CN102120307B - System and method for grinding industrial robot on basis of visual information - Google Patents

System and method for grinding industrial robot on basis of visual information Download PDF

Info

Publication number
CN102120307B
CN102120307B CN2010106033994A CN201010603399A CN102120307B CN 102120307 B CN102120307 B CN 102120307B CN 2010106033994 A CN2010106033994 A CN 2010106033994A CN 201010603399 A CN201010603399 A CN 201010603399A CN 102120307 B CN102120307 B CN 102120307B
Authority
CN
China
Prior art keywords
workpiece
grinding
industrial robot
computer
head
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.)
Active
Application number
CN2010106033994A
Other languages
Chinese (zh)
Other versions
CN102120307A (en
Inventor
区志财
乔红
苏建华
张波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang Institute of Automation of CAS
Original Assignee
Shenyang Institute of Automation of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenyang Institute of Automation of CAS filed Critical Shenyang Institute of Automation of CAS
Priority to CN2010106033994A priority Critical patent/CN102120307B/en
Publication of CN102120307A publication Critical patent/CN102120307A/en
Application granted granted Critical
Publication of CN102120307B publication Critical patent/CN102120307B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention relates to a system and a method for grinding an industrial robot on the basis of visual information. In the system, the image of a workpiece is collected by a video camera; the workpiece is positioned by computer recognition; the movement locus points of a mechanical arm from the current position to a fetching position are planned on line to control the mechanical arm to fetch the workpiece; and then, the revolving speed and the revolving direction of an abrasive belt are controlled by a preset operation step of grinding so as to control the industrial robot to move the workpiece to an appointed position and gesture for grinding and polishing the workpiece. In the system, the video camera collects the image of the workpiece in real time; the workpiece is identified and positioned by the computer to control the mechanical arm to fetch the workpiece; the movement of the industrial robot and the running state of a grinder in the workpiece grinding operation are controlled; the data end of a computer is connected with the data end of the control cabinet of the industrial robot; output locus points controls the industrial robot to move; and the data end of the computer is connected with the data end of the grinder to control the grinder to start and stop and control the revolving speed and the revolving direction of the abrasive belt.

