WO2020087909A1 - 一种基于多相机的光伏电池串定位方法 - Google Patents

一种基于多相机的光伏电池串定位方法 Download PDF

Info

Publication number
WO2020087909A1
WO2020087909A1 PCT/CN2019/087239 CN2019087239W WO2020087909A1 WO 2020087909 A1 WO2020087909 A1 WO 2020087909A1 CN 2019087239 W CN2019087239 W CN 2019087239W WO 2020087909 A1 WO2020087909 A1 WO 2020087909A1
Authority
WO
WIPO (PCT)
Prior art keywords
photovoltaic cell
cell string
camera
long
robot
Prior art date
Application number
PCT/CN2019/087239
Other languages
English (en)
French (fr)
Inventor
尹章芹
万皓
张冶
王杰高
Original Assignee
南京埃斯顿机器人工程有限公司
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 南京埃斯顿机器人工程有限公司 filed Critical 南京埃斯顿机器人工程有限公司
Publication of WO2020087909A1 publication Critical patent/WO2020087909A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/68—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere for positioning, orientation or alignment
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/68—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere for positioning, orientation or alignment
    • H01L21/681—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere for positioning, orientation or alignment using optical controlling means
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L31/00—Semiconductor devices sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation; Processes or apparatus specially adapted for the manufacture or treatment thereof or of parts thereof; Details thereof
    • H01L31/18—Processes or apparatus specially adapted for the manufacture or treatment of these devices or of parts thereof
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L31/00—Semiconductor devices sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation; Processes or apparatus specially adapted for the manufacture or treatment thereof or of parts thereof; Details thereof
    • H01L31/18—Processes or apparatus specially adapted for the manufacture or treatment of these devices or of parts thereof
    • H01L31/1876—Particular processes or apparatus for batch treatment of the devices
    • H01L31/188—Apparatus specially adapted for automatic interconnection of solar cells in a module
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/50—Photovoltaic [PV] energy
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
    • Y02P70/50—Manufacturing or production processes characterised by the final manufactured product

Definitions

  • the invention relates to a material automatic positioning method, in particular to a photovoltaic cell string positioning method based on multiple cameras.
  • solar energy as a renewable energy source, has the characteristics of inexhaustible, inexhaustible, clean and environmental protection, and Gradually used.
  • solar power generation and as the main carrier of photovoltaics, silicon crystal solar cells have to go through string welding and typesetting processes in the process of being produced by units.
  • the typesetting process is to arrange the rows of serially welded single-row batteries neatly in parallel to splice and package them into components.
  • the photovoltaic industry generally has high requirements for the neatness of typesetting (usually the deviation does not exceed 0.1mm); traditional typesetting is performed by manual, with the help of special mechanical devices Typesetting; but this method of purely relying on mechanical positioning cannot guarantee the accuracy of positioning, and manual contact will cause contact loss to the battery; in recent years, with the development of robot technology, industrial robots have begun to apply to typesetting; this avoids Direct manual contact, but the absolute positioning accuracy of the robot often fails to meet the requirements of typesetting accuracy;
  • multi-camera positioning methods based on camera self-calibration. Although it can be extended to determine the pose of objects in three-dimensional space, the operation process of self-calibration methods is cumbersome and the accuracy is low, generally -20mm; In addition, there is a multi-camera positioning method using image stitching technology, which improves the resolution of the image, but the positioning accuracy depends on the effect of camera distortion correction.
  • the Chinese invention patent application CN107611073A discloses a positioning method for solar cell string layout.
  • a backlight and camera solution is used to locate the characteristic points of the photovoltaic cell through a conveyor belt with light transmission properties, and two points are used to confirm a straight line.
  • Scheme calculate the deviation of the photovoltaic cell string from the standard cell string.
  • the equipment solution has certain limitations and high requirements on the mechanical structure. In practical applications, it affects the accuracy of positioning and reduces the working cycle.
  • the image stitching technology mainly uses the stitching of multiple pictures into one large picture.
  • the image stitching algorithm has a long execution time, and the stitching algorithm has high requirements for image feature extraction, and the photovoltaic cell string features are few.
  • the stitched visual pictures The accuracy is much worse, and according to the actual application, the photovoltaic cell string positioning is realized based on the splicing technology, which cannot meet the requirements of the scene in terms of accuracy and performance.
  • the problem to be solved by the present invention is to overcome the shortcomings of the prior art, and propose a multi-camera-based photovoltaic cell string positioning method, which uses localization features such as fitting, weight calculation, and deduction calculation to improve accuracy while improving accuracy , So that the on-site photovoltaic positioning accuracy can reach within 0.1mm, improving work efficiency.
  • the multi-camera photovoltaic cell string positioning method of the present invention has the following steps:
  • An angle camera is arranged above one corner of the tested photovoltaic cell string of the transmission device, and a long edge camera is arranged above the long edge of the tested photovoltaic cell string.
  • the field of view of the angle camera is opposite to one corner of the measured photovoltaic cell string, used to obtain the corner image of the photovoltaic cell string, and the field of view of the long edge camera is opposite to one long edge of the photovoltaic cell string, used to obtain Image of the long side of a photovoltaic cell string.
  • Establish the robot reference coordinate system r use the three-point method to teach three robot points r 0 (x 0 , y 0 , z 0 , a 0 , b 0 , c 0 ), r 1 (x 1 , y 1 , z 0 , a 0 , b 0 , c 0 ), r 2 (x 2 , y 2 , z 0 , a 0 , b 0 , c 0 ), using the rotational translation equation R
  • T to establish the reference coordinates of the robot Department r.
  • R represents the rotation matrix
  • T represents the translation matrix.
  • the robot points r 0 , r 1 , and r 2 have different x-axis and y-axis coordinates, the z-axis coordinate z 0 is the same, and the robot postures a 0 , b 0 , c 0 are the same.
  • (x 0 , y 0 ), (x 1 , y 1 ), (x 2 , y 2 ) represent the coordinates of the robot.
  • X 0 , X 1 , X 2 represent the camera row coordinate value
  • Y 0 , Y 1 , Y 2 represent the camera column coordinate
  • Step 1 Calculate the standard layout position c 3 (X 3 , Y 3 , a 3 ) of the photovoltaic cell string and the position c 4 (X 4 , Y 4 , a 4 ) of the photovoltaic cell string obtained by the long-side camera.
  • Step 1.1 The robot grabs the photovoltaic cell string and places it under the long-side camera and angular camera.
  • the corresponding point coordinates of the robot are: r 6 (x 6 , y 6 , z 6 , a 6 , b 6 , c 6 ),
  • the long-side camera positions the boundary of the photovoltaic cell string, and the angular position camera positions the boundary and intersection of the photovoltaic cell string boundary line.
  • Step 1.2 The corner camera captures the two boundaries of the corners of the photovoltaic cell string, and uses the sub-pixel-level edge fitting method to find the straight lines L1 and L2 of the corners of the corners of the photovoltaic cell string, of which the long side edge straight line L2
  • the angle with the standard line is a 3
  • the intersection point c 3 (X 3 , Y 3 , a 3 ) of the straight lines L1 and L2 at the edge of the corner of the photovoltaic cell string is the standard layout position of the photovoltaic cell string.
  • the straight line formed by c 0 (X 0 , Y 0 ) and c 1 (X 1 , Y 1 ) is the standard straight line.
  • Step 1.3 Each long-side camera captures the long-side boundary of the photovoltaic cell string, and uses sub-pixel-level edge fitting to find L3 of the straight line of the long-edge edge of the photovoltaic cell string.
  • a 4 , intersection point c 4 (X 4 , Y 4 , a 4 ) is the position of the photovoltaic cell string obtained by the long-side camera.
  • Step 2 Calculate the difference between the position of the photovoltaic cell string and the standard layout position of the photovoltaic cell string (Dx, Dy, Da).
  • Step 2.1 Measure half of the length and width of the photovoltaic cell string as l 1 and l 2 respectively .
  • Step 2.2 Calculate the rotation center of the photovoltaic cell string:
  • DX and DY indicate how many millimeters a pixel represents.
  • Step 2.3 The robot grabs a new photovoltaic cell string, shoots it with a long-side camera and an angular camera, and calculates the standard layout position c31 (x 31 , y 31 , a 31 ) of the new photovoltaic cell string and the photovoltaic obtained by the long-side camera
  • the battery string position c41 (x 41 , y 41 , a 41 ).
  • i the starting counting position of the corner coordinate
  • i 2
  • N the number of cameras.
  • Step 2.4 Use Ethernet communication to send the data to the typesetting robot.
  • the robot adds [Dx, Dy, Da] on the basis of point r6 and executes.
  • the two long-edge cameras are set at different positions on the same long edge or different long edges of the photovoltaic cell string (one long-edge camera is provided for each of the two long edges).
  • the field of view of the edge camera is directly on the long side of the photovoltaic cell string.
  • step 1.3 the second long-side positioning criterion is added.
  • the long-side camera captured the boundary of the photovoltaic cell string, and used the sub-pixel-level edge fitting method to find the straight line L5.
  • the angle between the straight line L5 and the standard straight line is a 5
  • the coordinate of the intersection point is c5 (x 5 , y 5 , a 5 ).
  • Step 2.3 Robot 3 grabs the new photovoltaic cell string, and uses the sum of the two long-side camera and angular camera to calculate the new intersection point and angle c31 (x 31 , y 31 , a 31 ) and c41 (x 41 , y 41 , a 41 ), c51 (x 51 , y 51 , a 51 ).
  • Da ((a 3 -a 31 ) + (a 4 -a 41 ) + (a 5 -a 51 )) / 3
  • the boundary of the battery string is positioned by straight line fitting through a multi-camera solution, and the straight line fitting solution at the sub-pixel level is used to increase the accuracy of edge positioning.
  • Calculate the angle change of the photovoltaic cell string use the weight to calculate Da, and calculate Da by the multi-angle angle change to reduce the interference of the battery string burr.
  • Vision locates the straight line of the local boundary and calculates the intersection point c 3 (X 3 , Y 3 ), and calculates the deviation (Dx, Dy) of the position of the battery string from the standard layout position of the battery string through the model form. (The angle deviation is calculated first, and then the position deviation is calculated.) The tempo has been doubled.
  • FIG. 1 is a schematic diagram of a device system in which a robot grabs a photovoltaic cell string with a light source between the camera and the photovoltaic cell string.
  • Fig. 2 is a schematic diagram of an equipment system where a light source is below a photovoltaic cell string and a robot grabs the photovoltaic cell string.
  • FIG. 3 is a schematic diagram of the equipment system of the photovoltaic cell string with the light source above the belt line.
  • Figure 4 shows the characteristic parameters to be calculated for the long-side camera and the angular camera.
  • Figure 5 is the robot three-point relationship to establish the coordinate system r.
  • Fig. 6 is a visual three-point relationship to establish a coordinate system A. Among them, a is the robot point and b is the visual point.
  • Embodiment 1 An embodiment of an angle camera and a long side camera.
  • the on-site tooling solution is divided into the belt line 1 incoming, visually positioned and grabbed by the robot 3. Another solution is that the robot grabs the photovoltaic cell string and is visually positioned.
  • the on-site tooling scheme is shown in Figures 1, 2, and 3. The difference between Figure 1 and Figure 2 is that Figure 1 selects an open-hole light source (2.1, 5.1), and Figure 2 can select a backlight (2.2, 5.2)
  • Figure 3 is a belt line with a robot to grab a string of photovoltaic cells.
  • the angle camera 2 is used to locate the corner of the photovoltaic cell string, and its field of view 2.3 is located at one corner of the photovoltaic cell string 4, that is, the intersection of the two sides.
  • the long side camera 5 is used to locate the long side of the photovoltaic cell string, and its field of view 5.3 covers the long side position of the photovoltaic cell string.
  • the length of the photovoltaic cell string is 2 * l 1 because of the installation of the angle camera 2 and the long side camera 5 distance is generally less than 2 * l 1, at greater than l 1/3.
  • the reference coordinate system r of the robot and the reference coordinate system A of the angular camera 2 are established, and the relationship between r and A is horizontal and vertical.
  • the establishment of the robot reference coordinate system r As shown in Figures 5 and 6, the establishment of the robot reference coordinate system r.
  • R represents the rotation matrix and T represents the translation matrix.
  • T represents the translation matrix.
  • the difference between the robot points r 0 , r 1 , and r 2 is that the x-axis and y-axis coordinates are different, the z-axis coordinate z 0 is the same, and the robot postures a 0 , b 0 , c 0 are the same.
  • (x 0 , y 0 ), (x 1 , y 1 ), (x 2 , y 2 ) represent the coordinates of the robot.
  • Step 1 As shown in Fig. 4, calculating the standard layout position c 3 (X 3 , Y 3 , a 3 ) of the photovoltaic cell string in the angular camera 2 includes the following steps:
  • Step 1.1 The robot grabs the battery slice under the long-side camera 5 and the angular camera 2.
  • the corresponding point coordinates of the robot 3 are: r 6 (x 6 , y 6 , z 6 , a 6 , b 6 , c 6 ),
  • the long-side camera 5 locates the boundary of the photovoltaic cell string
  • the angular camera 2 locates the boundary and intersection of the boundary line of the photovoltaic cell string.
  • Step 1.2 The angle camera 2 captures the boundary of the photovoltaic cell string, and uses the sub-pixel-level edge fitting method to find the straight line L1 and the straight line L2.
  • the angle of the straight line L2 and the standard straight line is a 3 , L1 and L2
  • the intersection point is c 3 (X 3 , Y 3 , a 3 ), as shown in Figure 3.
  • Step 1.3 The long-side camera 5 captures the boundary of the photovoltaic cell string, and uses the sub-pixel-level edge fitting method to find the straight line L3.
  • the angle between the straight line L3 and the standard straight line is a 4
  • the coordinates of the intersection point are
  • the c 4 (X 4 , Y 4 , a 4 ) standard straight line is the reference straight line in the long-side camera 5, and the standard straight line can be defined as a straight line formed by c0 and c1, as shown in FIG. 3.
  • Step 2 Calculate the difference between the position of the photovoltaic cell string and the standard layout position of the photovoltaic cell string (Dx, Dy, Da).
  • Step 2.1 Measurement confirms that half of the length and width of the photovoltaic cell string are l 1 and l 2, respectively .
  • Step 2.2 Calculate the rotation center of the photovoltaic cell string:
  • DX and DY indicate how many millimeters a pixel represents.
  • Step 2.3 The robot 3 grabs the new photovoltaic cell string 4, and uses the long-side camera 5 and the angular camera 2 to calculate new intersection points and angles c31 (x 31 , y 31 , a 31 ) and c41 (x 41 , y 41 , a 41 ).
  • Step 2.4 Use Ethernet communication to send the data to the typesetting robot.
  • the robot adds [Dx, Dy, Da] on the basis of point r6 and executes.
  • Embodiment 2 One camera is used to locate the angular position, and two cameras are used to locate the long side. The implementation steps are the same as in case 1. Only the weight of calculating Da is different.
  • step 1.3 the standard for long-side positioning is added.
  • the standard straight line is the reference straight line in the long-side camera.
  • the standard straight line can be defined as the straight line composed of c0 and c1, as shown in Figure 3.
  • Step 2.3 Robot 3 grabs the new photovoltaic cell string, and uses the sum of the two long-side camera and angular camera to calculate the new intersection point and angle c31 (x 31 , y 31 , a 31 ) and c41 (x 41 , y 41 , a 41 ), c51 (x 51 , y 51 , a 51 ).
  • Da ((a 3 -a 31 ) + (a 4 -a 41 ) + (a 5 -a 51 )) / 3

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Manufacturing & Machinery (AREA)
  • Theoretical Computer Science (AREA)
  • Power Engineering (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Computer Hardware Design (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Electromagnetism (AREA)
  • Sustainable Energy (AREA)
  • Sustainable Development (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Manipulator (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Container, Conveyance, Adherence, Positioning, Of Wafer (AREA)
  • Image Analysis (AREA)

Abstract

本发明公开了一种基于多相机的光伏电池串定位方法，通过角位相机获取光伏电池串的角部图像，长边位相机获取光伏电池串长边的图像。建立机器人参考坐标系r和角位相机参考坐标系。计算光伏电池串标准排版位置c 3(X 3,Y 3,a 3)和长边位相机获取的光伏电池串位置c 4(X 4,Y 4,a 4)，然后计算光伏电池串位置与光伏电池串标准排版位置的差异(Dx,Dy,Da)。最后利用以太网的通讯，将数据发给排版机器人，机器人在点位r6的基础上，加上[Dx,Dy,Da]，并执行。本发明通过拟合、权重计算、推演计算等定位局部特征等方法，提高精度的同时，让现场的光伏定位精度能达到0.1mm以内，提升工作的效率。

Description

一种基于多相机的光伏电池串定位方法 技术领域
本发明涉及一种物料自动定位方法，具体说是一种基于多相机的光伏电池串定位方法。
背景技术
随着能源与环境问题的日益突出，世界各国正将目光逐渐转移到新能源的利用上来；太阳能作为一种可再生能源，由于具有取之不尽、用之不竭、清洁环保等特点，而逐渐得到利用。目前，太阳能的主要利用形式是太阳能发电，而作为光伏的主要载体，硅晶太阳能电池在由单元生产为组件的过程中，要经过串焊、排版等工序。
其中的排版工序，就是将一排排串焊后的单排电池进行整齐的并行排列，以拼接并封装成组件。为了保证组件的美观性以及确保电流汇集的性能，光伏行业一般对排版的整齐性要求较高(通常是偏差不超过0.1mm)；传统的排版是由人工，借助特殊机械装置进行对位，实现排版；但这种单纯依靠机械定位的方式无法保证定位的准确性，而且人工的接触会给电池带来接触损耗；近年，随着机器人技术的发展，工业机器人开始应用到排版中；这避免了人工的直接接触，但是机器人的绝对定位精度也常常不能够达到排版精度要求；
现有多相机定位相关专利中，多为基于相机自标定的多相机定位方法，虽然可以延伸到三维空间的物体位姿确定，但是自标定的方法操作过程繁琐，而且精度较低，一般在1-20mm；此外，还有采用图像拼接技术的多相机定位法，这虽然提高了图像的分辨率，但定位精度依赖于相机畸变校正的效果。
中国发明专利申请CN107611073A公开了一种太阳能电池串排版的定位方法，专利CN107611073A中通过具备透光属性的传送带，利用背光与相机方案，定位光伏电池子的特 征点，并利用两点确认一条直线的方案，计算光伏电池串与标准电池串的偏差。该设备方案具备一定的局限性，并且对机械结构的要求高，在实际的应用中，影响定位的精度、降低工作节拍。而图像拼接技术，主要是利用将多张图片拼接成一张大的图片，图像拼接的算法执行时间长，而且拼接算法对图像的特征提取要求高，光伏电池串特征较少，拼接出的视觉图片，精准性差很多，而且根据实际的应用，基于拼接技术实现光伏电池串定位，在精度与性能上满足不了现场的要求。
发明内容
本发明所要解决的问题在于，克服现有技术存在的缺陷，提出了一种基于多相机的光伏电池串定位方法，通过拟合、权重计算、推演计算等定位局部特征等方法，提高精度的同时，让现场的光伏定位精度能达到0.1mm以内，提升工作的效率。
本发明基于多相机的光伏电池串定位方法，其步骤如下：
在传送装置被测光伏电池串一个边角上方设置一个角位相机，在被测光伏电池串长边上方设置一个长边位相机。所述角位相机的视野正对被测光伏电池串一个边角，用于获取光伏电池串的角部图像，所述长边位相机的视野正对光伏电池串的一个长边，用于获取光伏电池串长边的图像。
建立机器人参考坐标系r：利用三点法，在空间中示教三个机器人点r 0(x 0,y 0,z 0,a 0,b 0,c 0)，r 1(x 1,y 1,z 0,a 0,b 0,c 0)，r 2(x 2,y 2,z 0,a 0,b 0,c 0)，利用旋转平移方程R|T，建立机器人的参考坐标系r。其中R表示旋转矩阵，T表示平移矩阵。机器人点r 0,r 1,r 2的x轴、y轴坐标不同，z轴的坐标z 0相同，机器人的姿态a 0,b 0,c 0相同。(x 0,y 0),(x 1,y 1),(x 2,y 2)表示机器人的坐标。
建立角位相机参考坐标系A：将机器人切换到机器人参考坐标系r中，在空间中顺序走三个点r 3(x 3,y 3,z 1,a 1,b 1,c 1),r 4(x 3,y 4,z 1,a 1,b 1,c 1)，r 5(x 4,y 3,z 1,a 1,b 1,c 1)，因为机器人是带有吸取光伏串的夹具，所以机器人走三个点的时候，夹具上的特征点(夹具的末端位置)需要在角位相机的视野范围中，在角位相机中成像的坐标分别为：c 0(X 0,Y 0),c 1(X 1,Y 1),c 2(X 2,Y 2),利用旋转平移方程R1|T1，建立的角位相机参考坐标系(即：视觉考坐标系)A,其中：r 3,r 4,r 5的关系是，r 3和r 4在机器人x轴上的坐标相同，r 3和r 5在机器人y轴上的坐标相同，z轴的坐标z 1相同，机器人的姿态a 1,b 1,c 1相同。
R 1|T 1坐标系的叉乘表示：
X轴向量
Figure PCTCN2019087239-appb-000001
Y轴向量
Figure PCTCN2019087239-appb-000002
其中：X 0，X 1，X 2表示相机行坐标值，Y 0，Y 1，Y 2表示相机的列坐标。
则遵循右手定则的坐标系第三轴可表示为：
Figure PCTCN2019087239-appb-000003
假设
Figure PCTCN2019087239-appb-000004
则角位相机参考坐标系A的矩阵形式为：
Figure PCTCN2019087239-appb-000005
步骤1.计算光伏电池串标准排版位置c 3(X 3,Y 3,a 3)和长边位相机获取的光伏电池串位置c 4(X 4,Y 4,a 4)。
步骤1.1 机器人抓取光伏电池串放置在长边位相机和角位相机下方，机器人对应的点位坐标是：r 6(x 6,y 6,z 6,a 6,b 6,c 6),长边位相机定位的是光伏电池串边界，角位相机定位的是光伏电池串边界线的边界和交点。
步骤1.2 角位相机拍摄到光伏电池串角部位置的两个边界，采用亚像素级别的边缘拟合的方法，找出光伏电池串角部边缘直线L1、L2，其中，其中长边边缘直线L2与标准直线的夹角为a 3,光伏电池串角部边缘直线L1和L2的交点c 3(X 3,Y 3,a 3)即是光伏电池串标准排版位置。c 0(X 0,Y 0)和c 1(X 1,Y 1)构成的直线为所述标准直线。
步骤1.3 各个长边位相机拍摄到光伏电池串的长边边界，采用亚像素级别的边缘拟合的方法，找出光伏电池串长边边缘直线的L3，直线L3与标准直线之间的角度为a 4,交点c 4(X 4,Y 4,a 4)即是长边位相机获取的光伏电池串位置。
步骤2 计算光伏电池串位置与光伏电池串标准排版位置的差异(Dx,Dy,Da)。
步骤2.1 测量光伏电池串的长和宽的一半分别为l 1和l 2。
步骤2.2 计算光伏电池串的旋转中心：
Figure PCTCN2019087239-appb-000006
其中DX和DY表示一个像素代表多少毫米。
步骤2.3 机器人抓取新的光伏电池串,利用长边位相机和角位相机拍摄，计算新的光伏电池串标准排版位置c31(x 31,y 31,a 31)和长边位相机获取的光伏电池串位置c41(x 41,y 41,a 41)。
Figure PCTCN2019087239-appb-000007
Figure PCTCN2019087239-appb-000008
Figure PCTCN2019087239-appb-000009
其中，i表示角位坐标的起始计数位，i＝2，N表示相机的个数。
步骤2.4 利用以太网的通讯，将数据发给排版机器人，机器人在点位r6的基础上，加上[Dx,Dy,Da]，并执行。
当设置两个长边位相机时，两个长边位相机设置在光伏电池串同一长边的不同位置或不同的长边上(两个长边各设置一个长边位相机)，所述长边位相机的视野正对光伏电池串的一个长边。
在步骤1.3中，增加第二个长边定位的标准。长边位相机拍摄到光伏电池串的边界，采用亚像素级别的边缘拟合的方法，找出直线的L5，直线L5与标准直线之间的角度为a 5,交点的坐标是c5(x 5,y 5,a 5)。
步骤2.3:机器人3抓取新的光伏电池串,利用两个长边位相机和角位相机和，计算新的交点和角度c31(x 31,y 31,a 31)和c41(x 41,y 41,a 41)，c51(x 51,y 51,a 51)。
Da＝((a 3-a 31)+(a 4-a 41)+(a 5-a 51))/3
根据实际的现场情况,可以增扩更多的相机，用于长边定位。这样可以逐步提高光伏电池串定位的精度。
本发明方法，通过多相机方案，利用直线拟合定位电池串的边界，通过亚像素级别的直线拟合的方案，以增加边缘定位的精准性。计算光伏电池串的角度变化，采用权重计算Da，通过多边的角度变化计算Da，以降低电池串毛刺的干扰。视觉通过定位局部边界直线，并计算交点c 3(X 3,Y 3)，通过模型形式计算出电池串位置与电池串标准排版位置的偏差(Dx，Dy)，一次完成计算，无需二次定位(先计算角度偏差，再计算位置偏差)，节拍上提升了1倍。
附图说明
图1是光源在相机与光伏电池串之间，机器人抓取光伏电池串的设备系统示意图。
图2是光源在光伏电池串下方，机器人抓取光伏电池串的设备系统示意图。
图3是光源在皮带线上方的光伏电池串的设备系统示意图。
图4是长边位相机和角位相机需要计算的特征参数。
图5是机器人三点关系建立坐标系r。
图6是视觉的三点关系建立坐标系A。其中，a是机器人点位，b是视觉点位。
具体实施方式
下面结合实施例，对本发明作进一步详细说明。
(请按照前面确定的技术方案的步骤，详细描述本发明方法的实施例)
实施例1.一个角位相机和一个长边位相机的实施例。
基于光伏电池串4来料方案的不同，现场的工装方案分为皮带线1来料、用视觉定位并由机器人3抓取，另一种方案是机器人抓取光伏电池串并由视觉定位。针对不同需求，现场的工装方案如图1,2,3.图1与图2的区别在于，图1选择的是开孔光源(2.1，5.1)，图2可以选择背光源(2.2，5.2)，图3是皮带线配合机器人抓取光伏电池串。
在这几种方案中，图中，角位相机2用于定位光伏电池串的角，其视野2.3定位在光伏电池串4的一个角，即两条边的交点。长边位相机5用于定位光伏电池串的长边,其视野5.3覆盖光伏电池串长边位置，光伏电池串的长度是2*l 1，因为角位相机2与长边位相机5的安装距离一般为小于2*l 1，大于l 1/3处。
如图5和图6，建立机器人的参考坐标系r，建立角位相机2的参考坐标系A，r与A的 关系是水平垂直。
如图5和图6，机器人参考坐标系r的建立。利用三点法，在空间中示教三个机器人点r 0(x 0,y 0,z 0,a 0,b 0,c 0)r 1(x 1,y 1,z 0,a 0,b 0,c 0)，r 2(x 2,y 2,z 0,a 0,b 0,c 0)，利用旋转平移方程R|T，建立机器人的参考坐标系r，参考坐标系r一般与光伏电池串的排版机构一致。其中R表示旋转矩阵，T表示平移矩阵。机器人点r 0,r 1,r 2的区别是x轴，y轴坐标不同，z轴的坐标z 0相同，机器人的姿态a 0,b 0,c 0相同。(x 0,y 0),(x 1,y 1),(x 2,y 2)表示机器人的坐标。
如图6，视觉坐标系的建立。将机器人切换到参考坐标系r中，在空间中顺序走三个点r 3(x 3,y 3,z 1,a 1,b 1,c 1),r 4(x 3,y 4,z 1,a 1,b 1,c 1)，r 5(x 4,y 3,z 1,a 1,b 1,c 1)，因为机器人是带有吸取光伏串的夹具，所有机器人走的三个点时候，夹具上的特征点(夹具的末端位置)需要在角位相机2的视野2.3范围中，在角位相机2中成像的坐标分别为：c 0(X 0,Y 0),c 1(X 1,Y 1),c 2(X 2,Y 2),利用旋转平移方程R1|T1，建立的视觉考坐标系A,其中：r 3,r 4,r 5的关系是，r 3和r 4在机器人x轴上的坐标相同，r 3和r 5在机器人y轴上的坐标相同，z轴的坐标z 1相同，机器人的姿态a 1,b 1,c 1相同。
R 1|T 1坐标系的叉乘表示：
坐标系的叉乘表示：
X轴向量
Figure PCTCN2019087239-appb-000010
Y轴向量
Figure PCTCN2019087239-appb-000011
则遵循右手定则的坐标系第三轴可表示为：
Figure PCTCN2019087239-appb-000012
假设
Figure PCTCN2019087239-appb-000013
则相机坐标系的矩阵形式为：
Figure PCTCN2019087239-appb-000014
步骤1.如图4，计算角位相机2中光伏电池串标准排版位置c 3(X 3,Y 3,a 3)，具体包括如下步骤：
步骤1.1：机器人抓取电池片在长边位相机5和角位相机2下方，机器人3对应的点位坐标是：r 6(x 6,y 6,z 6,a 6,b 6,c 6),长边位相机5定位的是光伏电池串边界，角位相机2定位的是光伏电池串边界线的边界和交点。
步骤1.2：角位相机2拍摄到光伏电池串的边界，采用亚像素级别的边缘拟合的方法，找出直线的L1和直线L2，直线L2与标准直线的角度是a 3,L1和L2的交点是c 3(X 3,Y 3,a 3)，如图3所示。
步骤1.3：长边位相机5拍摄到光伏电池串的边界，采用亚像素级别的边缘拟合的方法，找出直线的L3，直线L3与标准直线之间的角度为a 4,交点的坐标是c 4(X 4,Y 4,a 4)标准直线是长边位相机5中的参考直线，可以定义标准直线为c0和c1构成的直线，如图3所示。
步骤2：计算光伏电池串位置与光伏电池串标准排版位置的差异(Dx,Dy,Da)。
步骤2.1：测量确认光伏电池串的长和宽的一半分别为l 1和l 2。
步骤2.2：计算光伏电池串的旋转中心：
Figure PCTCN2019087239-appb-000015
其中DX和DY表示一个像素代表多少毫米。
步骤2.3:机器人3抓取新的光伏电池串4,利用长边位相机5和角位相机2，计算 新的交点和角度c31(x 31,y 31,a 31)和c41(x 41,y 41,a 41)。
Da＝((a 3-a 31)+(a 4-a 41))/2
Figure PCTCN2019087239-appb-000016
Figure PCTCN2019087239-appb-000017
步骤2.4：利用以太网的通讯，将数据发给排版机器人，机器人在点位r6的基础上，加上[Dx,Dy,Da]，并执行。
实施例2.现场利用一个相机定位角位，两个相机定位长边。实施步骤与案例1相同。只是计算Da的权重不同。
在步骤1.3中，增加长边定位的标准。长边位相机拍摄到光伏电池串的边界，采用亚像素级别的边缘拟合的方法，找出直线的L5，直线L5与标准直线之间的角度为a 5,交点的坐标是c5(x 5,y 5,a 5)标准直线是长边位相机中的参考直线，可以定义标准直线为c0和c1构成的直线，可以定义如图3所示。
步骤2.3:机器人3抓取新的光伏电池串,利用两个长边位相机和角位相机和，计算新的交点和角度c31(x 31,y 31,a 31)和c41(x 41,y 41,a 41)，c51(x 51,y 51,a 51)。
Da＝((a 3-a 31)+(a 4-a 41)+(a 5-a 51))/3
根据实际的现场情况,可以增扩更多的相机，用于长边定位。

Claims (6)

  1. 一种基于多相机的光伏电池串定位方法，其步骤如下：
    在传送装置被测光伏电池串一个边角上方设置一个角位相机，在被测光伏电池串长边上方设置一个长边位相机；所述角位相机的视野正对被测光伏电池串一个边角，用于获取光伏电池串的角部图像，所述长边位相机的视野正对光伏电池串的一个长边，用于获取光伏电池串长边的图像；
    建立机器人参考坐标系r；
    建立角位相机参考坐标系A：
    Figure PCTCN2019087239-appb-100001
    步骤1.计算光伏电池串标准排版位置c 3(X 3,Y 3,a 3)和长边位相机获取的光伏电池串位置c 4(X 4,Y 4,a 4)；
    步骤2计算光伏电池串位置与光伏电池串标准排版位置的差异(Dx,Dy,Da)；利用以太网的通讯，将数据发给排版机器人，机器人在点位r6的基础上，加上[Dx,Dy,Da]，并执行；所述点位r6是，机器人抓取光伏电池串放置在长边位相机和角位相机下方时，机器人对应的点位，其坐标是：r 6(x 6,y 6,z 6,a 6,b 6,c 6)。
  2. 根据权利要求1所述的基于多相机的光伏电池串定位方法，其特征是：
    设置两个长边位相机，两个长边位相机设置在光伏电池串同一长边的不同位置或不同的长边上，所述长边位相机的视野正对光伏电池串的一个长边；
    在步骤1.3中，增加第二个长边定位的标准：长边位相机拍摄到光伏电池串的边界，采用亚像素级别的边缘拟合的方法，找出直线的L5，直线L5与标准直线之间的角度为a 5,交点的坐标是c5(x 5,y 5,a 5)；
    步骤2.3:机器人抓取新的光伏电池串,利用两个长边位相机和角位相机和，计算新 的交点和角度c31(x 31,y 31,a 31)和c41(x 41,y 41,a 41)，c51(x 51,y 51,a 51)：
    Da＝((a 3-a 31)+(a 4-a 41)+(a 5-a 51))/3。
  3. 根据权利要求1所述的基于多相机的光伏电池串定位方法，其特征是：所述建立机器人参考坐标系r是，利用三点法，在空间中示教三个机器人点r 0(x 0,y 0,z 0,a 0,b 0,c 0)，r 1(x 1,y 1,z 0,a 0,b 0,c 0)，r 2(x 2,y 2,z 0,a 0,b 0,c 0)，利用旋转平移方程R|T，建立机器人的参考坐标系r，其中R表示旋转矩阵，T表示平移矩阵；机器人点r 0,r 1,r 2的x轴、y轴坐标不同，z轴的坐标z 0相同，机器人的姿态a 0,b 0,c 0相同；(x 0,y 0),(x 1,y 1),(x 2,y 2)表示机器人的坐标。
  4. 根据权利要求1所述的基于多相机的光伏电池串定位方法，其特征是：
    所述建立角位相机参考坐标系A是，将机器人切换到机器人参考坐标系r中，在空间中顺序走三个点r 3(x 3,y 3,z 1,a 1,b 1,c 1),r 4(x 3,y 4,z 1,a 1,b 1,c 1)，r 5(x 4,y 3,z 1,a 1,b 1,c 1)，在角位相机中成像的坐标分别为：c 0(X 0,Y 0),c 1(X 1,Y 1),c 2(X 2,Y 2),利用旋转平移方程R1|T1，建立的角位相机参考坐标系A,其中：r 3,r 4,r 5的关系是，r 3和r 4在机器人x轴上的坐标相同，r 3和r 5在机器人y轴上的坐标相同，z轴的坐标z 1相同，机器人的姿态a 1,b 1,c 1相同；
    R 1|T 1坐标系的叉乘表示：
    X轴向量
    Figure PCTCN2019087239-appb-100002
    Y轴向量
    Figure PCTCN2019087239-appb-100003
    其中：X 0，X 1，X 2表示相机行坐标值，Y 0，Y 1，Y 2表示相机的列坐标；
    则遵循右手定则的坐标系第三轴可表示为：
    Figure PCTCN2019087239-appb-100004
    假设
    Figure PCTCN2019087239-appb-100005
    则角位相机参考坐标系A的矩阵形式为：
    Figure PCTCN2019087239-appb-100006
  5. 根据权利要求1所述的基于多相机的光伏电池串定位方法，其特征是：
    所述步骤1.计算光伏电池串标准排版位置c 3(X 3,Y 3,a 3)和长边位相机获取的光伏电池串位置c 4(X 4,Y 4,a 4)：
    步骤1.1机器人抓取光伏电池串放置在长边位相机和角位相机下方，机器人对应的点位坐标是：r 6(x 6,y 6,z 6,a 6,b 6,c 6),长边位相机定位的是光伏电池串边界，角位相机定位的是光伏电池串边界线的边界和交点；
    步骤1.2角位相机拍摄到光伏电池串角部位置的两个边界，采用亚像素级别的边缘拟合的方法，找出光伏电池串角部边缘直线L1、L2，其中，其中长边边缘直线L2与标准直线的夹角为a 3,光伏电池串角部边缘直线L1和L2的交点c 3(X 3,Y 3,a 3)，即是光伏电池串标准排版位置；c 0(X 0,Y 0)和c 1(X 1,Y 1)构成的直线为所述标准直线；
    步骤1.3各个长边位相机拍摄到光伏电池串的长边边界，采用亚像素级别的边缘拟合的方法，找出光伏电池串长边边缘直线的L3，直线L3与标准直线之间的角度为a 4,交点c 4(X 4,Y 4,a 4)即是长边位相机获取的光伏电池串位置。
  6. 根据权利要求1所述的基于多相机的光伏电池串定位方法，其特征是：
    所述步骤2.计算光伏电池串位置与光伏电池串标准排版位置的差异(Dx,Dy,Da)：
    步骤2.1测量光伏电池串的长和宽的一半分别为l 1和l 2；
    步骤2.2计算光伏电池串的旋转中心：
    Figure PCTCN2019087239-appb-100007
    其中DX和DY表示一个像素代表多少毫米；
    步骤2.3机器人抓取新的光伏电池串,利用长边位相机和角位相机拍摄，计算新的光伏电池串标准排版位置c31(x 31,y 31,a 31)和长边位相机获取的光伏电池串位置c41(x 41,y 41,a 41)：
    Figure PCTCN2019087239-appb-100008
    Figure PCTCN2019087239-appb-100009
    Figure PCTCN2019087239-appb-100010
    其中，i表示角位坐标的起始计数位，i＝2，N表示相机的个数。
PCT/CN2019/087239 2018-10-31 2019-05-16 一种基于多相机的光伏电池串定位方法 WO2020087909A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811282155.3A CN111127553B (zh) 2018-10-31 2018-10-31 一种基于多相机的光伏电池串定位方法
CN201811282155.3 2018-10-31

Publications (1)

Publication Number Publication Date
WO2020087909A1 true WO2020087909A1 (zh) 2020-05-07

Family

ID=70462029

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/087239 WO2020087909A1 (zh) 2018-10-31 2019-05-16 一种基于多相机的光伏电池串定位方法

Country Status (2)

Country Link
CN (1) CN111127553B (zh)
WO (1) WO2020087909A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114739254A (zh) * 2022-04-01 2022-07-12 山西漳山发电有限责任公司 一种可在线测量尺寸的标尺
CN115797310A (zh) * 2022-12-13 2023-03-14 无锡海纳智能科技有限公司 一种光伏电站组串倾角的确定方法及电子设备
WO2023060511A1 (zh) * 2021-10-14 2023-04-20 宁德时代新能源科技股份有限公司 确定电池箱的位置的方法及拆换电池箱的方法
CN116360467A (zh) * 2023-06-01 2023-06-30 中国科学院自动化研究所 光伏板列无人清扫装置自主巡行系统及自主巡行方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113043334B (zh) * 2021-02-23 2022-12-06 上海埃奇机器人技术有限公司 一种基于机器人的光伏电池串定位方法
CN114882023B (zh) * 2022-07-07 2022-11-18 苏州小牛自动化设备有限公司 电池串位姿校正方法、装置、控制设备及系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103692447A (zh) * 2013-12-27 2014-04-02 西安航天精密机电研究所 一种基于俯视和侧视相机的电动汽车电池定位方法
CN106934813A (zh) * 2015-12-31 2017-07-07 沈阳高精数控智能技术股份有限公司 一种基于视觉定位的工业机器人工件抓取实现方法
US20180222055A1 (en) * 2017-02-03 2018-08-09 Abb Schweiz Ag Calibration article for a 3d vision robotic system

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5227985A (en) * 1991-08-19 1993-07-13 University Of Maryland Computer vision system for position monitoring in three dimensions using non-coplanar light sources attached to a monitored object
CN204290879U (zh) * 2014-12-31 2015-04-22 苏州巨能图像检测技术有限公司 电池串的定位检测装置
CN107611073B (zh) * 2017-09-15 2023-11-21 苏州宏瑞达新能源装备有限公司 太阳能电池串排版的定位方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103692447A (zh) * 2013-12-27 2014-04-02 西安航天精密机电研究所 一种基于俯视和侧视相机的电动汽车电池定位方法
CN106934813A (zh) * 2015-12-31 2017-07-07 沈阳高精数控智能技术股份有限公司 一种基于视觉定位的工业机器人工件抓取实现方法
US20180222055A1 (en) * 2017-02-03 2018-08-09 Abb Schweiz Ag Calibration article for a 3d vision robotic system

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023060511A1 (zh) * 2021-10-14 2023-04-20 宁德时代新能源科技股份有限公司 确定电池箱的位置的方法及拆换电池箱的方法
CN114739254A (zh) * 2022-04-01 2022-07-12 山西漳山发电有限责任公司 一种可在线测量尺寸的标尺
CN115797310A (zh) * 2022-12-13 2023-03-14 无锡海纳智能科技有限公司 一种光伏电站组串倾角的确定方法及电子设备
CN116360467A (zh) * 2023-06-01 2023-06-30 中国科学院自动化研究所 光伏板列无人清扫装置自主巡行系统及自主巡行方法
CN116360467B (zh) * 2023-06-01 2023-08-15 中国科学院自动化研究所 光伏板列无人清扫装置自主巡行系统及自主巡行方法

Also Published As

Publication number Publication date
CN111127553B (zh) 2023-04-28
CN111127553A (zh) 2020-05-08

Similar Documents

Publication Publication Date Title
WO2020087909A1 (zh) 一种基于多相机的光伏电池串定位方法
CN105234943B (zh) 一种基于视觉识别的工业机器人示教装置及方法
CN107192331A (zh) 一种基于双目视觉的工件抓取方法
JP6860180B2 (ja) 太陽光パネル清掃ロボットの測位装置及びその測位方法
CN106272426A (zh) 太阳能电池片串焊前视觉定位及角度检测设备及检测方法
WO2021004416A1 (zh) 一种基于视觉信标建立信标地图的方法、装置
CN110470238A (zh) 一种手持激光三维扫描仪、扫描方法及装置
CN201707034U (zh) 一种基于多像机多视场的轨距检测系统定标装置
WO2018145632A1 (zh) 用于太阳能面板清扫机器人的导航系统及其导航方法
CN114705122B (zh) 一种大视场立体视觉标定方法
CN113043334B (zh) 一种基于机器人的光伏电池串定位方法
CN106060399A (zh) 一种双摄像头自动aa的方法及装置
CN109798867B (zh) 一种跨越式清洗车的定日镜异常姿态检测方法及装置
CN111896032B (zh) 一种单目散斑投射器位置的标定系统及方法
CN111126381A (zh) 一种基于r-dfpn算法的绝缘子倾斜定位与识别方法
WO2023206020A1 (zh) 电池极片的Overhang检测方法、装置、设备、存储介质
CN105488807A (zh) 一种远心镜头的标定和矫正方法
CN112101118A (zh) 一种远景与近景图像结合的光伏组件定位与匹配方法
CN112718555A (zh) 一种硅太阳能电池分选工序测试系统
CN104931421A (zh) 液晶面板视觉检测系统的定位对焦方法
CN106643735A (zh) 一种室内定位方法、装置和移动终端
CN104748714B (zh) 一种星载索网可展开天线形面精度测量方法
CN103322980A (zh) 基于立体视觉的水果姿态检测方法
CN108760023B (zh) 两端支撑太阳翼的视觉振动测量装置及方法
CN107917666A (zh) 双目视觉装置及坐标标定方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19878664

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19878664

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19878664

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 07.01.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 19878664

Country of ref document: EP

Kind code of ref document: A1