CN101968343B - Method for detecting automobile door based on robot monocular vision measurement - Google Patents
Method for detecting automobile door based on robot monocular vision measurement Download PDFInfo
- Publication number
- CN101968343B CN101968343B CN2010102866405A CN201010286640A CN101968343B CN 101968343 B CN101968343 B CN 101968343B CN 2010102866405 A CN2010102866405 A CN 2010102866405A CN 201010286640 A CN201010286640 A CN 201010286640A CN 101968343 B CN101968343 B CN 101968343B
- Authority
- CN
- China
- Prior art keywords
- car door
- measured
- point
- standard
- workpiece
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The invention discloses a method for detecting an automobile door based on robot monocular vision measurement, comprising the following steps of: (1) positioning a standard automobile door in a positioning workpiece, setting a plurality of measured points on the edges of the standard automobile door and a measured automobile door and setting a plurality of mark points on the positioning workpiece; (2) solving the three-dimensional coordinates of the first mark point; (3) solving the three-dimensional coordinates of the first measured point; (4) calculating the gap distance between the first mark point on the positioning workpiece and the first measured point on the standard automobile door; (5) repeating the steps (2)-(4) to accomplish the measurement on the gap distances among all mark points on the positioning workpiece and all measured points on the standard automobile door; and (6) replacing the standard automobile door, fixing the measured automobile door in the position of the standard automobile door to accomplish the measurement on the gap distances among the mark points on the positioning workpiece and the measured points on the standard automobile door. By adopting the method, the measuring precision can reach 0.18 mm.
Description
Technical field
The present invention relates to a kind of detection method of automobile cubing, the invention particularly relates to the detection method of the arrangements for automotive doors of measuring based on robot monocular vision.
Background technology
Auto parts must detect it before the entrucking after punching press or injection moulding are come out, and carry out by means of cubing and detect Chang Bixu.Cubing (Checking Fixture) is just causing the great attention of more and more enterprises as the important auxiliary products of die industry.Production from self needs and its higher added value, many mould enterprises with it as an important developing direction.The correlation technique of cubing as a kind of new technology, is also just introduced by increasing mould enterprise.
The automobile cubing is by detection architecture location structure, clamping structure, supporting construction, base, and other slave parts constitute.Existing design is according to concrete part, according to inspection requirements, one by one, design corresponding structure progressively, and then is assembled together, and finally realizes the exploitation of cubing.This is a complicacy and loaded down with trivial details process, and each design all will start anew, and is difficult to reuse previous design achievement; Can not directly bring utilization to expert's mature experience; Cause duplication of labour amount big, design loaded down with trivial details, to the exploitation designer require problems such as height.
At present, machine vision metrology has obtained development apace, makes vision sensor make the exploitation of cubing that great development arranged to the measurement of putting on the part through robot movement, has improved the flexibility of cubing design.Present detection method that adopts usually based on this technology moves through the computer-controlled robot then and measures each measured point for camera is contained on robot end's the flange, directly obtains the D coordinates value of measured point through coordinate transform.Yet, commercial now most of industrial robots, its positional precision is directly measured for robot and bring bigger error than the low one magnitude of its repeatable accuracy.
Summary of the invention
The objective of the invention is to overcome the deficiency of prior art, the detection method of the arrangements for automotive doors of measuring based on robot monocular vision of a kind of flexibility that has improved cubing and automaticity is provided.
The detection method of the arrangements for automotive doors of measuring based on robot monocular vision of the present invention, it may further comprise the steps:
(1) the standard car door is positioned in the location workpiece, sets a plurality of measured points at the edge of standard car door and tested car door and on the workpiece inwall of location, set a plurality of gauge points, set up the stereoscopic vision measuring system and demarcate the inside and outside parameter of camera; The method that standard car door and tested car door measured point are confirmed is to paste a gauge point in the same position of standard car door and tested car door earlier, and every then separated 3cm CW is pasted a series of measured points in vehicle door edge;
(2) computer-controlled robot move to first gauge point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first gauge point solves the three dimensional space coordinate of first gauge point;
(3) computer-controlled robot move to first measured point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first measured point solves the three dimensional space coordinate of first measured point;
(4) space that calculates location first gauge point on the workpiece and first measured point on the standard car door through formula apart from and deposit computing machine in;
(5) repeat described step (2)-(4) and accomplish the gauge points all on the workpiece of location and the space Determination of distance of all measured points on the standard car door;
(6) the standard car door is changed; Tested car door is fixed on the position of standard car door, accomplishes the gauge point of locating on the workpiece and the space Determination of distance of the measured point on the tested car door with the same trajectories of the mensuration of a plurality of gauge points of accomplishing the standard car door and measured point mobile robot successively according to step (2)-(5) then;
(7) the space distance of standard of comparison car door and tested car door and location workpiece is to judge the whether accords with production requirement of tested car door.
Adopt the beneficial effect of the inventive method to be: this device has adopted the distance error model, and the workpiece in a location of measured piece placed around checks through the distance of putting on point and the measured piece on the measurement and positioning workpiece whether product is qualified.Obtain its measuring accuracy through experiment and can arrive 0.18mm.The present technique scheme adopts based on the robotic vision measuring system, has improved the recycling of automobile cubing, has noncontact, fast, the flexible good and high outstanding advantage of automaticity of speed, has practiced thrift the inspection cost of auto parts greatly.
Description of drawings
Fig. 1 is the mensuration fundamental diagram that adopts the detection method examination criteria car door of the arrangements for automotive doors of measuring based on robot monocular vision of the present invention;
Fig. 2 adopts the detection method of the arrangements for automotive doors of measuring based on robot monocular vision of the present invention to detect the mensuration fundamental diagram of tested car door.
Embodiment
Below in conjunction with concrete embodiment, and, the present invention is done further explanation with reference to accompanying drawing:
The detection method of the arrangements for automotive doors of measuring based on robot monocular vision at first is positioned at the standard car door in the location workpiece, and set a plurality of measured points (like A, B) (method confirmed of standard car door and tested car door measured point can be to paste a gauge point in the same position of standard car door and tested car door earlier at the edge of standard car door and tested car door; Every then separated 3cm CW is pasted a series of measured points in vehicle door edge) and on the workpiece inwall of location, set a plurality of gauge points (like a, b); (setting up stereoscopic vision measuring system method can edit referring to Zhang Guangjun: " photoelectricity test technology ", China Measuring Press, 2008 to set up the stereoscopic vision measuring system; 304-307) and demarcate camera inner parameter and external parameter; The inside calibrating parameters of camera comprises (effective focal length f, the Δ x that comprehensively distorts, Δ y; Image planes center (Cx; Cy)), external parameter comprises (translation and rotation matrix have reflected that the three-dimensional world coordinate is tied to the transformational relation of camera coordinate system).The standard car door can adopt " 3-2-1 " positioning principle to fix.Mensuration of at first carrying out the standard car door as shown in Figure 1: the computer-controlled robot move to first gauge point near; Write down the three dimensional space coordinate of the robot of this moment, the image of first gauge point of camera shooting solves the three dimensional space coordinate of first gauge point; The computer-controlled robot move to first measured point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first measured point solves the three dimensional space coordinate of first measured point; Calculate the space distance of first measured point on location first gauge point on the workpiece and the mark standard car door and deposit computing machine in through formula
; Repeat the space Determination of distance that all measured points on gauge points all on the workpiece of location and the mark standard car door are accomplished in described step (2)-(4); First gauge point coordinate method for solving (can adopt the pin-hole imaging model); Be generally projection model and calibrating parameters according to video camera; To the image calibration that distorts, go out the three dimensional space coordinate of first gauge point again by stereo vision sensor three-dimensional measurement model solution respectively.Mensuration of carrying out tested car door as shown in Figure 2 then: the standard car door is changed; Tested car door is fixed on the position of standard car door, accomplishes the gauge point of locating on the workpiece and the space Determination of distance of the measured point on the tested car door with the same trajectories of the mensuration of a plurality of gauge points of accomplishing the standard car door and measured point mobile robot successively according to the assay method of standard car door then; The space distance of last standard of comparison car door and tested car door and location workpiece is to judge the whether accords with production requirement of tested car door.
(1) the standard car door is positioned in the location workpiece, sets a plurality of measured points at the edge of standard car door and tested car door and on the workpiece inwall of location, set a plurality of gauge points, set up the stereoscopic vision measuring system and demarcate the inside and outside parameter of camera;
(2) computer-controlled robot move to first gauge point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first gauge point solves the three dimensional space coordinate of first gauge point;
(3) computer-controlled robot move to first measured point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first measured point solves the three dimensional space coordinate of first measured point;
(4) space that calculates location first gauge point on the workpiece and first measured point on the standard car door through formula apart from and deposit computing machine in;
(5) repeat described step (2)-(4) and accomplish the gauge points all on the workpiece of location and the space Determination of distance of all measured points on the standard car door;
(6) the standard car door is changed; Tested car door is fixed on the position of standard car door, accomplishes the gauge point of locating on the workpiece and the space Determination of distance of the measured point on the tested car door with the same trajectories of the mensuration of a plurality of gauge points of accomplishing the standard car door and measured point mobile robot successively according to step (2)-(5) then;
(7) the space distance of standard of comparison car door and tested car door and location workpiece is to judge the whether accords with production requirement of tested car door.
The three-dimensional coordinate of the gauge point in the workpiece of location is (x
a, y
a, z
a), the three-dimensional coordinate of standard car door measured point is (x
A, y
A, z
A); The three-dimensional coordinate of the gauge point of location in the workpiece be (x '
a, y '
a, z '
a). the three-dimensional coordinate of tested car door be (x '
A, y '
A, z '
A); Distance between distance between more tested car door and the location workpiece and standard car door and the location workpiece, two distances equate then accords with production requirement.
Claims (1)
1. the detection method of the arrangements for automotive doors of measuring based on robot monocular vision is characterized in that it may further comprise the steps:
(1) the standard car door is positioned in the location workpiece, sets a plurality of measured points at the edge of standard car door and tested car door and on the workpiece inwall of location, set a plurality of gauge points, set up the stereoscopic vision measuring system and demarcate the inside and outside parameter of camera; The method that standard car door and tested car door measured point are confirmed is to paste a gauge point in the same position of standard car door and tested car door earlier, and every then separated 3cm CW is pasted a series of measured points in vehicle door edge;
(2) computer-controlled robot move to first gauge point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first gauge point solves the three dimensional space coordinate of first gauge point;
(3) computer-controlled robot move to first measured point near, write down the three dimensional space coordinate of robot of this moment, the image that camera is taken first measured point solves the three dimensional space coordinate of first measured point;
(4) space that calculates location first gauge point on the workpiece and first measured point on the standard car door through formula apart from and deposit computing machine in;
(5) repeat described step (2)-(4) and accomplish the gauge points all on the workpiece of location and the space Determination of distance of all measured points on the standard car door;
(6) the standard car door is changed; Tested car door is fixed on the position of standard car door, accomplishes the gauge point of locating on the workpiece and the space Determination of distance of the measured point on the tested car door with the same trajectories of the mensuration of a plurality of gauge points of accomplishing the standard car door and measured point mobile robot successively according to step (2)-(5) then;
(7) the space distance of standard of comparison car door and tested car door and location workpiece is to judge the whether accords with production requirement of tested car door.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102866405A CN101968343B (en) | 2010-09-19 | 2010-09-19 | Method for detecting automobile door based on robot monocular vision measurement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102866405A CN101968343B (en) | 2010-09-19 | 2010-09-19 | Method for detecting automobile door based on robot monocular vision measurement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101968343A CN101968343A (en) | 2011-02-09 |
| CN101968343B true CN101968343B (en) | 2012-07-04 |
Family
ID=43547533
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010102866405A Expired - Fee Related CN101968343B (en) | 2010-09-19 | 2010-09-19 | Method for detecting automobile door based on robot monocular vision measurement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101968343B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103438803B (en) * | 2013-09-18 | 2015-10-28 | 苏州晓创光电科技有限公司 | Computer vision technique accurately measures the method for Rectangular Parts size across visual field |
| CN105424012B (en) * | 2015-11-26 | 2017-11-03 | 广州机械科学研究院有限公司 | A kind of robot and the caliberating device and scaling method of machine coordinates relation |
| CN105773613B (en) * | 2016-03-30 | 2017-09-19 | 东莞市速美达自动化有限公司 | A kind of horizontal machine people camera coordinates system calibrating method |
| CN109945782B (en) * | 2019-04-02 | 2020-12-08 | 易思维（杭州）科技有限公司 | Method for detecting key position of super-long body-in-white |
| CN111707207A (en) * | 2020-06-22 | 2020-09-25 | 福州云睿自动化设备有限公司 | High-precision robot three-coordinate electronic gauge rapid measurement system and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6205243B1 (en) * | 1996-03-21 | 2001-03-20 | Viewpoint Corp. | System and method for rapid shape digitizing and adaptive mesh generation |
| CN1605827A (en) * | 2004-11-11 | 2005-04-13 | 天津大学 | Multifunctional vision sensor with binocular spatial structure |
| CN101334276A (en) * | 2007-06-27 | 2008-12-31 | 中国科学院自动化研究所 | Visual sense measurement method and device |
| CN101509878A (en) * | 2009-03-09 | 2009-08-19 | 北京航空航天大学 | Part vision detection device |
-
2010
- 2010-09-19 CN CN2010102866405A patent/CN101968343B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6205243B1 (en) * | 1996-03-21 | 2001-03-20 | Viewpoint Corp. | System and method for rapid shape digitizing and adaptive mesh generation |
| CN1605827A (en) * | 2004-11-11 | 2005-04-13 | 天津大学 | Multifunctional vision sensor with binocular spatial structure |
| CN101334276A (en) * | 2007-06-27 | 2008-12-31 | 中国科学院自动化研究所 | Visual sense measurement method and device |
| CN101509878A (en) * | 2009-03-09 | 2009-08-19 | 北京航空航天大学 | Part vision detection device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101968343A (en) | 2011-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101949687B (en) | Detection method of automobile door based on vision measurement | |
| CN101957176B (en) | Robot and camera combined calibration based method for detecting high-precision car door | |
| CN101968343B (en) | Method for detecting automobile door based on robot monocular vision measurement | |
| Yin et al. | Development and calibration of an integrated 3D scanning system for high-accuracy large-scale metrology | |
| US8019147B2 (en) | Three-dimensional data registration method for vision measurement in flow style based on double-sided target | |
| Du et al. | Online robot calibration based on vision measurement | |
| US8520067B2 (en) | Method for calibrating a measuring system | |
| CN105806251A (en) | Four-axis measuring system based on line laser sensor and measuring method thereof | |
| CN102607502B (en) | Automatic detection device and method for size of automobile rear axle assembly | |
| EP3431920A1 (en) | Device, system, and method for inspecting crankshaft shape | |
| US11511435B2 (en) | Robot-conveyor calibration method, robot system and control system | |
| CN206683593U (en) | Measured workpiece dimensional gaughing equipment based on 3D machine vision | |
| CN105403156A (en) | Three-dimensional measuring device and data fusion calibration method for three-dimensional measuring device | |
| CN106660181A (en) | System and method for placing an object, which is to be processed, on a production apparatus in a precise position | |
| KR20020095122A (en) | Calibration and Setting Method for Sensors in an Assembly Station and an Assembly Station | |
| CN103557813B (en) | The on-line measuring device of axial workpiece | |
| Andersen et al. | Fast calibration of industrial mobile robots to workstations using QR codes | |
| CN102602425B (en) | Locomotive limiting system and calibration method thereof | |
| CN103465246A (en) | Rough casting marking-off method and marking-off device | |
| CN104374318A (en) | Method for measuring surface of large mold through laser scanner | |
| CN112828878B (en) | Three-dimensional measurement and tracking method for large-scale equipment in butt joint process | |
| US10040496B2 (en) | Method for installing an add-on part on a vehicle body | |
| CN113624136B (en) | Part detection device and part detection device calibration method | |
| CN101634541B (en) | Detection method of armature stamping trough position degree | |
| Rejc et al. | Dimensional measurements of a gray-iron object using a robot and a laser displacement sensor |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120704 Termination date: 20210919 |