Description

A kind of industrial robot grinding system and method based on visual information
Technical field
The present invention relates to a kind of grinding system, particularly a kind of industrial robot grinding system and method based on visual information.
Background technology
In manufacturing industry, grinding-polishing is one crucial operation, and the quality of grinding has often determined the class of product.Traditional workpiece grinding mainly contains these three kinds of patterns of artificial grinding, special purpose machine tool grinding and Digit Control Machine Tool grinding.Yet artificial grinding workload is big, efficient is lower, and the homogeneity of workpiece processing is relatively poor; The special purpose machine tool versatility is bad, only is fit to produce in batches; The processing cost of Digit Control Machine Tool is higher.In addition, a large amount of dust of polishing departement have caused threat to workers'health.
Industrial robot has the automaticity height, flexibility is good, the operating space is little, can long-play, numerous advantages such as control able to programme, paid attention to by manufacturing industry gradually and use.Utilize robot to carry out the homogeneity that grinding had both guaranteed the product grinding, enhance productivity again, attracted researcher's concern.Document " the robot grinding experiment and the analysis of rough surface " (Ni Xiaobo, Jin Dewen, Zhang Ji river, Chinese mechanical engineering, 2004,15 (22): 1986-1989) analyzed tangential grinding force in the grinding process and the laterally origin cause of formation and the difference of grinding force; Document " based on the robot high-precision grinding modeling of SVM " (Yang Yang etc.; Robot; 2010,32 (2): 278-282) proposed a kind of grinding process modeling method that returns based on SVMs (SVM), improved in the robot grinding process control to stock removal; Document " the robot grinding system that is used for the complex space Machining of Curved Surface " (Hong Yunfei, Li Chengqun, negative ultra, Chinese mechanical engineering, 2006,150-153) a kind of robot grinding system that can be used for the complex space Machining of Curved Surface has been proposed.Patent CN101738981A has proposed a kind of method for grinding based on machine learning, in order to realize high-precision grinding action.These researchs then are to utilize fixture to cooperate the location in the grip operation paying close attention in the point set in the meticulous control of grinding action.
Therefore; How to make up a kind of industrial robot grinding system that utilizes visual information; Detect workpiece to be ground in real time, the tracing point of online planning mechanical paw from current location to next position can improve the flexibility and the intelligent degree of grinding action effectively.
Summary of the invention
In order to improve in the workpiece grinding process, the flexibility of workpiece placement location, grip intelligent the object of the present invention is to provide a kind of industrial robot grinding system and method based on visual information.This system is through the workpiece on video frequency pick-up head detection in real time and the positioning table, and online planning mechanical paw to the tracing point that grasps the some position, is realized the intellectuality of grip from current location, has improved the flexibility of grinding action.This system is according to pre-set grinding action step, the tracing point of the position and attitude of the workpiece grinding of mechanical paw from the current location to the appointment in the online planning grinding operation, and the rotating speed of controlling milling drum in real time with turn to, accomplish the automatic grinding of workpiece.
In order to realize said purpose; The invention provides a kind of industrial robot grinding system based on visual information; Said system comprises industrial robot, computer, mechanical paw, video frequency pick-up head, milling drum, and mechanical paw is installed on industrial robot, is used for grabbing workpiece; Directly over workbench, be fixed with video frequency pick-up head, be used for the workpiece image on the real-time collecting work platform; Graphics processing unit in the computer is connected with video frequency pick-up head through video signal cable; Computer Recognition and location workpiece, control mechanical paw grabbing workpiece, the state of the motion of the industrial robot in the grinding action of control workpiece and the operation of milling drum; The data terminal of computer is connected with the data terminal of the switch board of industrial robot, through the motion of output trajectory point control industrial robot; The data terminal of computer is connected with the data terminal of milling drum, the startup of control milling drum, stops with the rotating speed in abrasive band, turns to.
In order to realize said purpose, second aspect of the present invention provides a kind of industrial robot method for grinding that uses said industrial robot grinding system based on visual information, may further comprise the steps:
Step S1: set up the transformational relation between industrial robot coordinate system and the video frequency pick-up head coordinate system according to grasping calibration algorithm;
Step S2: the initial placing attitude of setting workpiece; Collection is in the image of the workpiece diverse location in the video frequency pick-up head visual field in the initial placing attitude; Utilize the Haar characteristic of traditional Haar feature extraction algorithm extraction workpiece, obtain the Haar features training collection of workpiece diverse location in the video frequency pick-up head visual field;
Step S3: be placed on workpiece on the workbench according to initial placing attitude, video frequency pick-up head is gathered image directly over workbench;
Step S4: the graphics processing unit of computer extracts the Haar characteristic to the workpiece image of input, does coupling to the Haar characteristic of extracting and the Haar characteristic of training set, detects and the location workpiece;
Step S5: the extracting control module of computer is according to the positional information of the workpiece of graphics processing unit output, and the movement locus point of online planning mechanical paw point from the current location to the grip is controlled industrial robot to transport motivation tool paw grabbing workpiece;
Step S6: the grinding action control module of computer is according to predefined grinding step, the rotating speed that the abrasive band of milling drum in the k step grinding is set with turn to, 1≤k≤N, N are the side numbers that workpiece needs grinding;
Step S7: the grinding action control module of computer is according to the position and attitude of workpiece in the predefined k step grinding; On-line computer tool paw goes on foot the movement locus point of grinding action appointed positions attitude from the current location attitude to k; The control robot moves to appointed positions to workpiece; K side of the sbrasive belt grinding workpiece of milling drum, 1≤k≤N;
Step S8: repeating step S6 and step S7, up to all sides that need grinding of workpiece all by grinding;
Step S9: industrial robot is put into appointed positions to the good workpiece of grinding.
Beneficial effect of the present invention: industrial robot, computer, mechanical paw, video frequency pick-up head and milling drum have been formed the industrial robot workpiece grinding system based on visual information.Be installed in the video frequency pick-up head of workbench top, detect and locate workpiece in real time, realize, improved the flexibility of workpiece placement and the intellectuality of grip the grip position choice.The industrial robot grinding system has been realized the automatic grinding to workpiece, has guaranteed the homogeneity of workpiece grinding, has improved the efficient of workpiece grinding.Advantage of the present invention has:
A) under the vision guide, industrial robot identification workpiece calculates the crawl position in computer, compare with traditional teach mode, has improved the flexibility of industrial robot extracting process.
B) serial signal line of milling drum is connected with the data terminal of computer, computer according to each the step grinding action demand, the rotating speed in real-time regulated abrasive band with turn to, guaranteed the quality of grinding.
C) said industrial robot grinding system based on visual information has stronger extensibility on hardware structure, sets up multirobot cooperation grinding system as expanding, or sets up the industrial robot grinding system based on the various visual angles vision.
Description of drawings
Fig. 1 is the pie graph based on the industrial robot grinding system of visual information of the embodiment of the invention;
Fig. 2 is the structural representation of each unit of computer-internal of the embodiment of the invention;
The flow chart based on the industrial robot method for grinding of the grip of visual information of Fig. 3 embodiment of the invention.
Critical piece explanation among the figure:
Industrial robot 1 industrial machine human body 11
Industrial robot switch board 12 computers 2
Graphics processing unit 21 grip control modules 22
Grinding action control module 23 data communication units 24
Serial communication unit 25 mechanical paws 3
Video frequency pick-up head 4 milling drums 5
Abrasive band 51 workbench 6
Workpiece 7 video camera head frames 8
The specific embodiment
Below in conjunction with accompanying drawing practical implementation of the present invention is elaborated.
As shown in Figure 1; Industrial robot grinding system based on visual information comprises industrial robot 1, computer 2, mechanical paw 3, video frequency pick-up head 4, milling drum 5, and said computer 2 comprises graphics processing unit 21, grip control module 22, grinding action control module 23, data communication units 24, serial communication unit 25.Mechanical paw 3 is installed on industrial robot 1, is used for grabbing workpiece; Directly over workbench, be fixed with video frequency pick-up head 4, be used for the workpiece image on the real-time collecting work platform; Graphics processing unit 21 in the computer 2 is connected with video frequency pick-up head 4 through video signal cable; Computer 2 identifications and location workpiece, control mechanical paw 3 grabbing workpieces, the state of the motion of the industrial robot 1 in the grinding action of control workpiece and the operation of milling drum 5; The data terminal of computer 2 is connected with the data terminal of the switch board 12 of industrial robot 1, through the motion of output trajectory point control industrial robot 1; The data terminal of computer 2 is connected with the data terminal of milling drum 5, the startup of control milling drum 5, stops with the rotating speed in abrasive band, turns to.
Industrial robot 1 comprises industrial machine human body 11 and industrial robot switch board 12.Industrial robot 1 is the solid high six-shaft industrial robot or the six-shaft industrial robot of other models.Video frequency pick-up head 4 is the DS-2CZ252P camera or the camera of other model.Milling drum 5 is the double end belt grinding machine or the belt grinding machine of other models.
Milling drum 5 carries frequency converter, has network control function, through the RS485 universal serial bus, the frequency converter of milling drum is connected with the serial communication unit 25 of computer 2.According to the serial order control word of frequency converter frequency adjustment, computer 2 according to required rotating speed with turn to, send corresponding control word to serial communication unit 25, thus tangible abrasive band rotating speed and the adjustment that turns to.Frequency converter is the universal frequency converter FR-E700 of Mitsubishi, the perhaps universal frequency converter of other models
Be depicted as the structural representation of computer 2 inner each unit of the embodiment of the invention like Fig. 2; Wherein: computer 2 comprises graphics processing unit 21, grip control module 22, grinding action control module 23, data communication units 24 and serial communication unit 25; Wherein: graphics processing unit 21 is connected with video frequency pick-up head 4, the image of video frequency pick-up head 4 inputs is carried out the crawl position of image processing, detection workpiece, location workpiece; Grip control module 22 links to each other with graphics processing unit 21, and according to the grip position of graphics processing unit 21 output, planning is from the current location tracing point of grip position on earth; In the grinding action control module 23 the predefined workpiece of storage need N step correspondence of a grinding N side the location of workpiece and attitude, abrasive band 51 rotating speed and turn to; Planning goes on foot the tracing point of the position of grinding from mechanical paw 3 current locations to k; 1≤k≤N, set the grinding of k step the abrasive band rotating speed with turn to; One end of data communication units 24 links to each other with grinding action control module 23 with grip control module 22 respectively, receives the movement locus point of the grip control module 22 and the mechanical paw 3 of grinding action control module 23 outputs; The other end of data communication units 24 is connected with industrial robot 1, to the movement locus point of industrial robot 1 output mechanical paw 3, through the motion of industrial robot 1 control mechanical paw 3; One end of serial communication unit 25 is connected with grinding action control module 23, receives the milling drum control instruction of grinding action control module 23 outputs; The other end of serial communication unit 25 is connected with milling drum 5, converts the control instruction of the milling drum 5 of grinding action control module 23 outputs into serial signal, and the startup of control milling drum 5 stops to turn to the rotating speed in abrasive band 51.
Be depicted as the flow chart of the industrial robot method for grinding of a kind of industrial robot grinding system based on visual information of the embodiment of the invention like Fig. 3, this method for grinding comprises the steps:
Step S1: set up the transformational relation between video frequency pick-up head coordinate system and the industrial robot coordinate system according to grasping calibration algorithm; Before workpiece 7 is grasped grinding action,, set the initial putting position that is suitable for grasping of industrial robot 1 according to the position of workbench 6 and the field range of video frequency pick-up head 4.
Step S2: the initial placing attitude of setting workpiece 7; Collection is in the image of workpiece 7 diverse location in video frequency pick-up head 4 visual fields in the initial placing attitude; Utilize the Haar characteristic of traditional Haar feature extraction algorithm extraction workpiece 7, obtain the Haar features training collection of workpiece 7 diverse location in video frequency pick-up head 4 visual fields; The initial placing attitude of said workpiece is the attitude that makes things convenient for mechanical paw to grasp.
Step S3: be placed on workpiece on the workbench according to initial placing attitude, video frequency pick-up head 4 is gathered the background image that includes workpiece directly over workbench;
Step S4: the workpiece image of 21 pairs of inputs of graphics processing unit of computer 2 extracts the Haar characteristic, does coupling to the Haar characteristic of extracting and the Haar characteristic of training set, detects and the location workpiece;
Step S5: the extracting control module 22 of computer 2 is according to the positional information of the workpiece of graphics processing unit 21 outputs; The movement locus point of online planning mechanical paw 3 point from the current location to the grip, control industrial robot 1 mechanically moving paw 3 grabbing workpieces;
Step S6: the grinding action control module 23 of computer 2 is according to predefined grinding step, the rotating speed that the abrasive band 51 of milling drum 5 in the k step grinding is set with turn to, 1≤k≤N, N are the side numbers that workpiece needs grinding; Preestablish and be the rotating speed in workpiece 7 residing positions and attitude when before graphics processing unit 21 detects workpiece, just setting 7 of workers and needing the order of N side of grinding, each side of grinding, abrasive band 51 and turn to.
Step S7: the grinding action control module 23 of computer 2 is according to the position and attitude of workpiece in the predefined k step grinding; The movement locus point of on-line computer tool paw 3 step grinding action appointed positions attitude from the current location attitude to k; Control robot 1 moves to appointed positions to workpiece; K side of abrasive band 51 grinding work pieces of milling drum 5,1≤k≤N;
Step S8: repeating step S6 and step S7, up to all sides that need grinding of workpiece all by grinding;
Step S9: industrial robot 1 is put into appointed positions to the good workpiece of grinding.
Grasp calibration algorithm and confirm the step (cylinder of a rule of needs) as follows of the transformational relation of industrial robot coordinate system and video frequency pick-up head coordinate system:
Step b1: the cylinder setting is placed on the workbench 6 of a level altitude; Move cylinder to different positions; Video frequency pick-up head 4 is gathered the cylinder image of diverse location; In computer 2, utilize oval extraction algorithm to extract the profile of the cylinder end face of diverse location, and calculate the coordinate Cn of cylinder end face center under video frequency pick-up head 4 coordinate systems of diverse location, the number of times of different position, n >=1 for cylinder moves;
Step b2: operation industrial robot 1 motion; Make the mechanical paw 3 of industrial robot 1 move to the cylinder top; And center line and the center line of mechanical paw 3 of the cylinder end face of each position are overlapped basically; Mechanical paw 3 front ends just contact with the cylinder end face, the pose Pn of industrial robot 1 when computer 2 reads and write down the diverse location cylinder from industrial robot 1 data terminal;
Step b3: according to the first pose P1, the second pose P2, the 3rd pose P3, the 4th pose P4 and the n pose Pn of the corresponding industrial robot 1 of coordinate C1, C2, C3, C4 and the Cn of cylinder end face center under the coordinate system of video frequency pick-up head 2; Setting coordinate and pose is quadric corresponding relation, utilizes the quadratic surface approximating method to simulate the coordinate system of video frequency pick-up head 4 and the transformation relation between industrial robot 1 coordinate system.
The calculation procedure of grip position selection algorithm is (with the tap is embodiment, and the crawl position is the circular hole of the water inlet end of tap) as follows:
Step c1: the gray-scale map modular converter of tap crawl position selection algorithm is to contain the tap gray-scale map of having powerful connections with the tap image transitions of having powerful connections that contains of the data terminal of video frequency pick-up head 4 input in computer 2;
Step c2: the canny edge extracting module of tap crawl position selection algorithm is carried out edge extracting to containing the tap gray-scale map of having powerful connections;
Step c3: the oval detection algorithm of tap crawl position selection algorithm carries out match to the edge of Canny operator extraction, obtains all ellipses in the image.
Step c4: the haar characteristic that the Haar feature extraction algorithm of tap crawl position selection algorithm is oval in advance;
Step c5: Haar characteristic that each that will obtain is oval and the Haar characteristic matching in the training set, judge whether ellipse is the circular hole of the water inlet end of tap.
Step c6: according to the oval information that coupling obtains, calculate oval center, this center is exactly the crawl position of tap.
The step of the movement locus point of on-line computer tool paw from current location to next position is following:
Steps d 1:, be transformed into current location point and next location point in the joint coordinate system of industrial robot according to the joint coordinate system of industrial robot and the transformational relation of cartesian coordinate system;
Steps d 2: in joint coordinate system, according to " crossing the cubic polynomial interpolation algorithm of path point ", the movement locus point of calculating machine paw.
The above; Be merely the specific embodiment among the present invention, but protection scope of the present invention is not limited thereto, anyly is familiar with this technological people in the technical scope that the present invention disclosed; Can understand conversion or the replacement expected, all should be encompassed within the protection domain of claims of the present invention.

Claims (5)

1. industrial robot grinding system based on visual information; It is characterized in that; System comprises industrial robot (1), computer (2), mechanical paw (3), video frequency pick-up head (4), milling drum (5), and mechanical paw (3) is installed on industrial robot (1), is used for grabbing workpiece; Directly over workbench, be fixed with video frequency pick-up head (4), be used for the workpiece image on the real-time collecting work platform; Graphics processing unit (21) in the computer (2) is connected with video frequency pick-up head (4) through video signal cable; Computer (2) identification and location workpiece, control mechanical paw (3) grabbing workpiece, the state of the motion of the industrial robot (1) in the grinding action of control workpiece and the operation of milling drum (5); The data terminal of computer (2) is connected with the data terminal of the switch board (12) of industrial robot (1), through the motion of output trajectory point control industrial robot (1); The data terminal of computer (2) is connected with the data terminal of milling drum (5), the startup of control milling drum (5), stops with the rotating speed in abrasive band, turns to;
Computer (2) comprises graphics processing unit (21), grip control module (22), grinding action control module (23), data communication units (24) and serial communication unit (25); Wherein: graphics processing unit (21) is connected with video frequency pick-up head (4), the image of video frequency pick-up head (4) input is carried out the crawl position of image processing, detection workpiece, location workpiece; Grip control module (22) links to each other with graphics processing unit (21), according to the grip position of graphics processing unit (21) output, and the tracing point of planning from current location to the grip position; In the grinding action control module (23) the predefined workpiece of storage need N step correspondence of a grinding N side the location of workpiece and attitude, abrasive band (51) rotating speed and turn to; Planning goes on foot the tracing point of the position of grinding from mechanical paw (3) current location to k; 1≤k≤N, set the grinding of k step the abrasive band rotating speed with turn to; One end of data communication units (24) links to each other with grinding action control module (23) with grip control module (22) respectively, receives the movement locus point of the mechanical paw (3) of grip control module (22) and grinding action control module (23) output; The other end of data communication units (24) is connected with industrial robot (1), to the movement locus point of industrial robot (1) output mechanical paw (3), through the motion of industrial robot (1) control mechanical paw (3); One end of serial communication unit (25) is connected with grinding action control module (23), receives the milling drum control instruction of grinding action control module (23) output; The other end of serial communication unit (25) is connected with milling drum (5), converts the control instruction of the milling drum (5) of grinding action control module (23) output into serial signal, and the startup of control milling drum (5) stops and the rotating speed of abrasive band (51) turns to.
2. industrial robot method for grinding that uses the said industrial robot grinding system based on visual information of claim 1 is characterized in that may further comprise the steps:
Step S1: set up the transformational relation between industrial robot coordinate system and the video frequency pick-up head coordinate system according to grasping calibration algorithm;
Step S2: the initial placing attitude of setting workpiece; Collection is in the image of the workpiece diverse location in the video frequency pick-up head visual field in the initial placing attitude; Utilize the Haar characteristic of traditional Haar feature extraction algorithm extraction workpiece, obtain the Haar features training collection of workpiece diverse location in the video frequency pick-up head visual field;
Step S3: be placed on workpiece on the workbench according to initial placing attitude, video frequency pick-up head is gathered image directly over workbench;
Step S4: the graphics processing unit of computer extracts the Haar characteristic to the workpiece image of input, does coupling to the Haar characteristic of extracting and the Haar characteristic of training set, detects and the location workpiece;
Step S5: the grip control module of computer is according to the positional information of the workpiece of graphics processing unit output, and the movement locus point of online planning mechanical paw point from the current location to the grip is controlled industrial robot to transport motivation tool paw grabbing workpiece;
Step S6: the grinding action control module of computer is according to predefined grinding step, the rotating speed that the abrasive band of milling drum in the k step grinding is set with turn to, 1≤k≤N, N are the side numbers that workpiece needs grinding;
Step S7: the grinding action control module of computer is according to the position and attitude of workpiece in the predefined k step grinding; On-line computer tool paw goes on foot the movement locus point of grinding action appointed positions attitude from the current location attitude to k; The control robot moves to appointed positions to workpiece; K side of the sbrasive belt grinding workpiece of milling drum, 1≤k≤N;
Step S8: repeating step S6 and step S7, up to all sides that need grinding of workpiece all by grinding;
Step S9: industrial robot is put into appointed positions to the good workpiece of grinding;
Said employing is grasped calibration algorithm and is confirmed that the step of transformational relation of industrial robot coordinate system and video frequency pick-up head coordinate system is following:
Step Sb1: the cylinder setting is placed on the workbench of a level altitude; Move cylinder to different positions; Video frequency pick-up head is gathered the cylinder image of diverse location; In computer, utilize oval extraction algorithm to extract the profile of the cylinder end face of diverse location, and calculate the coordinate Cn of cylinder end face center under the video frequency pick-up head coordinate system of diverse location, the number of times of different position, n >=1 for cylinder moves;
Step Sb2: operation industrial robot motion; Make the mechanical paw of industrial robot move to the cylinder top; And center line and the center line of mechanical paw of the cylinder end face of each position are overlapped basically; The mechanical paw front end just contacts with the cylinder end face, the pose Pn of computer industrial robot when the data terminal of industrial robot reads and write down the diverse location cylinder;
Step Sb3: according to the first pose P1, the second pose P2, the 3rd pose P3, the 4th pose P4 and the n pose Pn of the corresponding industrial robot of coordinate C1, C2, C3, C4 and the Cn of cylinder end face center under the coordinate system of video frequency pick-up head; Setting coordinate and pose is quadric corresponding relation, utilizes the quadratic surface approximating method to simulate the coordinate system of video frequency pick-up head and the transformation relation between the industrial robot coordinate system.
3. industrial robot method for grinding according to claim 2; It is characterized in that said preestablishing is the rotating speed in the residing position of workpiece and attitude when before graphics processing unit detects workpiece, just setting workpiece and needing the order of N side of grinding, each side of grinding, abrasive band and turns to.
4. industrial robot method for grinding according to claim 2 is characterized in that, the initial placing attitude of said workpiece is the attitude that makes things convenient for mechanical paw to grasp.
5. industrial robot method for grinding according to claim 2 is characterized in that, the step of the movement locus point of on-line computer tool paw from current location to next position is following:
Steps d 1:, be transformed into current location point and next location point in the joint coordinate system of industrial robot according to the joint coordinate system of industrial robot and the transformational relation of cartesian coordinate system;
Steps d 2: in joint coordinate system, according to the cubic polynomial interpolation algorithm of crossing path point, the movement locus point of calculating machine paw.
CN2010106033994A 2010-12-23 2010-12-23 System and method for grinding industrial robot on basis of visual information Active CN102120307B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2010106033994A CN102120307B (en) 2010-12-23 2010-12-23 System and method for grinding industrial robot on basis of visual information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2010106033994A CN102120307B (en) 2010-12-23 2010-12-23 System and method for grinding industrial robot on basis of visual information

Publications (2)

Publication Number Publication Date
CN102120307A CN102120307A (en) 2011-07-13
CN102120307B true CN102120307B (en) 2012-07-04

Family

ID=44249023

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2010106033994A Active CN102120307B (en) 2010-12-23 2010-12-23 System and method for grinding industrial robot on basis of visual information

Country Status (1)

Country Link
CN (1) CN102120307B (en)

Families Citing this family (48)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102564319B (en) * 2011-12-30 2014-03-12 清华大学 Method for detecting slip during linear delivery of wafer by using image processing technology
CN102632458A (en) * 2012-04-19 2012-08-15 浙江工业大学 Soft fixed grain air-pressure grinding wheel optical-finishing system with damage detection
JP5962242B2 (en) * 2012-06-15 2016-08-03 日本精工株式会社 Grinding equipment
CN102962749B (en) * 2012-11-14 2015-12-02 天津中屹铭科技有限公司 A kind of full-automatic grinding robot
CN103056759B (en) * 2012-12-24 2015-01-28 中国科学院自动化研究所 Robot grinding system based on feedback of sensor
CN103639867A (en) * 2013-12-06 2014-03-19 苏州逸美德自动化科技有限公司 Robot polisher
CN103692320B (en) * 2013-12-31 2016-05-11 深圳先进技术研究院 A kind of method and apparatus of realizing six axle polishing grinding machinery arm off-line programings
CN104907914A (en) * 2014-03-11 2015-09-16 张家港鹿鸣生物质能源设备有限公司 Polishing device for mold board of biomass granulator
CN103831695B (en) * 2014-03-28 2016-03-02 中国科学院自动化研究所 Large-scale free form surface robot polishing system
CN104972362B (en) * 2014-04-14 2017-10-31 沈阳远大科技园有限公司 Intelligent Force man-controlled mobile robot grinding system and method
CN104385282B (en) * 2014-08-29 2017-04-12 暨南大学 Visual intelligent numerical control system and visual measuring method thereof
CN104626169B (en) * 2014-12-24 2017-03-22 四川长虹电器股份有限公司 Robot part grabbing method based on vision and mechanical comprehensive positioning
CN105159231B (en) * 2015-07-22 2018-02-16 中国地质大学(武汉) A kind of on-line calculation method of digital control system double end machining locus optimization
CN105171595A (en) * 2015-08-13 2015-12-23 浙江勤龙机械科技股份有限公司 Full-automatic disc grinder
CN105196148B (en) * 2015-09-25 2017-08-08 广东省自动化研究所 A kind of adaptive polishing grinding system with intelligent loading and unloading function
CN105643399B (en) * 2015-12-29 2018-06-26 沈阳理工大学 The complex-curved automatic grinding-polishing system of robot and processing method based on Shared control
CN105479461A (en) * 2016-01-26 2016-04-13 先驱智能机械(深圳)有限公司 Control method, control device and manipulator system
CN106181699A (en) * 2016-07-21 2016-12-07 广东意戈力智能装备有限公司 A kind of flexible polishing grinding system
CN106736904B (en) * 2016-12-03 2019-03-12 浙江博来工具有限公司 A kind of intelligence packaged type optical axis milling robot
CN106774144B (en) * 2016-12-21 2019-04-12 上海华括自动化工程有限公司 A kind of intelligent CNC processing method based on industrial robot
CN106584273B (en) * 2017-01-31 2019-02-15 西北工业大学 A kind of online vision detection system for robot polishing
CN106695483B (en) * 2017-02-09 2018-12-21 泉州市科恩智能装备技术研究院有限公司 The precise positioning grinding device of fire-fighting pipette tips
CN107052950B (en) * 2017-05-25 2018-10-12 上海莫亭机器人科技有限公司 A kind of complex-curved sanding and polishing system and method
CN109129039B (en) * 2017-06-16 2019-12-13 上海铼钠克数控科技股份有限公司 system and method for sharpening a tool
CN107139003A (en) * 2017-06-27 2017-09-08 巨轮(广州)机器人与智能制造有限公司 Modularization vision system preparation method
CN107598775A (en) * 2017-07-27 2018-01-19 芜湖微云机器人有限公司 It is a kind of by laser detect the method that simultaneously multi-axis robot is polished
CN107738254B (en) * 2017-08-25 2019-12-24 中国科学院光电研究院 Conversion calibration method and system for mechanical arm coordinate system
CN107443212A (en) * 2017-08-31 2017-12-08 苏州永博电气有限公司 A kind of triangular drawbench of the brush processing with visual identity function
CN107471028B (en) * 2017-09-07 2019-05-07 中车唐山机车车辆有限公司 Grinding method and system
CN107617943A (en) * 2017-09-27 2018-01-23 安徽合力股份有限公司合肥铸锻厂 A kind of balance weight iron robot sanding apparatus and its control system
CN107553329A (en) * 2017-10-25 2018-01-09 天津通卓机器人有限公司 A kind of milling robot
CN107877366A (en) * 2017-11-03 2018-04-06 天津市三木森电炉股份有限公司 A kind of grinding workpieces automatic identification detecting system
CN107877382A (en) * 2017-11-24 2018-04-06 广东惠利普路桥信息工程有限公司 A kind of glass edge-grinding machine auto-feed control method
CN108226164B (en) * 2017-12-29 2021-05-07 深圳市智能机器人研究院 Robot polishing method and system based on visual detection
CN108805931A (en) * 2018-06-08 2018-11-13 歌尔股份有限公司 Position finding and detection method, device and the computer readable storage medium of AR products
CN108994830A (en) * 2018-07-12 2018-12-14 上海航天设备制造总厂有限公司 System calibrating method for milling robot off-line programing
TWI675719B (en) * 2018-08-07 2019-11-01 由田新技股份有限公司 Three-dimensional processing unit, processing system and processing method for irregular surfaces
CN109352512A (en) * 2018-12-04 2019-02-19 苏州丰川电子科技有限公司 Metal surface automatic machining device
CN109623567A (en) * 2018-12-10 2019-04-16 苏州丰川电子科技有限公司 Notebook computer shell machine for automatic working structure
CN109605189A (en) * 2018-12-10 2019-04-12 苏州丰川电子科技有限公司 Full-automatic accurate grinding device
CN109605212A (en) * 2018-12-10 2019-04-12 苏州丰川电子科技有限公司 High-precision working apparatus for electronic equipment casing
CN109590815B (en) * 2018-12-12 2020-05-29 上海卫星装备研究所 Intelligent polishing system, method and computer readable storage medium
CN110216530B (en) * 2019-07-18 2020-11-24 佛山市高明金石建材有限公司 Plate grinding device and grinding method
CN110552118A (en) * 2019-09-30 2019-12-10 琦星智能科技股份有限公司 cloth suction manipulator based on industrial robot vision and fabric grabbing method
CN111007818A (en) * 2019-12-13 2020-04-14 金星阀门有限公司 Valve production system based on machine vision and intelligent robot
CN111230862B (en) * 2020-01-10 2021-05-04 上海发那科机器人有限公司 Handheld workpiece deburring method and system based on visual recognition function
CN111923053A (en) * 2020-04-21 2020-11-13 广州里工实业有限公司 Industrial robot object grabbing teaching system and method based on depth vision
CN111823100A (en) * 2020-07-24 2020-10-27 前元运立(北京)机器人智能科技有限公司 Robot-based small-curvature polishing and grinding method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101396807A (en) * 2008-10-30 2009-04-01 上海大学 Cam grinding control system and method based on PC and infrared thermal imaging system
CN101738981A (en) * 2009-12-04 2010-06-16 清华大学 Machine learning-based robot grinding method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3753133B2 (en) * 2003-04-25 2006-03-08 マツダ株式会社 NC machining data generation method for machine tools

