the edge of the first top film (430);
the limiting elastic net (442) penetrates through all the membrane bottom inserting columns (440) and is fixed on the membrane bottom inserting columns (440) for assisting the membrane bottom inserting columns (440) to be inserted into the inserting holes (443);
the spacing elastic net (442) and the first covering film (430) are spaced more than 0.6 cm apart;
the surface, close to the first top film (430), of the first bearing block (420) is densely provided with insertion holes (443), and the insertion holes (443) are in one-to-one correspondence with the film bottom insertion columns (440) and are used for inserting the film bottom insertion columns (440); the length of the film bottom insertion column (440) is shorter than the depth of the insertion hole (443);
a meshed net body penetrating groove (444) is formed in the surface, close to the first top film (430), of the first bearing block (420);
the first bearing block (420) is internally provided with an adsorption magnet (700), and the adsorption magnet (700) plays a role of adsorbing and fixing the membrane bottom insertion column (440).
8. The deep learning based post-cleaning defect detection device for rail transit locomotives according to claim 7, wherein: the bottom of the membrane bottom insertion column (440) is hemispherical.
9. The deep learning based post-cleaning defect detection device for rail transit locomotives according to claim 7, wherein: the membrane bottom insertion column (440) comprises a foundation column and an iron ring body (441);
the foundation column is a plastic or rubber column and is fixed on the first covering top film (430);
the iron ring body (441) is sleeved and fixed on the foundation column and is 1.5 to 2.5 cm away from the fixed end of the foundation column;
the attracting magnet (700) is positioned within the first carrier block (420) and is 1.5 to 2.5 cm from the face of the first carrier block (420) that is proximate to the first cover top film (430).
10. The deep learning based post-cleaning defect detection device for rail transit locomotives according to any of claims 1 to 5, wherein: the dust-sticking cleaning assembly is also arranged in the detection workshop, and the transfer trolley (006) is cleaned once by the dust-sticking cleaning assembly after each use;
the dust-sticking cleaning assembly comprises a bearing frame (810), a receiving roller (820) and an adhesive tape body (830);
the bearing frame (810) is fixed on the ground and is of a frame structure and is used for bearing and fixing the folding roller (820);
the number of the winding and unwinding rollers (820) is more than 1.5 times of the length of the transfer trolley (006), motors are arranged in the winding and unwinding rollers, and the winding and unwinding rollers are transversely arranged and positioned on the bearing frame (810) and used for winding and unwinding the adhesive tape body (830);
the adhesive tape body (830) is a tape body with two surfaces coated with self-adhesive tapes, and two ends of the tape body are respectively wound and positioned on the two winding and unwinding rollers (820); the width of the adhesive tape body (830) is 0.5 to 0.6 times of the width of the transfer trolley (006).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410022021.7A CN117517327B (en) | 2024-01-08 | 2024-01-08 | Rail transit locomotive cleaning defect detection equipment based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410022021.7A CN117517327B (en) | 2024-01-08 | 2024-01-08 | Rail transit locomotive cleaning defect detection equipment based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117517327A true CN117517327A (en) | 2024-02-06 |
| CN117517327B CN117517327B (en) | 2024-03-15 |
Family
ID=89746127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410022021.7A Active CN117517327B (en) | 2024-01-08 | 2024-01-08 | Rail transit locomotive cleaning defect detection equipment based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117517327B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2934898A1 (en) * | 2008-08-08 | 2010-02-12 | Descamps Ventilation | Piece i.e. axle, controlling installation for high speed train, has control units controlling piece and with guiding system guiding turn to control displacement of turn along vertical plane parallel to longitudinal spindle of piece |
| US20100100275A1 (en) * | 2008-10-22 | 2010-04-22 | Mian Zahid F | Thermal imaging-based vehicle analysis |
| US20140142868A1 (en) * | 2012-11-18 | 2014-05-22 | Andian Technologies Ltd. | Apparatus and method for inspecting track in railroad |
| CN107243622A (en) * | 2017-08-10 | 2017-10-13 | 安徽理工大学 | The removable heavily loaded foundry robot of series parallel type |
| CN208239081U (en) * | 2018-04-04 | 2018-12-14 | 深圳市乾行达科技有限公司 | Car body defect detection equipment |
| KR102008973B1 (en) * | 2019-01-25 | 2019-08-08 | (주)나스텍이앤씨 | Apparatus and Method for Detection defect of sewer pipe based on Deep Learning |
| WO2020156543A1 (en) * | 2019-02-03 | 2020-08-06 | 北京新联铁集团股份有限公司 | Rail transit locomotive and vehicle inspection apparatus and system |
| EP3944981A1 (en) * | 2020-07-27 | 2022-02-02 | ALSTOM Transport Technologies | Device for detecting the degradation of a contact surface of a system for supplying power to a vehicle |
| CN117330498A (en) * | 2023-09-27 | 2024-01-02 | 广西交科集团有限公司 | Concrete bridge crack detecting system based on digital image processing |
-
2024
- 2024-01-08 CN CN202410022021.7A patent/CN117517327B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2934898A1 (en) * | 2008-08-08 | 2010-02-12 | Descamps Ventilation | Piece i.e. axle, controlling installation for high speed train, has control units controlling piece and with guiding system guiding turn to control displacement of turn along vertical plane parallel to longitudinal spindle of piece |
| US20100100275A1 (en) * | 2008-10-22 | 2010-04-22 | Mian Zahid F | Thermal imaging-based vehicle analysis |
| US20140142868A1 (en) * | 2012-11-18 | 2014-05-22 | Andian Technologies Ltd. | Apparatus and method for inspecting track in railroad |
| CN107243622A (en) * | 2017-08-10 | 2017-10-13 | 安徽理工大学 | The removable heavily loaded foundry robot of series parallel type |
| CN208239081U (en) * | 2018-04-04 | 2018-12-14 | 深圳市乾行达科技有限公司 | Car body defect detection equipment |
| KR102008973B1 (en) * | 2019-01-25 | 2019-08-08 | (주)나스텍이앤씨 | Apparatus and Method for Detection defect of sewer pipe based on Deep Learning |
| WO2020156543A1 (en) * | 2019-02-03 | 2020-08-06 | 北京新联铁集团股份有限公司 | Rail transit locomotive and vehicle inspection apparatus and system |
| EP3944981A1 (en) * | 2020-07-27 | 2022-02-02 | ALSTOM Transport Technologies | Device for detecting the degradation of a contact surface of a system for supplying power to a vehicle |
| CN117330498A (en) * | 2023-09-27 | 2024-01-02 | 广西交科集团有限公司 | Concrete bridge crack detecting system based on digital image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117517327B (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN207668706U (en) | Automatic welding equipment | |
| CN105817428B (en) | A kind of bearing detecting device for bearing automatic assembly production line | |
| CN108736088B (en) | Automatic battery pack disassembling system | |
| CN105382368B (en) | A kind of battery miaow head welding equipment | |
| CN110936925A (en) | Multi-degree-of-freedom car washing device and car washing method | |
| CN105973918A (en) | 3D ball detecting platform for industrial CT | |
| CN109761001B (en) | Intelligent and rapid robot carrying device based on machine vision | |
| CN117517327B (en) | Rail transit locomotive cleaning defect detection equipment based on deep learning | |
| CN216508708U (en) | Robot suitable for steel plate edge cleaning, recycling and flaw detection operation before welding | |
| CN103659229A (en) | Assembly line for installing motor shaft clamping springs | |
| CN105427735B (en) | A kind of robot used for teaching work platforms | |
| CN114295715A (en) | Digital fluorescent magnetic powder detection method for welded joint | |
| CN208717439U (en) | A kind of wheel web automatic rust-removing cleaning device | |
| CN213728044U (en) | Photovoltaic module cleans machine people system | |
| CN111855698A (en) | PCB two-side visual efficient detection system and two-side detection method | |
| CN209005156U (en) | A kind of gripping-type basketball pickup robot mobile based on omni-directional wheel | |
| CN211577017U (en) | X-ray flaw detector and running gear thereof | |
| US10897841B2 (en) | Rotary head of a surface mounter | |
| CN107285146A (en) | A kind of retractable cable dolly, cruising inspection system and control method | |
| CN211635137U (en) | Ping-pong ball picking device based on OpenMV | |
| CN209699904U (en) | A kind of tire-mold supply line | |
| WO2021156433A1 (en) | Inspection device and inspection unit | |
| CN206709353U (en) | Screen replacing device and air-conditioning system | |
| CN111855795A (en) | Horizontal magnetic particle flaw detector of large-scale bearing ring | |
| CN116620939B (en) | Polyimide film production detection device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |