WO2019090487A1 - 数控机床高动态大范围任意轮廓误差单目六维测量方法 - Google Patents

数控机床高动态大范围任意轮廓误差单目六维测量方法 Download PDF

Info

Publication number
WO2019090487A1
WO2019090487A1 PCT/CN2017/109782 CN2017109782W WO2019090487A1 WO 2019090487 A1 WO2019090487 A1 WO 2019090487A1 CN 2017109782 W CN2017109782 W CN 2017109782W WO 2019090487 A1 WO2019090487 A1 WO 2019090487A1
Authority
WO
WIPO (PCT)
Prior art keywords
coordinate system
camera
machine tool
contour
measurement
Prior art date
Application number
PCT/CN2017/109782
Other languages
English (en)
French (fr)
Inventor
刘巍
贾振元
李肖
潘翼
马鑫
马建伟
Original Assignee
大连理工大学
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 大连理工大学 filed Critical 大连理工大学
Priority to US16/338,971 priority Critical patent/US11014211B2/en
Priority to PCT/CN2017/109782 priority patent/WO2019090487A1/zh
Priority to EP17925531.0A priority patent/EP3511122B1/en
Publication of WO2019090487A1 publication Critical patent/WO2019090487A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B23Q17/2409—Arrangements for indirect observation of the working space using image recording means, e.g. a camera
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B23Q17/2414—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for indicating desired positions guiding the positioning of tools or workpieces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B23Q17/2452—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B23Q17/248—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves using special electromagnetic means or methods
    • B23Q17/249—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves using special electromagnetic means or methods using image analysis, e.g. for radar, infrared or array camera images
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/02—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness
    • G01B21/04—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness by measuring coordinates of points
    • G01B21/042—Calibration or calibration artifacts
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration

Definitions

  • the invention belongs to the field of machine tool dynamic error detection, and relates to a six-dimensional measurement method for high-dynamic arbitrary large-scale contour error of a numerical control machine tool with a monocular vision measurement technology combined with short-time stroboscopic illumination and a priori standard board.
  • Difficult to machine and variable curvature parts such as marine propellers, propellers for ships, turbines and engine blades are widely used in national major equipment projects.
  • the shape and geometric accuracy of such parts directly affect the performance of the equipment.
  • the multi-axis linkage machine tool can solve the problem of one-time clamping and processing of most complex variable curvature parts by virtue of the controllability of the tool and workpiece position angle during the machining process.
  • All kinds of difficult-to-machine and variable-curvature parts are processed by multi-axis CNC machine tools under special high-dynamic conditions with special interpolation contours. Especially under high feed conditions, CNC machine tools are especially prominent due to insufficient dynamic characteristics of machine tools.
  • Contour error is an important indicator for evaluating the dynamic performance of CNC machine tools. Therefore, the periodic evaluation of machine tool high dynamics, large range, and arbitrary contour space error (dynamic characteristics) is an important guarantee for evaluating the dynamic performance of CNC machine tools and improving machining accuracy.
  • the existing methods for measuring the contour error of CNC machine tools include ball bar measurement method, plane grating measurement method, R-test measurement method and visual measurement method.
  • the two-step method is used to obtain the measurement data by installing the ballbar instrument, and the four installation errors and six volume errors are separated according to the error identification model combined with the identification algorithm.
  • the ballbar device has high flexibility, it is a one-dimensional measuring device that can only measure the plane circle contour error and cannot achieve any contour error measurement. In addition, due to the mechanical structure of the rod length, it is difficult for the ballbar to measure small-circular circular contour errors that better reflect the dynamic performance of the machine.
  • R-test has high measurement accuracy, but its measurement range is small.
  • the X, Y, and Z single-direction measurement ranges are less than 12mm, and the linkage error of unrelated axes is introduced when performing measurement.
  • Soichi Ibaraki of Kyoto University, Japan, and "Diagnosis and compensation of motion errors in NC machine tools by arbitrary shape contouring error measurement” published by "Laser Metrology & Machine Performance V” proposed a method for machine tool installation and servo system error measurement and identification based on planar grating. Combined with the feedback signal of the NC position, the servo system is compensated to improve the contour accuracy of the machine tool.
  • this device is cumbersome to operate and cannot measure the swivel circle profile.
  • this method is limited by the camera bandwidth, and the camera frame rate is limited to increase, which causes the feature points in the captured image to be blurred under high feed conditions, which reduces the accuracy of the contour error.
  • This method can only measure two-dimensional contour error, and can not realize the three-dimensional solution of the contour error of CNC machine tools.
  • the technical problem to be solved by the invention is to overcome the defects of the prior art, and the existing single measurement method can not solve the six-dimensional measurement problem of high dynamics, large range and arbitrary contour error of the numerical control machine tool, and invents a high dynamic large-scale arbitrary contour of the numerical control machine tool.
  • Error monocular six-dimensional measurement method The measurement tooling and measurement system is designed. This method combines the error distribution principle to improve the field of view measurement accuracy of the coding element by using the small measurement field of view.
  • the monocular pose algorithm combined with the prior knowledge is used to improve the dimension of the visually measurable machine tool interpolation contour.
  • the entire machine motion profile is characterized by a selected reference element, and then the X, Y, Z, pitch, roll and yaw six-dimensional representation of the machine interpolation contour represented by the reference element in the machine coordinate system is obtained by reference conversion.
  • Information using this method to traverse the image of each frame to obtain the actual six-dimensional motion profile of the machine tool; by comparing with the nominal motion profile of the machine tool, the six-dimensional error generated by the interpolation contour of the CNC machine tool can be solved.
  • the measuring system of the method has low cost and is easy to operate.
  • the technical scheme adopted by the invention is a high dynamic large-scale arbitrary contour error monocular six-dimensional measuring method for a numerically controlled machine tool, which is characterized in that the method designs a measuring tooling and measuring system, and uses a monocular visual pose algorithm combined with prior knowledge. Improve the dimension and range of the visually measurable machine tool interpolation contour.
  • the whole machine motion profile is characterized by a selected reference element; combined with the error distribution principle, the field measurement accuracy of the coded element is improved by using the small measurement field of view;
  • the six-dimensional information of X, Y, Z, pitch, roll and yaw of the machine tool interpolation contour represented by the reference element in the machine coordinate system is obtained.
  • the actual six-dimensional motion profile of the machine tool is obtained by traversing the image of each frame.
  • the six-dimensional error generated by the interpolation contour of the CNC machine tool can be solved by comparing with the nominal motion profile of the machine tool; the specific steps of the method are as follows:
  • Step 1 Install the measurement tooling and measurement system
  • the measuring tooling is composed of a base body 10, a high-brightness short-time lighting unit 9, a priori standard board 7, and an encoding element 8.
  • the a priori standard board 7 is a transparent base material on which an encoding element 8 having a unique coding value and a matrix distribution is distributed.
  • the measuring system comprises a camera 1, a camera fixture 2 and a measuring tool; the camera 1 is fixed on the camera fixture 2, and the camera fixture 2 is mounted above the measuring tool to collect a sequence image during the measurement tooling movement; the assembled measuring tool is placed in the On the optical three-coordinate device platform, the spatial geometric relationship between the coding elements 8 is calibrated in the global coordinate system of the a priori standard plate by using an optical three-coordinate device; when the measurement system is arranged, the calibrated measurement tool is passed through the compression bolt 11 and compacted.
  • the nut 12 is fastened on the rotary table 3 of the numerical control machine tool 4 for characterizing the movement information of the machine tool;
  • Step 2 Establish a priori standard board global coordinate system
  • the global coordinate system O G X G Y G Z G 14 of the a priori standard plate is established on the measuring tool.
  • the origin is established on the center of the coding element 8 in the first column of the first row, defined as O G ; the direction of the X G coordinate axis
  • the origin O G points downward to the center point of the last row of the coding element 8 on the first column of the array; the direction of the Y G coordinate axis is directed to the right by the O G to the center point of the last column of the coding element 8 of the first row on the array;
  • the axis is determined by the right-hand rule; the spatial geometric relationship between the coding elements 8 is calibrated using the optical three-coordinate device in the global coordinate system O G X G Y G Z G 14 of the a priori standard plate, and each coding element 8 is obtained globally on the a priori standard plate.
  • the coding element 8 on the a priori standard board 7 carries the motion information of the numerical control machine tool 4, and the spatial positional relationship between the coding elements 8 is calibrated by high-precision equipment.
  • the size of the a priori standard board 7 can be manufactured as large as possible to meet the wide range of measurement requirements of contour error;
  • the third step is camera calibration
  • the camera imaging model expresses the one-to-one mapping relationship between the camera coordinate system and the world coordinate system.
  • the camera imaging model with distortion parameters is:
  • (X w , Y w , Z w ) is the three-dimensional coordinate of the center point of the coding element 8 in the world coordinate system
  • K is the internal parameter matrix of the camera 1
  • T is the outer parameter matrix of the camera 1
  • (u, v) is The center point of the coding element 8 is in the two-dimensional coordinates of the image plane
  • (u 0 , v 0 ) is the principal point coordinate of the image
  • (C x , C y ) is the equivalent focal length in the horizontal and vertical directions.
  • the rotation and translation transformation matrix between the camera coordinate system and the world coordinate system, respectively, ( ⁇ x , ⁇ y ) are the distortions of the image points in the x, y direction caused by the imperfection of the optical system; using the checkerboard calibration plate
  • the camera 1 measures a plurality of positions in the field of view 21 to obtain a calibration plate image, and calibrates the camera 1 distortion parameter and the internal and external parameter matrix by a calibration algorithm proposed by Zhang Zhengyou;
  • the fourth step is CNC machine tool high dynamic, large range interpolation contour high definition without blur acquisition and image processing
  • the CNC machine tool 4 is used to interpolate the contour image; because the accuracy of the contour error measurement requires high precision, the required field of view of the camera 21 is small; first adjust the camera 1 parameter to make it in the best view. Field and frame rate; subsequently, the camera 1 and the high-brightness short-time light-emitting unit 9 are synchronously triggered, and the light-emitting time and the light-emitting intensity of the high-brightness short-time light-emitting unit 9 are set to ensure the high-brightness short-time light-emitting unit 9 during the exposure time of the camera 1.
  • the code value represented by each code element 8 on the image is identified and the two-dimensional pixel coordinates of the center point of each coded element 8 after decoding are located by using the gray center of gravity method; the spherical flag is located by using the gray center of gravity method.
  • Point center whose calculation expression is:
  • (i, j) represents the image pixel point coordinates
  • m, n is the number of pixels contained in the horizontal and vertical directions of the image
  • (x, y) is the centroid coordinate of the image
  • f(i, j) is the pixel Gray value at coordinates (i, j);
  • the fifth step is the six-dimensional solution of high dynamic range and arbitrary contour error of CNC machine tools.
  • the method combines the error distribution principle, uses the small measurement field of view 21 to improve the measurement accuracy of the coding element in the field of view; and uses the monocular pose algorithm combined with the prior knowledge to improve the dimension and range of the visually measurable machine tool interpolation contour; Characterized by a selected reference element, The position of the point in the invisible area of the measurement field of view 21 is obtained by the pixel coordinates of the visible area point combined with the high-precision a priori constraint solution; the reference element is used to traverse all the images to obtain the motion profile of the reference element in the machine coordinate system.
  • the six-dimensional error of the interpolation contour of the CNC machine tool 4 can be solved by comparing with the nominal motion profile of the machine tool; the six-dimensional solution of the high-dynamic large-scale arbitrary contour error of the numerical control machine tool is as follows:
  • the camera 1 measures the field of view 21 as N ⁇ N (unit: mm), and the size of the a priori standard board 7 is M ⁇ M (unit: mm), and N is much smaller than M; the coordinate system involved is in addition to the a priori standard plate global coordinate system described above.
  • O G X G Y G Z G 14 also relates to camera coordinate system O C X C Y C Z C 13, machine coordinate system O M X M Y M Z M 15 and prior standard plate local coordinate system O Li X Li Y Li Z Li 22; camera coordinate system O C X C Y C Z C 13 origin is established at the optical center O C ; when the numerical control machine tool 4 is not moving, the four frames arranged in the field of view are arranged in the first frame image.
  • Coding element with Selected code element As a reference element; the motion profile synthesized by the CNC machine tool 4 in the interpolation motion axis of each axis passes the coding element To characterize; its coordinates under the a priori standard plate global coordinate system O G X G Y G Z G 14 are Take The machine coordinate system O M X M Y M Z M is established for the origin. The machine coordinate system O M X M Y M Z M 15 The coordinate axes are in the same direction as the movement axes of the CNC machine tool 4; the control machine drives the measuring tool 3 along the machine tool.
  • the X-axis direction vector is fitted to the camera coordinate system O C X C Y C Z C 13 three-dimensional coordinates (x, y, z); the machine coordinate system O M X M Y M Z M is determined according to the same rules
  • the Y axis of 15 and the Z axis of the machine coordinate system O M X M Y M Z M 15 are determined by the right-hand rule; the X and Y axes are established as follows:
  • (m x , n x , p x ) is the X-axis direction vector of the machine coordinate system O M X M Y M Z M 15
  • (m y , n y , p y ) is the Y-axis direction vector of the machine coordinate system O M X M Y M Z M 15 ,
  • ( C X, C Y, C Z) is the camera coordinate system O C X C Y C Z C 13
  • the random standard bed of the a priori standard board 7 is continuously performing the interpolation motion, and the coding element 8 thereon is continuously imaged on the camera 1; during the movement of the numerical control machine tool 4, the camera 1 collects the G frame image in the ith frame.
  • the four coding elements 8 appearing in the image in a rectangular arrangement in the field of view are The coordinates of the four code elements 8 at the a priori standard plate global coordinate system O G X G Y G Z G 14 are Corresponding two-dimensional pixel coordinates on the image
  • Establish the local coordinate system O Li X Li Y Li Z Li 22 of the a priori standard plate under the i-th frame, (i 1, 2...G); 23 is the coordinate origin, and the X Li and Y Li coordinate axes are respectively parallel to the X G and Y G directions of the a priori standard plate global coordinate system O G X G Y G Z G 14 , and the Z Li coordinate axis is determined by the right-hand rule; Then the three coordinates of the selected four coding elements 8 in the prior coordinate standard local coordinate system O Li X Li Y Li Z Li 22 are:
  • X i is the camera coordinate system O C X C Y C Z C 13 under the optical center O C to the i-th frame a priori standard board 7
  • Y i is the camera coordinate system O C X C Y C Z C 13 under the optical center O C to the i-th frame a priori standard board 7
  • Z i is the camera coordinate system O C X C Y C Z C 13 under the optical center O C to the i-th frame prior standard board Point distance
  • a' is the i-th frame a priori standard board 7 global coordinate system O G X G Y G Z G 14 versus Distance between b
  • b' is the i-th frame a priori standard plate global coordinate system O G X G Y G Z G 14 versus Distance between
  • c' is the ith frame a priori standard plate global coordinate system O G X G Y G Z G 14 versus Distance between;
  • ⁇ is a straight line versus Angle between ⁇ is a straight line
  • t' i , t' i is the ith frame a priori standard plate local coordinate system O Li X Li Y Li Z Li 22 origin
  • O Li X Li Y Li Z Li 22 origin Three-dimensional coordinates in the camera coordinate system O C X C Y C Z C 13
  • the elevation angle ⁇ ′ i , the roll angle ⁇ ′ i , and the yaw angle ⁇ ′ i between the two coordinate systems are solved by R′ i separation; the reference element is solved according to the known spatial constraint between the coded elements 8 of the a priori standard plate 7
  • the measured contour error is characterized by the machine coordinate system O M X M Y M Z M 15 , and the contour measured by camera 1 is coordinate-converted to correctly solve the machine contour error; machine coordinate system O M X M Y M Z M
  • the pose transformation formula of the 15 and the prior standard plate local coordinate system O Li X Li Y Li Z Li 22 is:
  • the contour L m solves the six-degree-of-freedom contour error E generated by the interpolation of the CNC machine tool 4:
  • the invention improves the interpolation speed of the visually measurable numerical control machine tool, breaks through the measurable speed limit of the machine tool, and expands the interpolation feed speed range of the numerical control machine tool for visual inspection.
  • the camera measurement is small.
  • the present invention adopts the visual pose algorithm combined with the prior space constraint between the coding elements on the large-scale standard board to realize the six-dimensional arbitrary contour error of the numerical control machine tool in the small field of view. Measurements expand the measurable range of motion and measurement dimensions of machine tools. Only a single camera, special measurement tooling and measurement system are used in the measurement process, which reduces the cost and improves the ease of operation.
  • Figure 1 is a high-dynamic large-scale arbitrary contour error six-dimensional measurement system diagram of CNC machine tools. Among them, 1-camera, 2-camera fixture, 3-turntable, 4-CNC machine.
  • Figure 2 is an exploded view of the measurement tooling. Among them, 5-pressure plate locking bolt, 6-press plate, 7-prior standard plate, 8-code element, 9- short time high Brightness lighting unit, 10-base, 11-compression bolt, 12-compression nut.
  • Figure 3 shows the layout of the measurement system and each coordinate system when the machine tool is not moving.
  • 13-camera coordinate system O C X C Y C Z C 14-a priori standard plate global coordinate system O G X G Y G Z G , 15-machine coordinate system O M X M Y M Z M ,16- Coding element 17-encoded element 18-encoding element 19-encoding element 20-first frame prior standard board local coordinate system O L1 X L1 Y L1 Z L1 , 21-camera measurement field of view.
  • FIG. 4 shows the image encoding element identification and positioning result of the 300th frame.
  • Figure 5 is an visual solution to the error produced by the machine's interpolation contour in the X direction.
  • Figure 6 shows the error that the machine tool interpolation contour produces in the Y direction.
  • Figure 7 is a visual solution to the error produced by the machine tool's interpolation profile in the Z direction.
  • Fig. 8 is an error of the pitch direction pitch angle ⁇ i generated by the machine tool interpolation contour visually solved.
  • Figure 9 is the error of the roll direction roll angle ⁇ i generated by the machine tool interpolation profile visually solved.
  • Figure 10 is the error of the yaw angle ⁇ i in the yaw direction generated by the machine tool interpolation contour.
  • the contour contour of the CNC machine tool is used as the research object, and the six-dimensional contour error is solved by the visual inspection method of the invention.
  • the six-dimensional solution steps of the isometric spiral contour error are as follows:
  • the first step is to design and install the measurement tooling and measurement system.
  • the measuring object is a self-built CNC machine tool 4, and the strokes of the CNC machine tool 4 in the X and Y directions are 800 mm and 900 mm, respectively.
  • the machine interpolation speed was chosen to be 3m/min.
  • the measurement system consists of a camera 1, a camera holder 2 and a measuring tool.
  • the camera 1 selected in the embodiment has a frame rate of 60 fps, a resolution of 3300 ⁇ 3300 pixels, and a camera 1 exposure time of 5000 ⁇ s.
  • the camera jig 2 can realize the six-pose adjustment of the camera 1.
  • the measuring tooling is composed of a base body 10, a high-brightness short-time lighting unit 9, a priori standard board 7, and an encoding element 8.
  • the measuring tool is mounted on a rotary table of a numerical control machine tool.
  • the a priori standard plate 7 is a glass transparent material having an outer dimension of 250 mm ⁇ 250 mm, and 196 photolithographically encoded decimal elements 8 are lithographically patterned thereon. Each code element 8 is sampled from [65, 637] and the code value is unique. The adjacent code element 8 has a center distance of 16 mm.
  • the assembly unit is assembled according to the exploded view of the tooling assembly.
  • the specific assembly sequence is: inserting the short-time high-brightness light-emitting unit 9 into the grooves on both sides of the base body 10, and placing the a priori standard plate 7 on the base body 10;
  • the a priori standard plate 7 is pressed by the two pressure plates 6, and the abutting standard plates 7 are pressed and fixed by the pressure plate 6 by the four platen locking bolts 5.
  • the assembled measuring tool is placed on the optical three-coordinate device platform to establish a priori standard plate global coordinate system O G X G Y G Z G 14, as shown in FIG.
  • the optical geometric relationship between the coding elements 8 is calibrated in the a priori standard plate global coordinate system O G X G Y G Z G 14 using an optical three-coordinate device, and the optical three-coordinate spatial measurement detection accuracy is 0.5 ⁇ m, and each coding element 8 is obtained.
  • the three-dimensional coordinates of the a priori standard plate global coordinate system O G X G Y G Z G 14 is
  • the calibrated measuring tool is fastened to the turntable of the numerically controlled machine tool 4 by means of the pressing bolt 11 and the pressing nut 12.
  • the camera 1 is fixed on the camera fixture 2 and adjusts the camera 1 six-dimensional attitude parameter to ensure that the camera fixture 2 is located above the measurement tool to collect measurements.
  • the imaging parameters of the camera 1 were adjusted, and the camera 1 measured the field of view 21 to be 40 mm x 40 mm, and the measurement distance was about 337 mm.
  • the high-precision two-dimensional checkerboard calibration plate is placed in the camera 1 measurement field of view 21 in 16 positions, and the camera 1 is triggered to capture the calibration plate image at each position.
  • the internal and external parameters and distortion parameters of camera 1 are calibrated using the calibration algorithm proposed by Zhang Zhengyou.
  • the five distortion coefficients are (0.10328, -0.23054, 0.00045, 0.00012, 0).
  • the four-dimensional interpolation isometric spiral contour image of the CNC machine tool is acquired and processed, and the camera 1 and the high-brightness short-time lighting unit 9 are synchronously triggered.
  • the X and Y axes of the numerical control machine tool 4 are driven to interpolate the contour to be tested, and the high-brightness short-time lighting unit 9 is set to have a lighting time of 700 ⁇ s per frame, thereby ensuring high contrast of the acquired sequence image.
  • the code value of the coded element 8 in each frame image is identified, and the center point pixel coordinate of the coded element 8 is located in combination with the formula (2) gray center of gravity method, and FIG. 4 is the coded element 8 identification and positioning image processing result. .
  • the fourth step is the six-dimensional solution of high dynamic range and arbitrary contour error of CNC machine tools.
  • the camera 1 used has a measurement field of view of 40 mm ⁇ 40 mm, and the a priori standard plate 7 has an outer dimension of 250 mm ⁇ 250 mm.
  • the measured isometric spiral has a contour range of [90 mm, 70 mm], and the camera 1 measures the field of view 21 less than The machine tool interpolation contour range.
  • the overall machine motion profile is characterized by an encoding element 8 selected on the first frame image. The position of this point in the invisible area is calculated by the spatial geometric relationship between the coding elements 8 on the large-size prior standard board 7, thereby realizing the measurement of the large-scale contour error of the numerical control machine tool 4.
  • the specific steps are as follows:
  • Step 1 Build a camera coordinate system O C X C Y C Z C 13 with reference to FIG. 3, and select the coding element of the sixth row and the ninth column with a code value of 235 in the first frame image.
  • the sixth row and the tenth column have a code value of 237 as the second code element.
  • Line 7 and column 9 are the fourth coding element of the coding element 8 having a code value of 283.
  • Step 2 The camera 1 captures a total of 500 images during the entire contour measurement process, and the embodiment analyzes and captures the 300th frame image, and selects the 9th row and the 8th column as the first coding element selected in the 300th frame.
  • the 9th row and the 9th column are selected as the second coding element of the 300th frame.
  • the 10th row and the 9th column are selected as the third coding element of the 300th frame.
  • the 10th row and the 8th column are selected as the fourth coding element of the 300th frame.
  • the coordinates of the four points in the global coordinate system O G -X G Y G Z G 14 of the a priori standard board are respectively (112.0010 , 128.0024, 0.0110), (127.9990, 128.0021, 0.0111), (128.0012, 144.0037, 0.0120) and (112.0008, 144.0057, 0.0146).
  • Calculate the coordinates of the four points in the local coordinate system O L300 -X L300 Y L300 Z L300 21 of the a priori standard board by formula (5) as (0,0,0), (15.9980, -0.0030, -0.0010), ( 16.0003, 16.0133, 0.0010) and (-0.0002, 16.0033, 0.0036).
  • the pixel coordinates of the image plane of the camera 1 are (1452.39, 1071.15), (2497.23, 683.88), (2884.75, 1729.45), and (1839.66, 2116.14).
  • the calculation results were (-10.9098 mm, -3.08975 mm, 337.696, -179.931°, 0.19436°, -0.00356046°).
  • Step 3 Calculate the 300th frame image machine coordinate system O M -X M Y M Z M 15 under M X 300 , M Y 300 , M Z 300 , pitch angle ⁇ 300 , roll angle ⁇ 300 , yaw angle ⁇ 300
  • the dimensional information is (12.3424, 2.3471, -0.051, -0.211°, 0.056°, -0.002°).
  • Step 4 traverses each frame of the image taken in accordance with the above method to obtain an interpolation contour under the machine coordinate system O M X M Y M Z M 15 characterized by the coding element 8 having an encoding value of 235, by comparing the actual six-degree-of-freedom contour L r and the nominal contour L m solve the six-degree-of-freedom error of the CNC machine tool 4 interpolation contour.
  • Figure 5, Figure 6, and Figure 7 respectively show the error of the machine tool interpolation contour in the X, Y, and Z directions.
  • Figure 8, Figure 9, and Figure 10 respectively show the pitch and deviation of the machine tool interpolation contour. Error in navigation and roll direction. It can be clearly seen from the figure that the error value of the machine tool interpolation contour can be conveniently obtained by this method.
  • the method adopts the monocular visual pose algorithm combined with the prior space constraints between the coding elements on the large-scale standard board to realize the six-dimensional measurement of the large-scale arbitrary contour error of the numerical control machine tool in the small field of view, and expands the measurable motion range and measurement dimension of the machine tool. .
  • the measurement system and method of the present invention are low in cost and simple in operation.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Optics & Photonics (AREA)
  • Mechanical Engineering (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Electromagnetism (AREA)
  • Theoretical Computer Science (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

数控机床(4)高动态大范围任意轮廓误差单目六维测量方法，属于机床动态误差检测领域，涉及一种单目视觉测量技术配合短时频闪照明、先验标准板(7)的数控机床(4)高动态任意大范围轮廓误差六维测量方法。该方法设计了测量工装和测量系统，利用单目视觉位姿算法结合先验知识提升视觉可测机床插补轮廓的维度与范围。结合误差分配原则，利用小测量视场(21)提升编码元(8)的视场测量精度；遍历拍摄每一帧图像即得到机床实际六维运动轮廓；通过与机床名义运动轮廓比较即可解算数控机床(4)插补轮廓产生的六维误差。通过单目视觉位姿算法结合基准转换方法，提高了视觉所测机床插补轮廓的维度；实现了小视场下数控机床(4)大范围任意轮廓误差六维度测量。

Description

数控机床高动态大范围任意轮廓误差单目六维测量方法 技术领域
本发明属于机床动态误差检测领域，涉及一种单目视觉测量技术配合短时频闪照明、先验标准板的数控机床高动态任意大范围轮廓误差六维测量方法。
背景技术
诸如船用螺旋桨、舰艇用螺旋推进器、涡轮机与发动机叶片等难加工变曲率零件被广泛用于国家重大装备工程中，此类零件的形位及几何轮廓精度直接影响着装备的工作性能。多轴联动机床相比于传统三轴数控机床凭借其在加工过程中刀具与工件位姿角的可随时调控性，解决了大多数复杂变曲率零件一次装夹、加工难题。各类难加工、变曲率零件由多轴数控机床在高动态条件下伴随特定插补轮廓加工而成，特别在高进给条件下，由于机床动态特性不足引起的数控机床误差尤为突出，使得机床运行过程中刀具与工件间的相对位置产生空间偏离，降低工件加工品质。轮廓误差是评估数控机床动态性能的重要指标。因此，机床高动态、大范围、任意轮廓空间误差(动态特性)定期评估是评价数控机床动态性能以及提升加工精度的重要保证。
现有的数控机床轮廓误差测量方法有球杆仪测量法、平面光栅测量法、R-test测量法与视觉测量法等。福州大学陈建雄在《International Journal of Machine Tools&Manufacture》第77卷第77期发表的《Geometric error measurement and identification for rotary table of multi-axis machine tool using double ball bar》提出了回转轴4项安装误差及6项体积误差球杆仪检测与辨识方法。采用两步法通过使安装球杆仪获得测量数据，根据误差辨识模型结合辨识算法分离4项安装误差及6项体积误差。虽球杆仪设备使用灵活性高，但其为一维测量设备，只能测量平面圆轮廓误差而不能实现任意轮廓误差测量。此外，受限于杆长机械结构，球杆仪很难测量能较好反映机床动态性能的小半径圆轮廓误差。瑞士学者B.Bringmann等在《CIRP Annals-Manufacturing Technology》第58卷第1期发表的《A method for direct evaluation of the dynamic 3D path accuracy of NC machine tools》提出了利用R-test测量的空间轮廓偏差来辅助辨识线性轴与回转轴误差的综合方法，在辨识的基础上通过加速度和加加速度参数再设置减小了机床轮廓运动偏差。R-test测量精度高，但其测量范围小，X、Y、Z单方向测量范围小于12mm，且在执行测量时会引入不相关轴的联动误差。日本京都大学Soichi Ibaraki等在《Laser Metrology&Machine Performance V》发表的《Diagnosis and compensation of motion errors in NC machine tools by arbitrary shape contouring error measurement》提出了基于平面光栅的机床安装及伺服系统误差测量与辨识方法。并结合数控位置的反馈信号补偿了伺服系统使机床轮廓精度提高。然而，此设备操作繁琐且无法测量转轴回转圆轮廓。
刘巍、严洪悦等人申请的发明专利CN 105798704A，“一种机床平面轮廓误差单目测量方法”，发明了一种数控机床轮廓误差单目测量方法，提高了测量效率并且降低了成本，实现了数控机床平面插补轮廓误差的二维测量。然而，此种方法受限于相机带宽，相机拍摄帧频提升有限，导致高进给条件下拍摄图像中的特征点模糊，降低了轮廓误差求解精度。此方法只能测量二维轮廓误差，不能实现数控机床轮廓误差的三维解算。
发明内容
本发明要解决的技术难题是克服现有技术缺陷，针对现有单一测量方法无法解决数控机床高动态、大范围、任意轮廓误差六维测量难题，发明了一种数控机床高动态大范围任意轮廓误差单目六维测量方法。设计了测量工装和测量系统，该方法结合误差分配原则，利用小测量视场提升编码元的视场测量精度；利用单目位姿算法结合先验知识提升视觉可测机床插补轮廓的维度与范围，整个机床运动轮廓由选定的一个参考元来表征，然后通过基准转换得到机床坐标系下参考元所表征的机床插补轮廓的X、Y、Z、俯仰、滚转及偏航六维信息，采用该方法遍历拍摄每一帧图像即得到机床实际六维运动轮廓；通过与机床名义运动轮廓比较即可解算数控机床插补轮廓产生的六维误差。该方法的测量系统成本低，操作简便。
本发明采用的技术方案是一种数控机床高动态大范围任意轮廓误差单目六维测量方法，其特征是，该方法设计了测量工装和测量系统，利用单目视觉位姿算法结合先验知识提升视觉可测机床插补轮廓的维度与范围，整个机床运动轮廓由选定的一个参考元来表征；结合误差分配原则，利用小测量视场提升编码元的视场测量精度；然后通过基准转换得到机床坐标系下参考元所表征的机床插补轮廓的X、Y、Z、俯仰、滚转及偏航六维信息，采用该方法遍历拍摄每一帧图像即得到机床实际六维运动轮廓；通过与机床名义运动轮廓比较即可解算数控机床插补轮廓产生的六维误差；方法的具体步骤如下：
第一步 安装测量工装与测量系统
测量工装由基体10、高亮度短时发光单元9、先验标准板7和编码元8组成，先验标准板7为透明基底材料，其上分布有编码值唯一且呈矩阵分布的编码元8；测量工装安装时将短时高亮度发光单元9固定在基体10两侧的凹槽中；将先验标准板7支撑在基体10上，利用两个压板6压紧先验标准板7；利用四个压板锁紧螺栓5通过压板6压紧并固定先验标准板7；
测量系统包括相机1、相机夹具2和测量工装；相机1固定在相机夹具2上，相机夹具2安装在测量工装上方，以采集测量工装运动过程中的序列图像；将组装好的测量工装放置在光学三坐标设备平台上，采用光学三坐标设备在先验标准板全局坐标系下校准编码元8间的空间几何关系；测量系统布置时，将校准好的测量工装通过压紧螺栓11与压紧螺母12紧固在数控机床4的回转台3上用于表征机床运动信息；
第二步 建立先验标准板全局坐标系
在测量工装上建立先验标准板全局坐标系OGXGYGZG14，其原点建立在第一行第一列的编码元8中心上，定义为OG；XG坐标轴的方向由原点OG向下指向阵列上第一列最后一行编码元8的中心点；YG坐标轴的方向由OG向右指向阵列上第一行最后一列编码元8的中心点；ZG坐标轴由右手法则确定；采用光学三坐标设备在先验标准板全局坐标系OGXGYGZG14下校准编码元8间的空间几何关系，得到各个编码元8在先验标准板全局坐标系OGXGYGZG14下的三维坐标；先验标准板7上的编码元8承载着数控机床4的运动信息，编码元8间的空间位置关系经过高精度设备校准，在保证校准精度的前提下先验标准板7的尺寸可以制造的尽可能大以满足轮廓误差大范围测量需求；
第三步 相机标定
相机成像模型表达了相机坐标系与世界坐标系的一一映射关系，带有畸变参数的相机成像模型为：
Figure PCTCN2017109782-appb-000001
其中，(Xw,Yw,Zw)为世界坐标系下编码元8中心点的三维坐标，K为相机1的内参数矩阵，T为相机1的外参数矩阵，(u,v)为编码元8中心点在像平面的二维坐标，(u0,v0)为图像的主点坐标，(Cx,Cy)为横纵方向的等效焦距，
Figure PCTCN2017109782-appb-000002
分别为相机坐标系与世界坐标系间的旋转和平移变换矩阵，(δx,δy)为因光学系统不完善所引起的像点在x,y方向的畸变量；采用棋盘格标定板在相机1测量视场21内摆放多个位置获取标定板图像，通过张正友提出的标定算法标定相机1畸变参数以及内外参数矩阵；
第四步 数控机床高动态、大范围插补轮廓高清晰无模糊采集与图像处理
在完成测量工装安装、布置的基础上，采集数控机床4插补轮廓图像；由于轮廓误差测量精度要求精度高，所需拍摄测量视场21小；首先调整相机1参数使其处于最佳拍摄视场与帧频下；随后，同步触发相机1与高亮度短时发光单元9，设置高亮度短时发光单元9的发光时间与发光强度，保证高亮度短时发光单元9在相机1曝光时间内透过先验标准板7基底为编码元8补光；选择能反映机床动态性能的高进给机床速度，依照程序指令驱动数控机床4各运动轴插补待测轮廓；在机床图像采集过程中，相机1固定不动，机床运动，在高亮度短时发光单元9的辅助下采集到编码元8的清晰无模糊序列图像；
图像采集后，识别图像上每个编码元8代表的编码值并利用灰度重心法定位解码后的每个编码元8的中心点的二维像素坐标；采用灰度重心法提取算法定位球形标志点中心，其计算表达式为：
Figure PCTCN2017109782-appb-000003
其中，(i,j)代表图像像素点坐标，m，n为图像在横、纵方向的所含的像素的数量；(x,y)为图像的质心坐标，f(i,j)为像素坐标(i,j)处的灰度值；
第五步 数控机床高动态大范围任意轮廓误差六维度解算
本方法结合误差分配原则，利用小测量视场21提升视场内编码元8测量精度；利用单目位姿算法结合先验知识提升视觉可测机床插补轮廓的维度与范围；整个机床运动轮廓由选定的一个参考元来表征， 在测量视场21不可见区域该点的位置由可见区域点的像素坐标结合高精度先验约束解算求得；通过基准转换并遍历所有图像得到该参考元表征在机床坐标系下的运动轮廓，通过与机床名义运动轮廓比较即可解算数控机床4插补轮廓的六维误差；数控机床高动态大范围任意轮廓误差六维度解算步骤具体为：
相机1测量视场21为N×N(单位mm)，先验标准板7外形尺寸为M×M(单位mm),N远小于M；涉及的坐标系除上述的先验标准板全局坐标系OGXGYGZG14外，还涉及相机坐标系OCXCYCZC13、机床坐标系OMXMYMZM15和先验标准板局部坐标系OLiXLiYLiZLi22；相机坐标系OCXCYCZC13原点建立在光心OC处；数控机床4未运动时，在第一帧图像中选定位于视野中成矩形排列的四个编码元
Figure PCTCN2017109782-appb-000004
和
Figure PCTCN2017109782-appb-000005
选定编码元
Figure PCTCN2017109782-appb-000006
作为参考元；数控机床4在各个轴的插补运动轴合成的运动轮廓通过编码元
Figure PCTCN2017109782-appb-000007
来表征；其在先验标准板全局坐标系OGXGYGZG14下的坐标为
Figure PCTCN2017109782-appb-000008
以
Figure PCTCN2017109782-appb-000009
为原点建立机床坐标系OMXMYMZM15，机床坐标系OMXMYMZM15各坐标轴与数控机床4各运动轴方向一致；控制机床带动测量工装3分别沿机床X轴方向移动数个位置，利用单目位姿位姿求解算法解算每个位置下
Figure PCTCN2017109782-appb-000010
相对于相机坐标系OCXCYCZC13三维坐标(x,y,z)，在此基础上拟合X轴方向向量；按照相同规则确定机床坐标系OMXMYMZM15的Y轴，机床坐标系OMXMYMZM15的Z轴由右手法则来确定；按如下公式建立X、Y轴：
Figure PCTCN2017109782-appb-000011
其中，
Figure PCTCN2017109782-appb-000012
为首帧图像中编码元P1116在相机坐标系OCXCYCZC13下的三维坐标；(x',y',z')为测量工装沿机床Y轴方向移动数个位置，并用单目位姿求解算法解算的每个位置下
Figure PCTCN2017109782-appb-000013
点相对于相机坐标系OCXCYCZC13的三维坐标；(mx,nx,px)为机床坐标系OMXMYMZM15的X轴方向向量，(my,ny,py)为机床坐标系OMXMYMZM15的Y轴方向向量，(CX,CY,CZ)为相机坐标系OCXCYCZC13下一点的三维坐标，(MX,MY,MZ)为机床坐标系OMXMYMZM15下一点的三维坐标，
Figure PCTCN2017109782-appb-000014
为相机坐标系OCXCYCZC13与机床坐标系OMXMYMZM15之间的转换矩阵；
测量时，先验标准板7随机床不断做插补运动，其上的编码元8不断的在相机1上成像；在数控机床4运动过程中，相机1共采集G帧图像，在第i帧图像中出现在视野中成矩形排列的四个编码元8为
Figure PCTCN2017109782-appb-000015
该四个编码元8中心在先验标准板全局坐标系OGXGYGZG14下的坐标为
Figure PCTCN2017109782-appb-000016
对应的在图像上的二维像素坐标
Figure PCTCN2017109782-appb-000017
建立第i帧下先验标准板局部坐标系OLiXLiYLiZLi22，(i＝1,2…G)；此坐标系以
Figure PCTCN2017109782-appb-000018
23为坐标原点，XLi、YLi坐标轴方向分别平行于与先验标准板全局坐标系 OGXGYGZG14的XG、YG方向，ZLi坐标轴由右手法则确定；则选定的四个编码元8中心在先验标准板局部坐标系OLiXLiYLiZLi22的三维坐标为：
Figure PCTCN2017109782-appb-000019
其中，Ti为第i帧图像先验标准板全局坐标系OGXGYGZG14与先验标准板局部坐标系先验标准板局部坐标系OLiXLiYLiZLi22间的转换矩阵；对于第i帧图像，i＝1,2…G，计算：
Figure PCTCN2017109782-appb-000020
其中，Xi为相机坐标系OCXCYCZC13下光心OC到第i帧先验标准板7上
Figure PCTCN2017109782-appb-000021
点的距离，Yi为相机坐标系OCXCYCZC13下光心OC到第i帧先验标准板7上
Figure PCTCN2017109782-appb-000022
点的距离，Zi为相机坐标系OCXCYCZC13下光心OC到第i帧先验标准板
Figure PCTCN2017109782-appb-000023
点的距离；a'为第i帧先验标准板7全局坐标系OGXGYGZG14下
Figure PCTCN2017109782-appb-000024
与
Figure PCTCN2017109782-appb-000025
间的距离；b'为第i帧先验标准板全局坐标系OGXGYGZG14下
Figure PCTCN2017109782-appb-000026
与
Figure PCTCN2017109782-appb-000027
间的距离；c'为第i帧先验标准板全局坐标系OGXGYGZG14下
Figure PCTCN2017109782-appb-000028
与
Figure PCTCN2017109782-appb-000029
间的距离；α为直线
Figure PCTCN2017109782-appb-000030
与
Figure PCTCN2017109782-appb-000031
间的夹角
Figure PCTCN2017109782-appb-000032
β为直线
Figure PCTCN2017109782-appb-000033
与
Figure PCTCN2017109782-appb-000034
间的夹角
Figure PCTCN2017109782-appb-000035
γ为直线
Figure PCTCN2017109782-appb-000036
与
Figure PCTCN2017109782-appb-000037
间的夹角
Figure PCTCN2017109782-appb-000038
Figure PCTCN2017109782-appb-000039
Figure PCTCN2017109782-appb-000040
Figure PCTCN2017109782-appb-000041
令k＝2cosα，q＝2cosβ，r＝2cosγ，c′2＝vZi 2，a′2＝ac′2＝avZi 2，b'2＝bc'2＝bvZ2，Xi＝xZi，Yi＝bZi，
Figure PCTCN2017109782-appb-000042
和
Figure PCTCN2017109782-appb-000043
为第i帧中的非共面且位置关系已知的三个空间点，满足k2+q2+r2-kqr-1≠0；方程(5)可改写为公式(6)：
Figure PCTCN2017109782-appb-000044
采用优化算法排除公式(6)的退化解得到4个可接受的零解，利用
Figure PCTCN2017109782-appb-000045
作为先验约束得到Xi、Yi和Zi的唯一解；则相机坐标系OCXCYCZC13下第i帧先验标准板7上
Figure PCTCN2017109782-appb-000046
和
Figure PCTCN2017109782-appb-000047
点的三维坐标
Figure PCTCN2017109782-appb-000048
可表示为如下公式：
Figure PCTCN2017109782-appb-000049
在已知
Figure PCTCN2017109782-appb-000050
和
Figure PCTCN2017109782-appb-000051
的基础上，利用Kabsch算法求解第i帧先验标准板局部坐标系OLiXLiYLiZLi22相对于相机坐标系OCXCYCZC13的旋转矩阵R′i与平移矩阵t′i，t′i即为第i帧先验标准板局部坐标系OLiXLiYLiZLi22原点
Figure PCTCN2017109782-appb-000052
在相机坐标系OCXCYCZC13下的三维坐标
Figure PCTCN2017109782-appb-000053
通过R′i分离求解两坐标系间的俯仰角θ′i、滚转角Φ′i、偏航角ψ′i；根据先验标准板7编码元8间已知空间约束解算参考元
Figure PCTCN2017109782-appb-000054
在第i帧中相机坐标系OCXCYCZC13下的三维坐标
Figure PCTCN2017109782-appb-000055
Figure PCTCN2017109782-appb-000056
测量的轮廓误差是表征在机床坐标系OMXMYMZM15下的，需将相机1测量的轮廓进行坐标转换以正确求解机床轮廓误差；机床坐标系OMXMYMZM15与先验标准板局部坐标系OLiXLiYLiZLi22的位姿转换公式为：
Figure PCTCN2017109782-appb-000057
其中，Mi为第i帧图像下先验标准板局部坐标系OLiXLiYLiZLi22与机床坐标系OMXMYMZM15间的转换矩阵；从Mi中分离出第i帧标准板相对于机床坐标系OMXMYMZM15的俯仰角θi、滚转角Φi、偏航角ψi；遍历所有图像序列得到由参考元表征的的整个机床六自由度轮廓Lr，Lr(MXi，MYi，MZi，θi，Φi，ψi)，i＝1,2…G；通过比较实际六自由度轮廓Lr与名义轮廓Lm求解数控机床4插补产生的六自由度轮廓误差E：
E＝Lr-Lm。         (9)
本发明的有益之处与现有视觉方法相比提高了视觉可测的数控机床插补速度，突破了机床可测速度限制，拓展了视觉检测的数控机床插补进给速度范围。此外，为了保证轮廓误差测量精度，相机测量较小，本发明采用视觉位姿算法结合大尺寸标准板上编码元间的先验空间约束，实现了小视场下数控机床大范围任意轮廓误差六维度测量，拓展了机床可测运动范围与测量维度。在测量过程中仅采用单个相机、特制的测量工装和测量系统，降低成本的同时提升操作的简便性。
附图说明
图1为数控机床高动态大范围任意轮廓误差六维测量系统图。其中，1-相机，2-相机夹具，3-回转台，4-数控机床。
图2为测量工装爆炸视图。其中，5-压板锁紧螺栓，6-压板，7-先验标准板，8-编码元，9-短时高 亮度发光单元，10-基体，11-压紧螺栓，12-压紧螺母。
图3为机床未运动时测量系统和各坐标系布置图。其中，13-相机坐标系OCXCYCZC，14-先验标准板全局坐标系OGXGYGZG，15-机床坐标系OMXMYMZM，16-编码元
Figure PCTCN2017109782-appb-000058
17-编码元
Figure PCTCN2017109782-appb-000059
18-编码元
Figure PCTCN2017109782-appb-000060
19-编码元
Figure PCTCN2017109782-appb-000061
20-首帧先验标准板局部坐标系OL1XL1YL1ZL1，21-相机测量视场。
图4为第300帧图像编码元识别与定位结果。
图5为视觉求解机床插补轮廓在X方向上产生的误差。
图6为视觉求解机床插补轮廓在Y方向上产生的误差。
图7为视觉求解机床插补轮廓在Z方向上产生的误差。
图8为视觉求解的机床插补轮廓产生的俯仰方向俯仰角θi的误差。
图9为视觉求解的机床插补轮廓产生的滚转方向滚转角Φi的误差。
图10为视觉求解的机床插补轮廓产生的偏航方向偏航角ψi的误差。
具体实施方式
以下结合技术方案和附图详细叙述本发明的具体实施方式。
为了体现轨迹的任意性，以数控机床平面插补等角螺旋线轮廓为研究对象，利用本发明的视觉检测方法对轮廓六维误差进行求解。待测量的等角螺旋线轮廓方程为，r＝0.189e0.221θ，θ＝[0,7.3π]。等角螺旋线轮廓误差六维求解步骤具体如下：
第一步设计安装测量工装与测量系统
如附图1所示，测量对象为自行搭建的数控机床4，数控机床4在X、Y方向上的行程分别为800mm和900mm。为了反映机床机床的动态性能，机床插补速度选为3m/min。测量系统包含相机1，相机夹具2和测量工装。实施例中选用的相机1帧频为60fps，分辨率为3300×3300像素，相机1曝光时间为5000μs。相机夹具2可实现相机1的六姿态调整。测量工装由基体10、高亮度短时发光单元9、先验标准板7和编码元8组成，测量工装安装在数控机床回转台上，
附图2为测量工装爆炸视图，先验标准板7为玻璃透明材料，外形尺寸为250mm×250mm，其上光刻有196个呈矩阵排布的十进制编码元8。每个编码元8从[65，637]中抽样取值，码值唯一。相邻编码元8中心距为16mm。
按照附图2所示测量工装爆炸图组装各单元，具体组装顺序为：将短时高亮度发光单元9插入到基体10两侧的凹槽中，将先验标准板7放到基体10上；利用两个压板6压紧先验标准板7，利用四个压板锁紧螺栓5通过压板6压紧并固定先验标准板7。
将组装好的测量工装放置在光学三坐标设备平台上，建立先验标准板全局坐标系OGXGYGZG14，如图3所示。采用光学三坐标设备在先验标准板全局坐标系OGXGYGZG14下校准编码元8间的空间几何关系，光学三坐标空间测量检测精度为0.5μm，得到各个编码元8在先验标准板全局坐标系OGXGYGZG14下的三维坐标。
将校准好的测量工装通过压紧螺栓11与压紧螺母12紧固在数控机床4的回转台上。安装时，相机1固定在相机夹具2上并调整相机1六维姿态参数，确保相机夹具2位于测量工装上方以采集测量 工装运动过程中的编码元8的序列图像。调整相机1成像参数，相机1测量视场21为40mm×40mm，测量距离约为337mm。
第二步相机标定
将高精度二维棋盘格标定板放在相机1测量视场21中摆放16个位置，在每个位置下触发相机1拍摄标定板图像。结合公式(1)采用张正友提出的标定算法标定相机1的内外参数以及畸变参数。标定的横纵方向的等效焦距(αx，αy)＝(14959.25，14959.68)，图像的主点坐标为(u0，ν0)＝(2557.11，2599.79)，求解的用于表达畸变量的五个畸变系数为(0.10328，-0.23054，0.00045，0.00012，0)。
第三步数控机床高动态大范围插补轮廓高清晰无模糊采集与图像处理
采集与处理数控机床4平面插补等角螺旋线轮廓图像，同步触发相机1与高亮度短时发光单元9。驱动数控机床4的X、Y轴插补待测轮廓，设置高亮度短时发光单元9在每一帧的发光时间为700μs，保证采集序列图像的高对比度。采集完图像后识别每一帧图像中编码元8的码值，并结合公式(2)灰度重心法定位编码元8的中心点像素坐标，附图4为编码元8识别与定位图像处理结果。
第四步数控机床高动态大范围任意轮廓误差六维度解算
采用的相机1测量视场21为40mm×40mm，先验标准板7外形尺寸为250mm×250mm，所测量的等角螺旋线的轮廓范围为[90mm，70mm]，相机1测量视场21小于被测机床插补轮廓范围。整个机床运动轮廓由在第一帧图像上选定的一个编码元8表征。在不可见区域此点的位置通过大尺寸先验标准板7上的编码元8间的空间几何关系推算，以此实现数控机床4大范围轮廓误差测量。具体步骤过程为：
步骤1参照附图3建立相机坐标系OCXCYCZC13，在首帧图像中选取先验标准板上的第6行第9列码值为235的编码元
Figure PCTCN2017109782-appb-000062
作为参考元，第6行第10列码值为237的编码元8作为第二个编码元
Figure PCTCN2017109782-appb-000063
第7行第10列码值为285的编码元8作为第三个编码元
Figure PCTCN2017109782-appb-000064
第7行第9列作为码值为283的编码元8第四个编码元
Figure PCTCN2017109782-appb-000065
以码值为235的编码元
Figure PCTCN2017109782-appb-000066
为原点建立首帧先验标准板局部坐标系OL1XL1YL1ZL120以及机床坐标系OMXMYMZM15，根据公式(3)建立相机坐标系OCXCYCZC13与机床坐标系OMXMYMZM15间的转换关系为：
Figure PCTCN2017109782-appb-000067
步骤2在整个轮廓测量过程中相机1共拍摄500张图像，实施例分析拍摄到第300帧图像，选定第9行第8列作为第300帧选定的第一个编码元
Figure PCTCN2017109782-appb-000068
选定第9行第9列作为第300帧的第二个编码元
Figure PCTCN2017109782-appb-000069
选定第10行第9列作为第300帧的第三个编码元
Figure PCTCN2017109782-appb-000070
选定第10行第8列作为第300帧的第四个编码元
Figure PCTCN2017109782-appb-000071
建立第300帧下先验标准板局部坐标系OL300-XL300YL300ZL30021，该四点在先验标准板全局坐标系OG-XGYGZG14的坐标分别为(112.0010，128.0024，0.0110)、(127.9990，128.0021，0.0111)、(128.0012，144.0037，0.0120)和(112.0008，144.0057，0.0146)。通过公式(5)计算该四点在先验标准板局部坐标系OL300-XL300YL300ZL30021下的坐标为(0，0，0)、(15.9980，-0.0030，-0.0010)、(16.0003，16.0013，0.0010)和(-0.0002，16.0033，0.0036)。在相机1像平面的像素坐标为(1452.39， 1071.15)、(2497.23，683.88)、(2884.75，1729.45)和(1839.66，2116.14)。根据公式5～7计算第300帧图像先验标准板7在相机坐标系OCXCYCZC13下的CX300、CY300、CZ300、俯仰角θ'300、滚转角Φ′300、偏航角ψ'300。计算结果为(-10.9098mm，-3.08975mm，337.696，-179.931°，0.19436°，-0.00356046°)。
步骤3计算第300帧图像机床坐标系OM-XMYMZM15下的MX300、MY300、MZ300、俯仰角θ300、滚转角Φ300、偏航角ψ300六维信息为(12.3424，2.3471，-0.051，-0.211°，0.056°，-0.002°)。
步骤4按照上述方法遍历拍摄的每一帧图像，得到采用编码值为235的编码元8表征的机床坐标系OMXMYMZM15下的插补轮廓，通过比较实际六自由度轮廓Lr与名义轮廓Lm求解数控机床4插补轮廓的六自由度误差。图5、图6、图7分别为视觉求解机床插补轮廓在X、Y、Z方向上产生的误差，图8、图9、图10分别为视觉求解的机床插补轮廓产生的俯仰、偏航以及滚转方向的误差。由图中可以清楚的看出利用该方法可以方便的求出机床插补轮廓的误差值。
该方法采用单目视觉位姿算法结合大尺寸标准板上编码元间的先验空间约束，实现了小视场下数控机床大范围任意轮廓误差六维度测量，拓展了机床可测运动范围与测量维度。本发明的测量系统及方法成本低，操作简便。

Claims (1)

  1. 一种数控机床高动态大范围任意轮廓误差单目六维测量方法，其特征是，该方法采用了特殊设计的测量工装和测量系统，利用单目视觉位姿算法结合先验知识提升视觉可测机床插补轮廓的维度与范围，整个机床运动轮廓由选定的一个参考元来表征；结合误差分配原则，利用小测量视场提升编码元的视场测量精度；然后通过基准转换得到机床坐标系下参考元所表征的机床插补轮廓的X、Y、Z向、俯仰、滚转及偏航六维信息，采用该方法遍历拍摄每一帧图像即得到机床实际六维运动轮廓；通过与机床名义运动轮廓比较解算出数控机床插补轮廓产生的六维误差；方法的具体步骤如下：
    第一步 安装的特制测量工装与测量系统
    特制的测量工装由基体(10)、高亮度短时发光单元(9)、先验标准板(7)和编码元(8)组成，先验标准板(7)为透明基底材料，其上分布有编码值唯一且呈矩阵分布的编码元(8)；测量工装安装时将短时高亮度短时发光单元(9)插入基体(10)两侧的凹槽中；将先验标准板(7)支撑在基体(10)上，利用两个压板(6)压紧先验标准板(7)；利用四个压板锁紧螺栓(5)通过压板(6)压紧并固定先验标准板(7)；
    测量系统包括相机(1)、相机夹具(2)和测量工装；相机(1)固定在相机夹具(2)上，相机夹具(2)安装在测量工装上方，以采集测量工装运动过程中的序列图像；将组装好的测量工装放置在光学三坐标设备平台上，采用光学三坐标设备在先验标准板全局坐标系下校准编码元(8)间的空间几何关系；测量系统布置时，将校准好的测量工装通过压紧螺栓(11)与压紧螺母(12)紧固在数控机床(4)的回转台(3)上；
    第二步 建立先验标准板全局坐标系
    在测量工装上建立先验标准板全局坐标系OGXGYGZG(14)，其原点建立在第一行第一列的编码元(8)中心上，定义为OG；XG坐标轴的方向由原点OG向下指向阵列上第一列最后一行编码元(8)的中心点；YG坐标轴的方向由OG向右指向阵列上第一行最后一列编码元(8)的中心点；ZG坐标轴由右手法则确定；采用光学三坐标设备在先验标准板全局坐标系OGXGYGZG(14)下校准编码元(8)间的空间几何关系，得到各个编码元(8)在先验标准板全局坐标系OGXGYGZG(14)下的三维坐标；先验标准板(7)上的编码元(8)承载着数控机床(4)的运动信息，编码元(8)间的空间位置关系经过高精度设备校准，在保证校准精度的前提下先验标准板(7)的尺寸可以制造的尽可能大以满足轮廓误差大范围测量需求；
    第三步 相机标定
    相机成像模型表达了相机坐标系与世界坐标系的一一映射关系，带有畸变参数的相机成像模型为：
    Figure PCTCN2017109782-appb-100001
    其中，(Xw,Yw,Zw)为世界坐标系下编码元(8)中心点的三维坐标，K为相机(1)的内参数矩阵，T为相机(1)的外参数矩阵，(u,v)为编码元(8)中心点在像平面的二维坐标，(u0,v0)为图像的主点坐 标，(Cx,Cy)为横纵方向的等效焦距，
    Figure PCTCN2017109782-appb-100002
    I分别为相机坐标系与世界坐标系间的旋转和平移变换矩阵，(δx,δy)为因光学系统不完善所引起的像点在x,y方向的畸变量；采用棋盘格标定板在相机(1)测量视场(21)内摆放多个位置获取标定板图像，通过张正友提出的标定算法标定相机(1)畸变参数以及内外参数矩阵；
    第四步 数控机床高动态、大范围插补轮廓高清晰无模糊采集与图像处理
    在完成测量工装安装、布置的基础上，采集数控机床(4)插补轮廓图像；由于轮廓误差测量精度要求精度高，所需拍摄测量视场(21)小；首先调整相机(1)参数使其处于最佳拍摄视场与帧频下；随后，同步触发相机(1)与高亮度短时发光单元(9)，设置高亮度短时发光单元(9)的发光时间与发光强度，保证高亮度短时发光单元(9)在相机(1)曝光时间内透过先验标准板7基底为编码元(8)补光；选择能反映机床动态性能的高进给机床速度，依照程序指令驱动数控机床(4)各运动轴插补待测轮廓；在机床图像采集过程中，相机(1)固定不动，机床运动，在高亮度短时发光单元(9)的辅助下采集到编码元(8)的清晰无模糊序列图像；
    图像采集后，识别图像上每个编码元(8)代表的编码值并利用灰度重心法定位解码后的每个编码元(8)的中心点的二维像素坐标；采用灰度重心法提取算法定位球形标志点中心，其计算表达式为：
    Figure PCTCN2017109782-appb-100003
    其中，(i,j)代表图像像素点坐标，m，n为图像在横、纵方向的所含的像素的数量；(x,y)为图像的质心坐标，f(i,j)为像素坐标(i,j)处的灰度值；
    第五步 数控机床高动态大范围任意轮廓误差六维度解算
    本方法结合误差分配原则，利用小测量视场(21)提升视场内编码元(8)测量精度；利用单目位姿算法结合先验知识提升视觉可测机床插补轮廓的维度与范围；整个机床运动轮廓由选定的一个参考元来表征，在测量视场(21)不可见区域该点的位置由可见区域点的像素坐标结合高精度先验约束解算求得；通过基准转换并遍历所有图像得到该参考元表征在机床坐标系下的运动轮廓，通过与机床名义运动轮廓比较即可解算数控机床(4)插补轮廓的六维误差；数控机床高动态大范围任意轮廓误差六维度解算步骤具体为：
    相机(1)测量视场(21)为N×N(单位mm)，先验标准板(7)外形尺寸为M×M(单位mm),N远小于M；涉及的坐标系除上述的先验标准板全局坐标系OG-XGYGZG(14)外，还涉及相机坐标系OC-XCYCZC(13)、机床坐标系OM-XMYMZM(15)和先验标准板局部坐标系OLi-XLiYLiZLi(22)；相机坐标系OC-XCYCZC(13)原点建立在光心OC处；数控机床(4)未运动时，在第一帧图像中选定位于视野中成矩形排列的四个编码元
    Figure PCTCN2017109782-appb-100004
    和
    Figure PCTCN2017109782-appb-100005
    选定编码元
    Figure PCTCN2017109782-appb-100006
    作为参考元；数控机床(4)在各个轴的插补运动轴合成的运动轮廓通过编码元
    Figure PCTCN2017109782-appb-100007
    来表征；其在先验标准板全局坐标系OG-XGYGZG(14)下的坐标为
    Figure PCTCN2017109782-appb-100008
    以
    Figure PCTCN2017109782-appb-100009
    为原点建立机床坐标系OM-XMYMZM(15)，机床坐标系OM-XMYMZM(15)各坐标轴与数控机床(4)各运动轴方向一致；控制机床带动测量工装分别沿机床X轴方向移动数个位置，利用单目位姿位姿求解算法解算每个位置下
    Figure PCTCN2017109782-appb-100010
    相对于相机坐标系OC-XCYCZC(13)三维坐标(x,y,z)，在此基础上拟合X轴方向向量；按照相同规则确定机床坐标系OM-XMYMZM(15)的Y轴，机床坐标系OM-XMYMZM(15)的Z轴由右手法则来确定；按如下公式建立X、Y轴：
    Figure PCTCN2017109782-appb-100011
    其中，
    Figure PCTCN2017109782-appb-100012
    为首帧图像中编码元
    Figure PCTCN2017109782-appb-100013
    在相机坐标系OCXCYCZC(13)下的三维坐标；(x',y',z')为测量工装沿机床Y轴方向移动数个位置，并用单目位姿求解算法解算的每个位置下P1116点相对于相机坐标系OCXCYCZC13的三维坐标；(mx,nx,px)为机床坐标系OMXMYMZM(15)的X轴方向向量，(my,ny,py)为机床坐标系OMXMYMZM(15)的Y轴方向向量，(CX,CY,CZ)为相机坐标系OCXCYCZC(13)下一点的三维坐标，(MX,MY,MZ)为机床坐标系OMXMYMZM(15)下一点的三维坐标，
    Figure PCTCN2017109782-appb-100014
    M为相机坐标系OCXCYCZC13与机床坐标系OMXMYMZM(15)之间的转换矩阵；
    测量时，先验标准板(7)随机床不断做插补运动，其上的编码元(8)不断的在相机(1)上成像；在数控机床(4)运动过程中，相机(1)共采集G帧图像，在第i帧图像中出现在视野中成矩形排列的四个编码元(8)为
    Figure PCTCN2017109782-appb-100015
    该四个编码元(8)中心在先验标准板全局坐标系OGXGYGZG(14)下的坐标为
    Figure PCTCN2017109782-appb-100016
    对应的在图像上的二维像素坐标为
    Figure PCTCN2017109782-appb-100017
    建立第i帧下先验标准板局部坐标系OLiXLiYLiZLi(22)，(i＝1,2…G)；此坐标系以
    Figure PCTCN2017109782-appb-100018
    为坐标原点，XLi、YLi坐标轴方向分别平行于与先验标准板全局坐标系OGXGYGZG14的XG、YG方向，ZLi坐标轴由右手法则确定；则选定的四个编码元(8)中心在先验标准板局部坐标系OLiXLiYLiZLi(22)的三维坐标为：
    Figure PCTCN2017109782-appb-100019
    其中，Ti为第i帧图像先验标准板全局坐标系OGXGYGZG(14)与先验标准板局部坐标系先验标准板局部坐标系OLiXLiYLiZLi(22)间的转换矩阵；对于第i帧图像，i＝1,2…G，计算：
    Figure PCTCN2017109782-appb-100020
    其中，Xi为相机坐标系OCXCYCZC(13)下光心OC到第i帧先验标准板(7)上
    Figure PCTCN2017109782-appb-100021
    点的距离，Yi为相机坐标系OCXCYCZC(13)下光心OC到第i帧先验标准板7上
    Figure PCTCN2017109782-appb-100022
    点的距离，Zi为相机坐标系OCXCYCZC(13)下光心OC到第i帧先验标准板
    Figure PCTCN2017109782-appb-100023
    点的距离；a'为第i帧先验标准板(7)全局坐标系OGXGYGZG下
    Figure PCTCN2017109782-appb-100024
    与
    Figure PCTCN2017109782-appb-100025
    间的距离；b'为第i帧先验标准板全局坐标系OGXGYGZG(14)下
    Figure PCTCN2017109782-appb-100026
    与
    Figure PCTCN2017109782-appb-100027
    间的距离；c'为第i帧先验标准板全局坐标系OGXGYGZG(14)下
    Figure PCTCN2017109782-appb-100028
    与
    Figure PCTCN2017109782-appb-100029
    间的距离；α为直线
    Figure PCTCN2017109782-appb-100030
    与
    Figure PCTCN2017109782-appb-100031
    间的夹角
    Figure PCTCN2017109782-appb-100032
    β为直线
    Figure PCTCN2017109782-appb-100033
    与
    Figure PCTCN2017109782-appb-100034
    间的夹角
    Figure PCTCN2017109782-appb-100035
    γ为直线
    Figure PCTCN2017109782-appb-100036
    与
    Figure PCTCN2017109782-appb-100037
    间的夹角
    Figure PCTCN2017109782-appb-100038
    Figure PCTCN2017109782-appb-100039
    Figure PCTCN2017109782-appb-100040
    Figure PCTCN2017109782-appb-100041
    令k＝2cosα，q＝2cosβ，r＝2cosγ，c'2＝vZi 2，a'2＝ac'2＝avZi 2，b'2＝bc'2＝bvZ2，Xi＝xZi，Yi＝bZi，
    Figure PCTCN2017109782-appb-100042
    和
    Figure PCTCN2017109782-appb-100043
    为第i帧中的非共面且位置关系已知的三个空间点，满足k2+q2+r2-kqr-1≠0；方程(5)改写为公式(6)：
    Figure PCTCN2017109782-appb-100044
    采用优化算法排除公式(6)的退化解得到4个可接受的零解，利用
    Figure PCTCN2017109782-appb-100045
    作为先验约束得到Xi、Yi和Zi的唯一解；则相机坐标系OCXCYCZC(13)下第i帧先验标准板(7)上
    Figure PCTCN2017109782-appb-100046
    和
    Figure PCTCN2017109782-appb-100047
    点的三维坐标
    Figure PCTCN2017109782-appb-100048
    可表示为如下公式：
    Figure PCTCN2017109782-appb-100049
    在已知
    Figure PCTCN2017109782-appb-100050
    和
    Figure PCTCN2017109782-appb-100051
    的基础上，利用Kabsch算法求解第i帧先验标准板局部坐标系OLiXLiYLiZLi22相对于相机坐标系OC-XCYCZC(13)的旋转矩阵Ri'与平移矩阵ti'，ti'即为第i帧先验标准板局部坐标系OLiXLiYLiZLi(22)原点
    Figure PCTCN2017109782-appb-100052
    在相机坐标系OCXCYCZC(13)下的三维坐标ti'＝
    Figure PCTCN2017109782-appb-100053
    通过Ri'分离求解两坐标系间的俯仰角θi'、滚转角Φi'、偏航角ψi'；根据先验标准板(7)编码元(8)间已知空间约束解算参考元
    Figure PCTCN2017109782-appb-100054
    在第i帧中相机坐标系OCXCYCZC(13)下的三维坐标
    Figure PCTCN2017109782-appb-100055
    测量的轮廓误差是表征在机床坐标系OMXMYMZM15下的，需将相机(1)测量的轮廓进行坐标转换以正确求解机床轮廓误差；机床坐标系OMXMYMZM(15)与先验标准板局部坐标系OLiXLiYLiZLi(22)的位姿转换公式为：
    Figure PCTCN2017109782-appb-100056
    其中，Mi为第i帧图像下先验标准板局部坐标系OLiXLiYLiZLi(22)与机床坐标系OMXMYMZM(15)间的转换矩阵；从Mi中分离出第i帧标准板相对于机床坐标系OMXMYMZM(15)的俯仰角θi、滚转角Φi、偏航角ψi；遍历所有图像序列得到由参考元表征的的整个机床六自由度轮廓Lr，Lr(MXi，MYi，MZi，θi，Φi，ψi)，i＝1,2…G；通过比较实际六自由度轮廓Lr与名义轮廓Lm求解数控机床(4)插补产生的六自由度轮廓误差E：
    E＝Lr-Lm。    (9)
PCT/CN2017/109782 2017-11-07 2017-11-07 数控机床高动态大范围任意轮廓误差单目六维测量方法 WO2019090487A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US16/338,971 US11014211B2 (en) 2017-11-07 2017-11-07 Monocular vision six-dimensional measurement method for high-dynamic large-range arbitrary contouring error of CNC machine tool
PCT/CN2017/109782 WO2019090487A1 (zh) 2017-11-07 2017-11-07 数控机床高动态大范围任意轮廓误差单目六维测量方法
EP17925531.0A EP3511122B1 (en) 2017-11-07 2017-11-07 Monocular vision six-dimensional measurement method for high-dynamic large-range arbitrary contouring error of cnc machine tool

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/109782 WO2019090487A1 (zh) 2017-11-07 2017-11-07 数控机床高动态大范围任意轮廓误差单目六维测量方法

Publications (1)

Publication Number Publication Date
WO2019090487A1 true WO2019090487A1 (zh) 2019-05-16

Family

ID=66438693

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/109782 WO2019090487A1 (zh) 2017-11-07 2017-11-07 数控机床高动态大范围任意轮廓误差单目六维测量方法

Country Status (3)

Country Link
US (1) US11014211B2 (zh)
EP (1) EP3511122B1 (zh)
WO (1) WO2019090487A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111127565A (zh) * 2019-12-24 2020-05-08 易视智瞳科技（深圳）有限公司 标定方法、标定系统和计算机可读存储介质
CN112489009A (zh) * 2020-11-27 2021-03-12 芜湖哈特机器人产业技术研究院有限公司 基于视觉图像的泵体口环位姿检测方法
CN113034571A (zh) * 2021-04-16 2021-06-25 广东工业大学 一种基于视觉—惯性的物体三维尺寸测量方法
CN113592955A (zh) * 2021-07-27 2021-11-02 中国科学院西安光学精密机械研究所 一种基于机器视觉的圆形工件平面坐标高精度定位方法
CN114111578A (zh) * 2021-11-29 2022-03-01 哈尔滨工业大学 一种大口径元件的位姿自动确定方法
CN115937203A (zh) * 2023-01-09 2023-04-07 浙江智慧视频安防创新中心有限公司 一种基于模板匹配的视觉检测方法、装置、设备及介质
CN116595810A (zh) * 2023-07-18 2023-08-15 北矿机电科技有限责任公司 摇床接矿执行器偏移距离预测方法、系统、设备及介质

Families Citing this family (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6684684B2 (ja) * 2016-09-21 2020-04-22 株式会社Ｓｃｒｅｅｎホールディングス 試料容器の位置ずれ検出方法およびそれを用いた撮像方法、ならびに試料容器の位置ずれ検出装置
CN109360243B (zh) * 2018-09-28 2022-08-19 安徽爱观视觉科技有限公司 一种多自由度可动视觉系统的标定方法
US11158070B2 (en) * 2019-10-23 2021-10-26 1974266 AB Ltd (TAGDit) Data processing systems and methods using six-dimensional data transformations
CN112381881B (zh) * 2020-10-26 2023-05-23 西安航天精密机电研究所 一种基于单目视觉的大型刚体构件自动对接方法
CN112435302B (zh) * 2020-12-09 2024-05-31 北京理工大学 一种基于高精度转台和平行光管的远距离大视场鱼眼相机标定方法
CN113191997B (zh) * 2021-01-06 2022-02-01 天津大学 一种基于机器视觉的卡簧测量方法
CN112928324B (zh) * 2021-01-29 2022-03-25 蜂巢能源科技有限公司 模组部件装配方法及模组部件装配装置
CN113052913A (zh) * 2021-02-08 2021-06-29 南京航空航天大学 一种二级组合视觉测量系统中转位姿高精度标定方法
DE102021103493B3 (de) * 2021-02-15 2022-06-02 Novanta Europe Gmbh Verfahren für eine Scanfeldkorrektur mindestens einer Laserscannervorrichtung, Laserscannervorrichtung, Streumusterelement, Streumusterhaltevorrichtung und Scanfeldkorrektursystem
CN112882644B (zh) * 2021-02-24 2022-07-29 中国电建集团河南工程有限公司 一种输电线路百米弧垂值自动提取方法
CN113070644A (zh) * 2021-04-09 2021-07-06 三代光学科技（天津）有限公司 一种微结构阵列复合加工方法
CN113112545B (zh) * 2021-04-15 2023-03-21 西安电子科技大学 基于计算机视觉的手持移动打印装置定位方法
CN113029009B (zh) * 2021-04-30 2022-08-02 高速铁路建造技术国家工程实验室 一种双视角视觉位移测量系统及方法
CN113369990B (zh) * 2021-07-06 2022-05-10 成都飞机工业（集团）有限责任公司 一种用于非接触式测量孔的在线检测装置及其使用方法
CN113701798B (zh) * 2021-08-29 2024-01-19 孙恺欣 一种二维栅格标准板的校准装置和方法
CN113820070B (zh) * 2021-09-23 2024-04-30 浙江理工大学 基于机器视觉和扭摆法的刚体转动惯量测量方法及系统
CN114102256B (zh) * 2021-11-04 2022-09-06 清华大学 机床旋转轴几何误差识别方法、装置及存储介质
CN115265910B (zh) * 2022-07-13 2023-07-28 哈尔滨工业大学 基于偏心和倾斜误差传递的回转设备转动惯量测量方法
CN115619781B (zh) * 2022-12-15 2023-09-19 杭州三坛医疗科技有限公司 一种精度检测方法、装置、电子设备及存储介质
CN115673876B (zh) * 2022-12-28 2023-04-18 苏州猎奇智能设备有限公司 基于视觉的运动系统热变形二维测量方法及使用方法
CN116339238B (zh) * 2023-03-31 2024-04-05 中国科学院西安光学精密机械研究所 一种独立光束扫描的五轴激光加工设备光束运动控制方法
CN116400647B (zh) * 2023-06-08 2023-09-29 成都飞机工业（集团）有限责任公司 多轴插补运动控制方法、装置、存储介质及电子设备
CN116673792B (zh) * 2023-08-04 2023-11-10 成都飞机工业（集团）有限责任公司 一种加工中心旋转轴误差源剥离特征件及加工评价方法
CN117284500B (zh) * 2023-11-24 2024-02-09 北京航空航天大学 一种基于单目视觉和激光的盘绕式伸展臂位姿调整方法
CN117908464B (zh) * 2024-03-20 2024-06-14 山东大学 一种机床轮廓误差预测方法、系统、装置及可读存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5956253A (en) * 1997-09-09 1999-09-21 Glassline Corporation Camera controlled CNC apparatus for processing blanks
CN101329164A (zh) * 2007-03-30 2008-12-24 三丰株式会社 用于立体视觉探针的全局标定方法
CN105043259A (zh) * 2015-08-25 2015-11-11 大连理工大学 基于双目视觉的数控机床旋转轴误差检测方法
CN105252341A (zh) * 2015-09-02 2016-01-20 大连理工大学 五轴数控机床动态误差视觉测量方法
JP2016040531A (ja) * 2014-08-12 2016-03-24 コニカミノルタ株式会社 加工工具の測定方法及び測定装置
CN105798704A (zh) * 2016-04-25 2016-07-27 大连理工大学 一种机床平面轮廓误差单目测量方法
CN107971831A (zh) * 2017-11-07 2018-05-01 大连理工大学 数控机床高动态大范围任意轮廓误差单目六维测量方法

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5956523A (en) 1996-08-09 1999-09-21 Advantech Co., Ltd. Method and apparatus for reducing the number of RS232/RS485 transmission converters required for communicating between a PC and a plurality of instruments
US6437823B1 (en) * 1999-04-30 2002-08-20 Microsoft Corporation Method and system for calibrating digital cameras
US20060017720A1 (en) * 2004-07-15 2006-01-26 Li You F System and method for 3D measurement and surface reconstruction
US8103376B2 (en) * 2008-09-05 2012-01-24 Chung Yuan Christian University System and method for the on-machine 2-D contour measurement
CN101995850B (zh) * 2010-11-05 2012-02-29 江俊逢 一种计算机辅助数字控制方法与系统
WO2015111200A1 (ja) * 2014-01-24 2015-07-30 三菱電機株式会社 工具形状測定装置および工具形状測定方法
CN107533325B (zh) * 2016-03-16 2020-01-21 三菱电机株式会社 机械运动轨迹测定装置
CN110226137A (zh) * 2016-11-25 2019-09-10 格罗弗治公司 借助图像跟踪进行制造
US10558197B2 (en) * 2017-02-28 2020-02-11 Sap Se Manufacturing process data collection and analytics
US10234848B2 (en) * 2017-05-24 2019-03-19 Relativity Space, Inc. Real-time adaptive control of additive manufacturing processes using machine learning
US10192301B1 (en) * 2017-08-16 2019-01-29 Siemens Energy, Inc. Method and system for detecting line defects on surface of object
CN107589720B (zh) * 2017-09-11 2019-10-11 大连理工大学 一种等效平面交叉耦合控制方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5956253A (en) * 1997-09-09 1999-09-21 Glassline Corporation Camera controlled CNC apparatus for processing blanks
CN101329164A (zh) * 2007-03-30 2008-12-24 三丰株式会社 用于立体视觉探针的全局标定方法
JP2016040531A (ja) * 2014-08-12 2016-03-24 コニカミノルタ株式会社 加工工具の測定方法及び測定装置
CN105043259A (zh) * 2015-08-25 2015-11-11 大连理工大学 基于双目视觉的数控机床旋转轴误差检测方法
CN105252341A (zh) * 2015-09-02 2016-01-20 大连理工大学 五轴数控机床动态误差视觉测量方法
CN105798704A (zh) * 2016-04-25 2016-07-27 大连理工大学 一种机床平面轮廓误差单目测量方法
CN107971831A (zh) * 2017-11-07 2018-05-01 大连理工大学 数控机床高动态大范围任意轮廓误差单目六维测量方法

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111127565B (zh) * 2019-12-24 2023-09-01 易视智瞳科技（深圳）有限公司 标定方法、标定系统和计算机可读存储介质
CN111127565A (zh) * 2019-12-24 2020-05-08 易视智瞳科技（深圳）有限公司 标定方法、标定系统和计算机可读存储介质
CN112489009B (zh) * 2020-11-27 2022-07-26 芜湖哈特机器人产业技术研究院有限公司 基于视觉图像的泵体口环位姿检测方法
CN112489009A (zh) * 2020-11-27 2021-03-12 芜湖哈特机器人产业技术研究院有限公司 基于视觉图像的泵体口环位姿检测方法
CN113034571B (zh) * 2021-04-16 2023-01-20 广东工业大学 一种基于视觉—惯性的物体三维尺寸测量方法
CN113034571A (zh) * 2021-04-16 2021-06-25 广东工业大学 一种基于视觉—惯性的物体三维尺寸测量方法
CN113592955A (zh) * 2021-07-27 2021-11-02 中国科学院西安光学精密机械研究所 一种基于机器视觉的圆形工件平面坐标高精度定位方法
CN113592955B (zh) * 2021-07-27 2024-04-09 中国科学院西安光学精密机械研究所 一种基于机器视觉的圆形工件平面坐标高精度定位方法
CN114111578A (zh) * 2021-11-29 2022-03-01 哈尔滨工业大学 一种大口径元件的位姿自动确定方法
CN114111578B (zh) * 2021-11-29 2024-05-24 哈尔滨工业大学 一种大口径元件的位姿自动确定方法
CN115937203A (zh) * 2023-01-09 2023-04-07 浙江智慧视频安防创新中心有限公司 一种基于模板匹配的视觉检测方法、装置、设备及介质
CN116595810A (zh) * 2023-07-18 2023-08-15 北矿机电科技有限责任公司 摇床接矿执行器偏移距离预测方法、系统、设备及介质
CN116595810B (zh) * 2023-07-18 2023-09-26 北矿机电科技有限责任公司 摇床接矿执行器偏移距离预测方法、系统、设备及介质

Also Published As

Publication number Publication date
EP3511122B1 (en) 2020-04-29
EP3511122A1 (en) 2019-07-17
EP3511122A4 (en) 2019-10-23
US11014211B2 (en) 2021-05-25
US20200061769A1 (en) 2020-02-27

Similar Documents

Publication Publication Date Title
WO2019090487A1 (zh) 数控机床高动态大范围任意轮廓误差单目六维测量方法
CN107971831B (zh) 数控机床高动态大范围任意轮廓误差单目六维测量方法
CN106981083B (zh) 双目立体视觉系统摄像机参数的分步标定方法
CN108198224B (zh) 一种用于立体视觉测量的线阵相机标定装置及标定方法
CN103292695B (zh) 一种单目立体视觉测量方法
CN108844459A (zh) 一种叶片数字化样板检测系统的标定方法及装置
CN113674345B (zh) 一种二维像素级三维定位系统及定位方法
Boochs et al. Increasing the accuracy of untaught robot positions by means of a multi-camera system
CN109712139B (zh) 基于线性运动模组的单目视觉的尺寸测量方法
CN106500625B (zh) 一种远心立体视觉测量方法
CN110125455A (zh) 一种用于机器人钻孔中优化钻头位姿的方法
CN110695982A (zh) 一种基于三维视觉的机械臂手眼标定方法和装置
CN114283203A (zh) 一种多相机系统的标定方法及系统
CN107478203A (zh) 一种基于激光扫描的3d成像装置及成像方法
CN115861445B (zh) 一种基于标定板三维点云的手眼标定方法
CN111207670A (zh) 一种线结构光标定装置及方法
CN108871190B (zh) 一种用于双目立体视觉测量中的手持球型靶标及测量方法
CN116182703B (zh) 一种线结构光传感器标定方法及系统
Li et al. Monocular-vision-based contouring error detection and compensation for CNC machine tools
CN111311659B (zh) 一种基于斜交平面镜三维成像的校准方法
CN102538707B (zh) 一种对工件进行三维定位的装置及方法
CN105678847B (zh) 线激光用于slm显微立体视觉的小尺度物体表面重构方法
CN208350997U (zh) 一种目标物运动监测系统
Xu et al. Three degrees of freedom global calibration method for measurement systems with binocular vision
JPH05248819A (ja) カメラによる測定対象物の三次元位置測定のための較正対象データの較正方法及び三次元位置測定方法

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2017925531

Country of ref document: EP

Effective date: 20190326

NENP Non-entry into the national phase

Ref country code: DE