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101396807A (en) * 2008-10-30 2009-04-01 上海大学 Cam grinding control system and method based on PC and infrared thermal imaging system
CN101738981A (en) * 2009-12-04 2010-06-16 清华大学 Machine learning-based robot grinding method

Non-Patent Citations (13)

* Cited by examiner, † Cited by third party
Title
JP特开2004-322265A 2004.11.18
孔令富
孔令富;李晓辉;李林.基于双目主动视觉监测平台的刀具匹配方法.《计算机仿真》.2010,第27卷(第10期),290-293. *
徐敏
朱建华
李剑峰
李剑峰;朱建华;汤青;陈力;徐敏;郭永康.机器人—三维扫描系统联合扫描及其工业应用.《光电工程》.2007,第34卷(第2期),15-21. *
李晓辉
李林.基于双目主动视觉监测平台的刀具匹配方法.《计算机仿真》.2010,第27卷(第10期),290-293.
汤青
稻叶善治.机床和智能机器人的融合.《世界制造技术与装备市场》.2009,(第4期),37-41. *
郭永康.机器人—三维扫描系统联合扫描及其工业应用.《光电工程》.2007,第34卷(第2期),15-21.
陈力

Also Published As

Publication number Publication date
CN102120307A (en) 2011-07-13

Similar Documents

Publication Publication Date Title
CN102120307B (en) System and method for grinding industrial robot on basis of visual information
CN105397244B (en) A kind of pile end board multistation robot Intelligent welding system
CN106909216B (en) Kinect sensor-based humanoid manipulator control method
CN105184019A (en) Robot grabbing method and system
CN103707299A (en) Method of implementing real-time bending follow of bending robot
CN109623656B (en) Mobile double-robot cooperative polishing device and method based on thickness online detection
CN106853639A (en) A kind of battery of mobile phone automatic assembly system and its control method
CN205032395U (en) Robot vision sorts equipment
CN105302959A (en) Offline programming method for six-axis grinding and polishing industrial robot
CN203508417U (en) Sorting system of industrial robot
CN104626169A (en) Robot part grabbing method based on vision and mechanical comprehensive positioning
CN103350421A (en) Automatic glaze spraying controlling method and controlling device for simulating skilled worker operation
CN108858193B (en) Mechanical arm grabbing method and system
CN102830798A (en) Mark-free hand tracking method of single-arm robot based on Kinect
CN105500370A (en) Robot offline teaching programming system and method based on somatosensory technology
CN110666801A (en) Grabbing industrial robot for matching and positioning complex workpieces
CN102126221B (en) Method for grabbing object by mechanical hand based on image information
Djajadi et al. A model vision of sorting system application using robotic manipulator
CN107671838A (en) Robot teaching record system, the processing step and its algorithm flow of teaching
CN206484559U (en) Robot teaching system based on three-dimensional force sensor
CN111515945A (en) Control method, system and device for mechanical arm visual positioning sorting and grabbing
Duţă et al. Control and decision-making process in disassembling used electronic products
CN205121556U (en) Robot grasping system
Ge Programming by demonstration by optical tracking system for dual arm robot
CN109093599A (en) A kind of crawl teaching method and system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant