WO2019047683A1 - 一种智能打磨方法 - Google Patents

一种智能打磨方法 Download PDF

Info

Publication number
WO2019047683A1
WO2019047683A1 PCT/CN2018/100476 CN2018100476W WO2019047683A1 WO 2019047683 A1 WO2019047683 A1 WO 2019047683A1 CN 2018100476 W CN2018100476 W CN 2018100476W WO 2019047683 A1 WO2019047683 A1 WO 2019047683A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
grinding
target
floating
sanding
Prior art date
Application number
PCT/CN2018/100476
Other languages
English (en)
French (fr)
Inventor
黄约
刘运毅
唐明福
黎相成
Original Assignee
南宁宇立仪器有限公司
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 南宁宇立仪器有限公司 filed Critical 南宁宇立仪器有限公司
Publication of WO2019047683A1 publication Critical patent/WO2019047683A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B1/00—Processes of grinding or polishing; Use of auxiliary equipment in connection with such processes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B41/00—Component parts such as frames, beds, carriages, headstocks
    • B24B41/007—Weight compensation; Temperature compensation; Vibration damping
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
    • B24B49/16—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation taking regard of the load
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B9/00—Machines or devices designed for grinding edges or bevels on work or for removing burrs; Accessories therefor
    • B24B9/02—Machines or devices designed for grinding edges or bevels on work or for removing burrs; Accessories therefor characterised by a special design with respect to properties of materials specific to articles to be ground
    • B24B9/04—Machines or devices designed for grinding edges or bevels on work or for removing burrs; Accessories therefor characterised by a special design with respect to properties of materials specific to articles to be ground of metal, e.g. skate blades

Definitions

  • the invention relates to the field of intelligent robots, in particular to an intelligent grinding method.
  • 201310746856.9 discloses a polishing and grinding force-position mixing control method and system, the method comprising: a pressure sensor obtaining a grinding tool and a machining part The current grinding force Fc; the comparator calculates a difference between the desired grinding force Fd between the sanding tool and the workpiece and the current grinding force Fc; the fuzzy controller according to the rate of change Ec of the difference, The fuzzy control algorithm is used to output the control variable U to control the grinding force between the grinding tool and the workpiece.
  • This method improves the control precision when the polishing robot generates contact force with the external environment, and ensures the constant grinding force during the grinding process.
  • the main disadvantages are: (1) poor real-time performance; controlling the movement of the robot when the change of the force is detected, and the constant force effect is achieved by the change of the motion state of the robot, the robot The reaction speed is difficult to keep up with the speed of force change, so the constant force grinding effect is poor, the grinding efficiency is low; (2) the system is complex; the developer needs to acquire the force sensor signal, analyze and process, obtain the control strategy, and then control the robot movement; (3) There is no accurate position sensing; constant force grinding is only the control of force. Without position control, it is difficult to accurately obtain the amount to be polished and the remaining amount after grinding.
  • the prior art For the measurement of workpiece position and workpiece size, the prior art generally uses optical techniques such as laser measurement or industrial cameras.
  • the main disadvantages of this method are: (1) a large amount of dust is generated at the grinding site, and in some cases, water or coolant is sprayed. These will affect the measurement. (2) Vibration occurs during grinding, which places high demands on laser measurement and anti-vibration performance of industrial cameras. (3) When the actual grinding is performed, the workpiece may be deformed due to the grinding force, and the robot body may also be slightly deformed, and the abrasive grinding tool may also be deformed. Actually, when the workpiece to be ground is measured by laser measurement and industrial camera, there is no grinding force, and these deformations have not yet been generated, so the amount of deformation is not measured, which affects the final grinding effect.
  • the invention discloses an intelligent grinding method.
  • the floating grinding structure comprises a displacement sensor, so that the entire grinding system (including the robot, the floating grinding structure, the grinding tool, the displacement sensor, the workpiece to be ground and other fixtures) are all part of the position sensing system.
  • the deformation of any part of the floating sanding structure is ultimately reflected on the output of the displacement sensor. Therefore, by combining the motion trajectory curve with the displacement sensor displacement data, the actual size curve of the grinding target can be fitted, thereby formulating a targeted polishing strategy to improve the grinding efficiency and the grinding precision.
  • the robotic polishing apparatus comprises a robot, a floating sanding structure and a sanding tool, wherein the floating sanding structure comprises a fixing portion, a floating portion movable relative to the fixed portion, and a displacement sensor for measuring a relative displacement of the fixing portion and the floating portion; Need to install robots, floating sanding structures, sanding targets and sanding tools;
  • the robot moves according to a predetermined trajectory, so that the sanding tool or the floating sanding structure contacts the sanding target surface and moves relative to the sanding target, scans and/or polishes, and obtains the robot motion track coordinate data 1 and the displacement sensor data 2;
  • the translation relationship transformation matrix and the rotation relationship transformation matrix operation superimpose the data 1 and the data 2 in a specific coordinate system to obtain the actual feature data 3 of the polishing target;
  • the specific coordinate system is a geodetic coordinate system or a base coordinate system or a tool Coordinate system or user coordinate system;
  • the method for acquiring data 4 includes one or more of the following methods: (1) grinding the theoretical data of the target itself; and (2) analyzing the data by a linear fitting method. 3 obtaining the theoretical reference data 4; (3) obtaining the actual feature data 5 of the peripheral position or the like of the grinding target according to the method of step S2, and obtaining the theoretical reference data 4 by the linear fitting method;
  • the grinding target is polished by the same robot, floating sanding structure, sanding target and sanding tool installation method as in step S1.
  • steps of S2 and S3 may be replaced by:
  • the method for acquiring the data 4 includes one or two of the following methods: (1) grinding the theoretical data of the target itself; and (2) moving the robot according to a predetermined trajectory, so that The sanding tool or the floating sanding structure contacts the surface of the polishing target peripheral position or the like and moves relative to the peripheral position or the like of the polishing target to perform scanning and/or grinding to obtain the robot motion track coordinate data 1a and the displacement sensor data 2a;
  • the translation relationship transformation matrix and the rotation relationship transformation matrix operation of the spatial point superimpose the data 1a and the data 2a in a specific coordinate system to obtain the actual feature data 5 of the peripheral position or the like of the polished target, and then obtain the linear characteristic method to obtain the actual feature data 5
  • Theoretical reference data 4 ;
  • the robot moves according to a predetermined trajectory, so that the sanding tool or the floating sanding structure contacts the sanding target surface and moves relative to the sanding target, scans and/or polishes, and obtains the robot motion track coordinate data 1 and the displacement sensor data 2;
  • the translation relationship transformation matrix and the rotation relationship transformation matrix operation, the data 1 and the data 2 are superimposed in a specific coordinate system to obtain the actual feature data 3 of the polishing target;
  • the specific coordinate system is a geodetic coordinate system or a base coordinate system or a tool coordinate system or a user coordinate system.
  • the floating portion can float axially or/and radially with respect to the fixed portion.
  • the polishing strategy includes: (1) adjusting the movement track of the robot; (2) adjusting the grinding time, the grinding pressure, and the number of grinding; (3) replacing the grinding tool and/or the abrasive; and (4) adjusting the rotation speed of the polishing tool.
  • the method further includes the step S6: S6. acquiring the robot motion trajectory coordinate data and the displacement sensor data while grinding the robot polishing device, and obtaining the actual feature data 3i of the polished target after the grinding with the same scanning method as step S2 or S3, and The theoretical reference data 4 of the polished target is compared to adjust the polishing strategy or to determine whether the polishing is finished.
  • the robot polishing device collects the robot coordinate data and the displacement sensor data while the grinding work is performed. .
  • the floating sanding structure is constantly floating.
  • the floating sanding structure uses the same constant force when the sanding tool or the floating sanding structure contacts the sanding target and the polishing target peripheral position or the like position surface for scanning and/or grinding.
  • the robot, the floating sanding structure, the grinding target and the grinding tool are installed in the following manner: (1) the front end of the robot is connected with the floating sanding structure, the front end of the floating sanding structure is connected with the grinding tool, and the grinding target is fixed at other places; (2) The front end of the robot is connected to the floating sanding structure, the front end of the floating sanding structure is connected to the grinding target, and the grinding tool is fixed in other places; (3) the front end of the robot is connected with the grinding tool, and the grinding target and the floating sanding structure are connected and fixed in other places; (4) The robot is connected to the sanding target, and the sanding tool is connected to the floating sanding structure and fixed elsewhere.
  • the intelligent grinding method described above can determine the shape and size of the portion to be polished of the polished target surface and the position of the portion to be polished of the target surface to be polished according to the combination of the motion trajectory coordinate data and the displacement sensor data, where the portion to be polished is described herein, Including welding seams, flashing edges, burrs or other objects to be polished, etc., fitting the actual size curve of the grinding target, thereby controlling the robot polishing device to quickly reach the position to be polished according to the actual size curve, and according to the shape of the portion to be polished And the size, the targeted setting of the grinding time and the grinding pressure, the invention has the following advantages:
  • the present invention adopts a floating sanding structure, and the floating portion can float relative to the fixed portion.
  • the floating function can automatically adapt to the dimensional difference of the workpiece itself and the positioning accuracy error of the workpiece.
  • the position and size data are measured by a grinding tool or a floating sanding structure in contact with the object to be measured, so dust and vibration have little effect on the measurement.
  • the method can comprehensively collect the displacement sensor data and the coordinate data of the robot motion track, and the two have a good correspondence relationship, and real-time synchronization can be realized.
  • the system is relatively simple: the actual characteristics of the grinding target and the theoretical characteristic data of the grinding target are obtained by the above method, and the amount of the grinding target can be obtained by subtracting the two, and then the targeted grinding strategy can be formulated without complicated mechanical data. Conversion analysis processing, high processing efficiency, good grinding effect, and relatively simple process.
  • the floating grinding structure includes a displacement sensor, so that the entire grinding system (including robot, floating grinding structure, grinding tool, displacement sensor, workpiece to be ground and other fixtures) is part of the position sensing system, floating
  • the deformation of any part of the sanding structure is finally reflected on the output of the displacement sensor; further, due to the constant force floating of the floating sanding structure, and the robot grinding device judges that the result of the grinding is achieved by the relative displacement between the floating portion and the fixed portion. Then, whether the grinding target is deformed or the grinding tool wears out, the displacement data of the displacement sensor can be automatically compensated, thereby greatly improving the grinding precision.
  • the method has high processing speed, high automation degree, good grinding effect and wide application prospect.
  • Figure 1-1 is a schematic diagram of data 1 in the first embodiment.
  • 1-2 is a schematic diagram of data 2 in the first embodiment.
  • 1-3 are schematic views of data 10 in the first embodiment.
  • 1-4 are schematic views of data 20 in the first embodiment.
  • 1-5 are schematic views of data 3 in the first embodiment.
  • 1-6 are schematic views of data 4 in the first embodiment.
  • 1-7 are schematic views of data 6 in the first embodiment.
  • 2-1 is a diagram showing data 12 and data 22 in the second embodiment.
  • 2-2 is a schematic diagram of data 13 and data 23 in the second embodiment.
  • 2-3 is a schematic diagram of data 15 and data 25 in the second embodiment.
  • 2-4 are schematic views of data 52 in the second embodiment.
  • 2-5 are schematic views of data 53 in the second embodiment.
  • 2-6 are schematic views of data 4 in the second embodiment.
  • 2-7 are schematic diagrams of data 3 in the second embodiment.
  • 2-8 are schematic diagrams of data 6 in the second embodiment.
  • 2-9 is a schematic view showing a cross section of a weld bead and a peripheral position in the second embodiment.
  • FIG. 3 is a schematic cross-sectional view of a floating sanding structure.
  • the mark shown in the figure is: 1: floating shaft; 2: constant pressure outer cylinder; 3: constant pressure inner cylinder; 4: floating piston; 5: first chamber; 6: second chamber; One air inlet; 8: second air inlet; 9: rolling body.
  • This embodiment is a specific application of the method of the present invention in single weld grinding.
  • the robotic polishing apparatus comprises a robot, a floating sanding structure and a sanding tool, wherein the floating sanding structure comprises a fixing portion, a floating portion movable relative to the fixed portion, and a displacement sensor for measuring a relative displacement of the fixed portion and the floating portion; Install robots, floating sanding structures, sanding targets and sanding tools.
  • the installation method is that the robot is connected to the floating sanding structure through a flange, and the front end of the floating sanding structure is connected to the angle grinder, and the angle grinder is equipped with a grinding wheel.
  • the robot moves according to a predetermined trajectory, so that the sanding tool or the floating sanding structure contacts the sanding target surface and moves relative to the sanding target, scans and/or polishes, and obtains the robot motion track coordinate data 1 and the displacement sensor data 2;
  • the translation relationship transformation matrix and the rotation relationship transformation matrix operation superimpose the data 1 and the data 2 in a specific coordinate system to obtain the actual feature data 3 of the polishing target;
  • the specific coordinate system is a geodetic coordinate system or a base coordinate system or a tool Coordinate system or user coordinate system.
  • the specific actual feature data acquisition process is:
  • S2.1 establishes the user coordinate system: the starting point of the movement of the grinding wheel on the grinding target is taken as the origin O of the coordinate point, and the user coordinate system Pxyz is established, wherein the X axis is perpendicular to the weld direction, and the Y axis is the straight line extending direction of the weld and the pointing movement The direction of the track, the Z axis is determined by the right hand rule.
  • S2.2 Scanning The robot moves along the set Y-axis direction, so that the grinding wheel is in contact with the grinding target, that is, it also contacts the weld seam and scans the weld seam, and scans the point where the robot moves along the X-axis direction.
  • step S2.2 Conversion Matrix Operation: The data 1 and data 2 obtained in step S2.2 are converted to point coordinate data under the established user coordinate system Pxyz to obtain data 10 and data 20.
  • the formula for shifting or rotating the arbitrary point A in space to obtain point B is:
  • the spatial point A(x, y, z) is rotated by the angle ⁇ around the Z axis to point B (x1, y1, z1):
  • the spatial point A(x, y, z) rotates the ⁇ angle around the Y axis to point B (x1, y1, z1):
  • data 1 and data 2 are converted to the Pxyz coordinate system, as shown in Figures 1-3 and 1-4, the coordinate axis unit is the robot length unit, and data 10 and data 20 are obtained.
  • the amount of motion in the X direction is small and can be simplified to the motion in the Y and Z directions.
  • S2.4 data 10 and data 20 are added to obtain actual feature data 3 of the Pxyz coordinate system weld.
  • Figure 1-5 when representing data 3, do the equal interval processing and remove the zero point (ie, the entire curve is translated downward).
  • the method for acquiring data 4 includes one or more of the following methods: (1) grinding the theoretical data of the target itself; and (2) analyzing the data by a linear fitting method. 3 obtaining the theoretical reference data 4; (3) obtaining the actual feature data 5 of the peripheral position or the like of the sharpening target according to the method of step S2, and obtaining the theoretical reference data 4 by the linear fitting method.
  • This embodiment adopts the method (2), and the theoretical reference data 4 is obtained by analyzing the data 3 by the linear fitting method.
  • the line type fitting method adopted in this embodiment is a least square method, and the baseline of the actual feature data 3 is extracted by the least square method, which is the theoretical reference data 4 to which the weld is to be polished.
  • z n is the Z coordinate data of the data 3
  • y n is the Y coordinate data of the data 3, so n equations can be obtained, and the unknowns to be sought are only a and b.
  • This embodiment uses the least squares method to solve for the optimal a and b.
  • AX B, X is the amount to be requested, and A and B are known quantities.
  • the B matrix is: Where z n is the Z coordinate data of data 3.
  • the X matrix is:
  • the baseline of the data 3 is obtained by using the above formula, that is, the theoretical reference data 4 is obtained.
  • step S5. Grinding is performed using the same robot, floating sanding structure, sanding target, and sanding tool mounting method as in step S1.
  • step S6 Obtaining the robot motion trajectory coordinate data 1i and the displacement sensor data 2i while grinding the robot polishing device, obtaining the actual feature data 3i of the polished target after the grinding in the same manner as step S2, and comparing with the theoretical reference data 4 of the grinding target Yes, to adjust the polishing strategy.
  • Step S6 is repeated until the weld actual feature data 3i to be polished and the theoretical reference data 4 are within the allowable range, and it is judged that the polishing is finished.
  • the robotic grinding device can collect the robot coordinate data and the displacement sensor data while grinding.
  • This embodiment is a specific application of the method of the present invention in single weld grinding.
  • the robotic polishing apparatus comprises a robot, a floating sanding structure and a sanding tool, wherein the floating sanding structure comprises a fixing portion, a floating portion movable relative to the fixed portion, and a displacement sensor for measuring a relative displacement of the fixed portion and the floating portion; Install robots, floating sanding structures, sanding targets and sanding tools.
  • the installation method is that the robot is connected to the floating sanding structure through a flange, and the front end of the floating sanding structure is connected to the angle grinder, and the angle grinder is equipped with a grinding wheel.
  • the robot moves according to a predetermined trajectory, so that the grinding tool or the floating sanding structure is in contact with and relatively moved with the peripheral position of the object to be polished, and is scanned and/or polished to obtain the robot motion trajectory coordinate data 1 and the displacement sensor data 2;
  • the translation relationship transformation matrix and the rotation relationship transformation matrix operation superimpose the data 1 and the data 2 in a specific coordinate system to obtain the feature data 5 around the polishing target, and obtain the theoretical reference data 4 by the linear fitting method.
  • the specific coordinate system is a geodetic coordinate system or a polar coordinate system or a tool coordinate system or a user coordinate system.
  • the specific actual feature data acquisition process is:
  • S2.1 establish the user coordinate system: the starting point of the movement of the grinding wheel on the grinding target is used as the original coordinate point O of the scanning weld, and the user coordinate system Pxyz is established, wherein the X axis is in the direction of the weld and points in the opposite direction of the movement track, Y The axis is perpendicular to the weld direction and the Z axis is determined by the right hand rule.
  • S2.2 Scan The robot moves along the negative direction of the set X axis, so that the grinding wheel and the left and right sides of the grinding target are in contact for scanning. As shown in Figure 2-1 and Figure 2-2, the robot is scanned along X.
  • step S2.3 conversion matrix operation the operation method is the same as the mathematical method described in Embodiment 1; the data 12 and data 22 obtained in step S2.2 are converted to the point coordinate data of the established user coordinate system Pxyz, as shown in FIG. 2 As shown in 3, data 15 and data 25 are obtained.
  • the specific method is to assume that the weld is on a flat surface, and the weld seams are smoothly transitioned on both sides of the weld.
  • the cross section of the weld and the surrounding location can be simplified as shown in Figure 2-9.
  • D represents the left side of the weld
  • E represents the right side of the weld
  • F represents the weld
  • the length of the DF is L df
  • the length of the EF is L ef
  • L df and L ef can be set in advance in the trajectory of the robot and is known the amount.
  • the robot moves according to a predetermined trajectory, so that the grinding wheel connected to the front part of the floating part contacts the grinding target surface and moves relative to the grinding target, and obtains the robot motion trajectory coordinate data 1 and the displacement sensor data 2; the translation matrix of the translation relationship through the spatial point and The rotation relationship transformation matrix operation superimposes data 1 and data 2 in a specific coordinate system to obtain actual feature data 3 of the polishing target;
  • the specific coordinate system is a geodetic coordinate system or a base coordinate system or a tool coordinate system or a user coordinate system ;
  • Data 3 is shown in Figure 2-7.
  • step S5. Grinding is performed using the same robot, floating sanding structure, sanding target, and sanding tool mounting method as in step S1.
  • Step S6 is repeated until the weld bead feature data 3i to be polished and the theoretical reference data 4 are within the allowable range, and the end of the sanding is judged.
  • the robotic grinding device can collect the robot coordinate data and the displacement sensor data while the grinding work is performed.
  • the floating sanding structure in order to realize the automatic compensation of the deformation of the sharpening tool, it is more preferable that the floating sanding structure is a constant force floating when scanning or grinding, that is, a constant pressure is applied to the floating portion, so that the floating portion can be relatively fixed.
  • This embodiment further provides a preferred floating sanding structure in which the floating portion can be axially floated relative to the fixed portion. As shown in FIG. 3, it includes a floating shaft 1, a constant pressure outer cylinder 2, a constant pressure inner cylinder 3, and a floating piston 4: the constant pressure inner cylinder 3 and the floating piston 4 are respectively sleeved outside the floating shaft 1.
  • the constant pressure outer cylinder 2 is sleeved outside the constant pressure inner cylinder 3 and the floating piston 4, and a chamber is formed between the constant pressure outer cylinder 2 and the constant pressure inner cylinder 3, and one end of the floating piston 4 and the floating shaft 1 fixedly connected, the other end is located in the chamber and divides the chamber into a first chamber 5 and a second chamber 6, and a first air inlet 7 and a second air inlet are disposed on the side wall of the constant pressure outer cylinder 2
  • the port 8 the first air inlet 7 is in communication with the first chamber 5, and the second air inlet 8 is in communication with the second chamber 6.
  • the floating sanding structure of the embodiment When the floating sanding structure of the embodiment is in an operating state, the first chamber 5 or the second chamber 6 is inflated through the first inlet port 7 or the second inlet port 8. When inflating into the first chamber 5 through the first air inlet 7, the floating piston 4 is moved toward the second chamber 6 by the air flow. Since the floating piston 4 is fixedly connected with the floating shaft 1, the floating piston 4 drives the floating The shaft 1 moves together and the floating shaft 1 is in an extended state. When the airflow input from the first chamber 5 is constant, the floating piston 4 is forced to be constant, and therefore the force transmitted to the floating shaft 1 through the floating piston 4 is constant, that is, the axial constant force floats.
  • the floating piston 4 When the second air inlet 8 is inflated into the second chamber 6, the floating piston 4 is moved toward the first chamber 5 by the air flow, and the floating piston 4 drives the floating shaft 1 to move together, and the floating shaft 1 is retracted. status.
  • the airflow input from the second chamber 6 is constant, the floating piston 4 is forced to be constant, and therefore the force transmitted to the floating shaft 1 through the floating piston 4 is also constant. It is to be noted that since the floating piston 4 is movable in the chamber formed by the constant pressure outer cylinder 2 and the constant pressure inner cylinder 3, the boundary between the first chamber 5 and the second chamber 6 is not clearly defined.
  • the floating piston 4 and the constant pressure outer cylinder 2 are not necessarily completely dense, so that when the first intake port 7 is inflated into the first chamber 5, the gas can also pass through the floating piston. 4 and the gap between the constant pressure outer cylinder 2 and the second air inlet 8; when the second air inlet 8 is inflated into the second chamber 6, the gas can also pass through the floating piston 4 and the constant pressure outer cylinder The gap between the bodies 2 and the first air inlet 7 are discharged.
  • a linear guide mechanism is provided between the floating shaft 1 and the constant pressure inner cylinder block 3, and the linear guide mechanism allows only axial linear motion between the floating shaft 1 and the constant pressure inner cylinder block 3.
  • the linear guide mechanism includes: a first sliding slot disposed on an outer surface of the floating shaft 1, a second sliding slot disposed on an inner surface of the constant pressure inner cylinder 3 and opposite to the first sliding slot, and a first sliding slot a plurality of rolling bodies 9 in the groove and the second chute.
  • the floating portion of the embodiment is the floating shaft 1 and/or the floating piston 4, and the fixed portion is the constant pressure outer cylinder 2 and/or the constant pressure inner cylinder 3, in order to detect the relative displacement of the floating portion and the fixed portion, A displacement sensor capable of measuring the relative displacement between them is mounted on the floating portion and the fixed portion.
  • the constant force referred to in this embodiment is an approximate constant force in the engineering sense.
  • the mounting manner of the robot, the floating sanding structure, the grinding target and the grinding tool can also be replaced with other forms, for example: (1) the front end of the robot is connected to the floating sanding structure, and the front end of the floating sanding structure is connected to the grinding target, and the grinding is performed. The tool is fixed in other places; (2) the front end of the robot is connected to the grinding tool, the grinding target is connected with the floating grinding structure and fixed in other places; (3) the robot is connected to the grinding target, and the grinding tool and the floating sanding structure are connected and fixed elsewhere.
  • the present invention adopts a floating sanding structure, and the floating portion can float relative to the fixed portion.
  • the floating function can automatically adapt to the dimensional difference of the workpiece itself and the positioning accuracy error of the workpiece.
  • the position and size data are measured by a grinding tool or a floating sanding structure in contact with the object to be measured, so dust and vibration have little effect on the measurement.
  • the method can comprehensively collect the displacement sensor data and the coordinate data of the robot motion track, and the two have a good correspondence relationship, and real-time synchronization can be realized.
  • the system is relatively simple: the actual characteristics of the grinding target and the theoretical characteristic data of the grinding target are obtained by the above method, and the amount of the grinding target can be obtained by subtracting the two, and then the targeted grinding strategy can be formulated without complicated mechanical data. Conversion analysis processing, high processing efficiency, good grinding effect, and relatively simple process.
  • the floating grinding structure includes a displacement sensor, so that the entire grinding system (including robot, floating grinding structure, grinding tool, displacement sensor, workpiece to be ground and other fixtures) is part of the position sensing system, floating
  • the deformation of any part of the sanding structure is finally reflected on the output of the displacement sensor; further, due to the constant force floating of the floating sanding structure, and the robot grinding device judges that the result of the grinding is achieved by the relative displacement between the floating portion and the fixed portion. Then, whether the grinding target is deformed or the grinding tool wears out, the displacement data of the displacement sensor can be automatically compensated, thereby greatly improving the grinding precision.

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Finish Polishing, Edge Sharpening, And Grinding By Specific Grinding Devices (AREA)
  • Manipulator (AREA)

Abstract

一种智能打磨方法，包括以下步骤：机器人按照预定轨迹运动，使浮动打磨结构的浮动部或浮动部前部连接的打磨工具接触到打磨目标表面相对于打磨目标运动，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；获取特定坐标系下打磨目标的理论参数数据4；数据3减去数据4，获得打磨目标的待打磨量数据6，并根据数据6的特征，制度打磨策略。

Description

一种智能打磨方法 技术领域
本发明涉及智能机器人领域，特别是一种智能打磨方法。
背景技术
目前我国在金属制品生产中普遍采用熟练工人通过电、气动研磨工具手工完成。手工打磨劳动强度大、费时、效率低，且打磨时产生的易燃易爆粉尘，对操作工人来说无疑是存在潜在危险。现在也有厂家采用机器人进行打磨，但传统的机器人打磨方法都采用力传感器作为测量元件，实时监测打磨压力的大小并反馈至机器人系统，再由机器人系统控制机器人改变轨迹，以达到恒力打磨的效果。如申请号为201310746856.9的专利申请“一种抛光打磨力-位混合控制方法和系统”，公开了一种抛光打磨力-位混合控制方法和系统，方法包括：压力传感器获得打磨工具与加工件之间的当前打磨力Fc；比较器计算所述打磨工具与加工件之间的期望打磨力Fd与所述当前打磨力Fc之间的差值；模糊控制器根据所述差值的变化率Ec，采用模糊控制算法输出控制变量U控制打磨工具与加工件之间的打磨力，该方法虽然在打磨抛光机器人与外部环境产生接触力时提高控制精度，保证打磨过程中打磨力的恒定。但类似于该申请通过力传感器来控制打磨精度的方法，主要存在的缺点有：(1)实时性差；检测到力的变化时再控制机器人运动，通过机器人运动状态的变化达到恒力效果，机器人的反应速度难以跟上力变化的速度，因此恒力打磨效果差，打磨效率低；(2)系统复杂；开发者需要获取力传感器的信号，分析处理，得出控制策略，再控制机器人运动；(3)无准确的位置感知；恒力打磨只是力的控制，没有位置的控制，难以准确获取待打磨的量和打磨后的剩余量。对于工件位置和工件尺寸的测量，现有技术一般采用激光测量或工业相机等光学技术。该方法主要存在的缺点有：(1)打磨现场有大量粉尘产生，有的场合还会喷水或冷却液。这些都会影响测量。(2)打磨时会产生振动，对激光测量和工业相机的抗振性能提出来很高的要求。(3)实际打磨时，由于打磨力的作用，工件有可能会变形，机器人本体也可能有轻微形变，磨料磨具也可能会变形。而实际在用激光测量和工业相机测量待磨工件时，并没有打磨力的存在，这些变形还没有产生，所以测量不到这些变形量，影响了最终的打磨效果。
发明内容
本发明公开了一种智能打磨方法，浮动打磨结构包含有位移传感器，使得整个打磨系统 (包括机器人、浮动打磨结构、打磨工具、位移传感器、待磨工件及其它夹具)都是位置感知系统的一部分，浮动打磨结构中任意部分的变形都最终反应在位移传感器的输出上。因此，能通过运动轨迹曲线与位移传感器位移数据结合，拟合出打磨目标的实际尺寸曲线，从而制定有针对性的打磨策略，提高打磨效率和打磨精度。
为实现上述目的，本发明的技术方案为：
一种智能打磨方法，其特征在于包括以下步骤：
S1.机器人打磨设备包括机器人、浮动打磨结构和打磨工具，其中所述浮动打磨结构包括固定部、能相对固定部移动的浮动部以及测量固定部和浮动部相对位移的位移传感器；按照打磨工作的需求安装机器人、浮动打磨结构、打磨目标和打磨工具；
S2.机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标表面并相对于打磨目标运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；所述特定坐标系为大地坐标系或基坐标系或工具坐标系或用户坐标系；
S3.获取特定坐标系下打磨目标的理论参考数据4，数据4的获取方法包括以下方式的一种或多种：(1)打磨目标自身的理论数据；(2)通过线性拟合方法分析数据3获得理论参考数据4；(3)按步骤S2的方法获取打磨目标周边位置或类似位置的实际特征数据5，通过线性拟合方法计算获得理论参考数据4；
S4.数据3减去数据4，获得打磨目标的待打磨量数据6，并根据数据6的特征，制定打磨策略；
S5.采用与步骤S1相同的机器人、浮动打磨结构、打磨目标和打磨工具的安装方式对打磨目标进行打磨。
进一步的，所述S2和S3的步骤可以替换为：
所述S2和S3的步骤替换为：
S2.获取特定坐标系下打磨目标的理论参考数据4，数据4的获取方法包括以下方式的一种或两种：(1)打磨目标自身的理论数据；(2)机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标周边位置或类似位置的表面并相对于打磨目标周边位置或类似位置运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1a和位移传感器数据2a；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1a和数据2a在特定坐标系下进行叠加，获得打磨目标周边位置或类似位置的实际特征数据5，再通过线性拟合方法计算获得 理论参考数据4；
S3.机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标表面并相对于打磨目标运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；
所述特定坐标系为大地坐标系或基坐标系或工具坐标系或用户坐标系。
进一步的，所述步骤S1中，所述浮动部能相对固定部轴向或/和径向浮动。
进一步的，所述打磨策略包括：(1)调整机器人运动轨迹；(2)调整打磨时间、打磨压力和打磨次数；(3)更换打磨工具和/或磨料；(4)调整打磨工具转速。
进一步的，还包括步骤S6：S6.机器人打磨设备打磨的同时获取机器人运动轨迹坐标数据和位移传感器数据，以与步骤S2或S3相同的扫描方法获得打磨后打磨目标的实际特征数据3i，并与打磨目标的理论参考数据4做比对，以调整打磨策略或判断打磨是否结束。
进一步的，所述步骤S5中，机器人打磨设备在打磨工作的同时采集机器人坐标数据和位移传感器数据。。
优选的，所述浮动打磨结构恒力浮动。
优选的，所述打磨工具或浮动打磨结构接触到打磨目标及打磨目标周边位置或类似位置表面进行扫描和/或打磨时，浮动打磨结构采用相同的恒力。
进一步的，所述步骤S1中，机器人、浮动打磨结构、打磨目标和打磨工具的安装方式为：(1)机器人前端连接浮动打磨结构，浮动打磨结构前端连接打磨工具，打磨目标固定在其他地方；(2)机器人前端连接浮动打磨结构，浮动打磨结构前端连接打磨目标，打磨工具固定在其他地方；(3)机器人前端连接打磨工具，打磨目标和浮动打磨结构连接并固定在其他地方；(4)机器人连接打磨目标，打磨工具和浮动打磨结构连接并固定在其他地方。
以上所述的智能打磨方法，可根据运动轨迹坐标数据与位移传感器数据的结合确定打磨目标表面待打磨部位的形状和尺寸以及打磨目标表面待打磨的部位的位置，这里所述的待打磨部位，包括焊缝、飞边、毛刺或其他待打磨物体等，拟合出打磨目标的实际尺寸曲线，从而根据实际尺寸曲线，控制机器人打磨设备快速到达待打磨的位置，并根据待打磨的部位的形状和尺寸，针对性的设置打磨时间和打磨压力，本发明具有以下优点：
(1)适应性好：本发明采用了浮动打磨结构，浮动部能相对固定部浮动。在打磨和位置测量时，浮动功能能够自动的适应工件本身的尺寸差异和工件的定位精度误差。且采用打磨工具或浮动打磨结构和被测物体接触的方法来测量位置和尺寸数据，因此粉尘和振动对测量 的影响小。
(2)准确的位置感知：浮动打磨结构内部有位移传感器，能测量浮动部相对固定部的位置变化，为智能打磨系统提供了打磨目标的尺寸信息，结合位移传感器有效的获取的位置信息，即可以准确获取打磨目标具体部位的待打磨量和打磨后的剩余量。
(3)实时性好：在打磨和位置测量步骤中，本方法能综合采集位移传感器数据和机器人运动轨迹坐标数据，两者有较好的对应关系，可实现实时的同步。
(4)系统相对简单：通过上述方法获取打磨目标的实际特征和打磨目标的理论特征数据，两者相减就能得到待打磨量，进而可制定针对性的打磨策略，无须经过复杂的力学数据转换分析处理，处理效率高，打磨效果好，过程相对简单。
(5)具有自动补偿功能：浮动打磨结构包含有位移传感器，使得整个打磨系统(包括机器人、浮动打磨结构、打磨工具、位移传感器、待磨工件及其它夹具)都是位置感知系统的一部分，浮动打磨结构中任意部分的变形都最终反应在位移传感的输出上；又进一步由于浮动打磨结构恒力浮动，且机器人打磨设备判断打磨的结果是通过浮动部与固定部之间的相对位移实现的，那么无论是打磨目标发生形变，或是磨具发生了磨损，位移传感器的位移数据都能获得自动补偿，从而大大提高了打磨精度。
(6)本方法处理速度快、自动化程度高、打磨效果好，具有广泛的应用前景。
附图说明
图1-1是实施例1中数据1示意图。
图1-2是实施例1中数据2示意图。
图1-3是实施例1中数据10示意图。
图1-4是实施例1中数据20示意图。
图1-5是实施例1中数据3示意图。
图1-6是实施例1中数据4示意图。
图1-7是实施例1中数据6示意图。
图2-1是实施例2中数据12和数据22示意图。
图2-2是实施例2中数据13和数据23示意图。
图2-3是实施例2中数据15和数据25示意图。
图2-4是实施例2中数据52示意图。
图2-5是实施例2中数据53示意图。
图2-6是实施例2中数据4示意图。
图2-7是实施例2中数据3示意图。
图2-8是实施例2中数据6示意图。
图2-9是实施例2中焊缝及周边位置的截面的示意图。
图3是浮动打磨结构的截面结构示意图。
图中所示标记为：1：浮动轴；2：恒压外缸体；3：恒压内缸体；4：浮动活塞；5：第一腔室；6：第二腔室；7：第一进气口；8：第二进气口；9：滚动体。
具体实施方式
以下将结合具体实施例对本发明作进一步说明，但本发明的保护范围不限于以下实施例。
实施例1
本实施例为本发明的方法在单焊缝打磨中的具体应用。
S1.机器人打磨设备包括机器人、浮动打磨结构和打磨工具，其中所述浮动打磨结构包括固定部、能相对固定部移动的浮动部以及测量固定部和浮动部相对位移的位移传感器；按照打磨的情况安装机器人、浮动打磨结构、打磨目标和打磨工具。本实施例中，安装的方法为机器人通过法兰连接浮动打磨结构，浮动打磨结构前端连接角磨机，角磨机上装有砂轮。
S2.机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标表面并相对于打磨目标运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；所述特定坐标系为大地坐标系或基坐标系或工具坐标系或用户坐标系。
本实施例中，具体的实际特征数据获取过程为：
S2.1建立用户坐标系：以砂轮在打磨目标上运动的起始点作为坐标点原点O，建立用户坐标系Pxyz，其中X轴与焊缝方向垂直，Y轴为焊缝直线延伸方向且指向运动轨迹方向，Z轴按右手定则确定。
S2.2扫描：机器人沿着设定的Y轴方向运动，使砂轮和打磨目标接触，即同时也会与焊缝接触并对焊缝进行扫描，扫描得机器人沿着X轴方向移动的点的坐标数据和浮动部相对固定部移动的位移量，其中位移量即浮动部的浮动量，由位移传感器测得；如图1-1所示，机器人运动轨迹坐标数据记为数据1，如图1-2所示，位移传感器数据记为数据2，坐标轴单位为机器人长度单位。
S2.3转换矩阵运算：将步骤S2.2获得的数据1和数据2转换到建立的用户坐标系Pxyz下的点坐标数据，得到数据10和数据20。
由标准的坐标转换公式可知，空间任意点A平移或旋转得到点B的公式为：
空间点A(x，y，z)平移至点B(x1，y1，z1)：
x1＝x+Δx
y1＝y+Δy
z1＝z+Δz
Figure PCTCN2018100476-appb-000001
通过矩阵I换算得到平移后点的位置：
Figure PCTCN2018100476-appb-000002
空间点A(x，y，z)绕Z轴旋转θ角至点B(x1，y1，z1)：
x1＝x*cosθ-y*sinθ
y1＝x*sinθ+y*cosθ
z1＝z
Figure PCTCN2018100476-appb-000003
通过矩阵Ⅲ换算得到旋转后点的位置：
Figure PCTCN2018100476-appb-000004
同理，空间点A(x，y，z)绕X轴旋转α角至点B(x1，y1，z1)：
Figure PCTCN2018100476-appb-000005
空间点A(x，y，z)绕Y轴旋转β角至点B(x1，y1，z1)：
Figure PCTCN2018100476-appb-000006
使用以上已知标准的数学方法，将数据1和数据2转换到Pxyz坐标系下，如图1-3和图1-4所示，坐标轴单位为机器人长度单位，得到数据10和数据20。
由图1-3可知，X方向上的运动量很小，可以简化为Y和Z向的运动。
S2.4数据10和数据20相加，得到Pxyz坐标系焊缝的实际特征数据3。为了便于分析，如图1-5所示，在表示数据3时，做等间隔处理并把零点去掉(即曲线整体向下平移)。
S3.获取特定坐标系下打磨目标的理论参考数据4，数据4的获取方法包括以下方式的一种或多种：(1)打磨目标自身的理论数据；(2)通过线性拟合方法分析数据3获得理论参考数据4；(3)按步骤S2的方法获取打磨目标周边位置或类似位置的实际特征数据5，通过线性拟合方法计算获得理论参考数据4。
本实施例采用方法(2)，通过线性拟合方法分析数据3获得理论参考数据4。
本实施例采用的线型拟合方法为最小二乘法，用最小二乘法提取实际特征数据3的基线，该基线就是焊缝需要打磨到的理论参考数据4。
由于X方向上的运动量很小，数据3仅简化为Y向和Z向来分析。假设焊缝的理论参考数据4是一条直线，用方程z＝ay+b表示，因此，理论曲线各点的数学表达式为:
z 1＝ay 1+b
z 2＝ay 2+b
z 3＝ay 3+b
……..
z n＝ay n+b
z n是数据3的Z坐标数据，y n是数据3的Y坐标数据，因此可获得n个方程，而待求的未知量只有a和b。本实施例采用最小二乘法来求解最优的a和b。
最小二乘法的典型公式如下：
AX＝B，X是待求的量，A、B是已知量。
设A矩阵为，
Figure PCTCN2018100476-appb-000007
其中y n是数据3的Y坐标数据。
B矩阵为:
Figure PCTCN2018100476-appb-000008
其中z n是数据3的Z坐标数据。
X矩阵为:
Figure PCTCN2018100476-appb-000009
由矩阵变换公式得以下：
A T*AX＝A T*B；
(A T*A) -1*A T*A X＝(A T*A) -1*A T*B；
X＝(A T*A) -1*A T*B
如图1-6所示，采用以上公式得到数据3的基线，即获得理论参考数据4。
S4.数据3减去数据4，获得待打磨部位的待打磨量数据6，并根据数据6的特征，制定打磨策略；数据6如图1-7所示，从待打磨量数据6的特征看出，P1、P2、P3、P4、P5位置有较大的待打磨量，调整机器人运动速度，使得机器人在这几个位置的运动速度变慢，以达到增加打磨时间的效果。
S5.采用与步骤S1相同的机器人、浮动打磨结构、打磨目标和打磨工具的安装方式进行打磨。
S6.机器人打磨设备打磨的同时获取机器人运动轨迹坐标数据1i和位移传感器数据2i，以与步骤S2相同的方法获得打磨后打磨目标的实际特征数据3i，并与打磨目标的理论参考数据4 做比对，以调整打磨策略。
需要指出的是，理论参考数据4也可以按照上述方法在每轮打磨后进行修正。
重复步骤S6，直到待打磨的焊缝实际特征数据3i和理论参考数据4相差在允许范围以内，判断打磨结束。
步骤S2和S5中，机器人打磨设备均可以在打磨工作的同时采集机器人坐标数据和位移传感器数据。
实施例2
本实施例为本发明的方法在单焊缝打磨中的具体应用。
S1.机器人打磨设备包括机器人、浮动打磨结构和打磨工具，其中所述浮动打磨结构包括固定部、能相对固定部移动的浮动部以及测量固定部和浮动部相对位移的位移传感器；按照打磨的情况安装机器人、浮动打磨结构、打磨目标和打磨工具。本实施例中，安装的方法为机器人通过法兰连接浮动打磨结构，浮动打磨结构前端连接角磨机，角磨机上装有砂轮。
S2.机器人按照预定轨迹运动，使打磨工具或浮动打磨结构和待打磨目标周边位置接触并相对运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标周边的特征数据5，通过线性拟合方法计算获得理论参考数据4。所述特定坐标系为大地坐标系或极坐标系或工具坐标系或用户坐标系。
本实施例中，具体的实际特征数据获取过程为：
S2.1建立用户坐标系：以砂轮在打磨目标上运动的起始点作为扫描焊缝原始坐标点O，建立用户坐标系Pxyz，其中X轴在焊缝方向上且指向运动轨迹的反方向，Y轴与焊缝方向垂直，Z轴按右手定则确定。
S2.2扫描：机器人沿着设定的X轴负方向运动，使砂轮和打磨目标左侧和右侧接触进行扫描，如图2-1和图2-2所示，扫描得机器人沿着X轴负方向移动的点的机器人运动轨迹坐标数据12和13和浮动部相对固定部移动的位移量数据22和23，其中位移量即浮动部的浮动量，由位移传感器测得。
S2.3转换矩阵运算：运算方法和实施例1所述的数学方法同理；步骤S2.2获得的数据12和数据22转换到建立的用户坐标系Pxyz下的点坐标数据，如图2-3所示，得到数据15和数据25。
S2.4数据15和数据25相加，如图2-4所示，得到Pxyz坐标系焊缝左侧的实际特征数据52。
S2.5与步骤S2.3和S2.4同理计算出Pxyz坐标系焊缝右侧的实际特征数据53，实际特征数据53如图2-5所示。
S3.5用线性拟合方法拟合实际特征数据52和53，拟合的结果作为理论参考数据4，理论参考数据4如图2-6所示。
具体方法为假设焊缝位于一个平面上，焊缝两边和焊缝位置平滑过渡。焊缝及周边位置的截面可简化为如图2-9所示。其中D代表焊缝左侧，E代表焊缝右侧，F代表焊缝，DF长度为L df，EF长度为L ef，L df和L ef可在机器人运动轨迹中事先设定，是已知量。设D点坐标为(Y d,Z d)，E点坐标为(Y e,Z e)，F点坐标为(Y f,Z f)。由三角函数可得：
L1/(L df+L ef)＝L2/L df
L2＝(L1*L df)/(L df+L ef)
其中，L1＝Z e–Z d；L2＝Z f–Z d
因此，F点的纵坐标为：
Z f＝Z d+(L1*L df)/(L df+L ef)
S3.机器人按照预定轨迹运动，使浮动部前部连接的砂轮接触到打磨目标表面并相对于打磨目标运动，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；所述特定坐标系为大地坐标系或基坐标系或工具坐标系或用户坐标系；数据3如图2-7所示。
S4.数据3减去数据4，获得打磨目标的待打磨量数据6，并根据数据6的特征，制定打磨策略；由于机器人在Y向的运动量很小，工程处理上可忽略，且对数据做去零点等处理，数据52、数据53、数据3、数据4和数据6表示如图2-8所示。
从待打磨量数据6的特征看出，P至Q位置都有较大的待打磨量，调整机器人运动速度，使得机器人在这个区域的运动速度变慢，以达到增加打磨时间的效果。
S5.采用与步骤S1相同的机器人、浮动打磨结构、打磨目标和打磨工具的安装方式进行打磨。
S6.机器人打磨设备打磨的同时获取机器人运动轨迹坐标数据1i和位移传感器数据2i，以与步骤S3相同的方法获得打磨后打磨目标的实际特征数据3i，并与打磨目标的理论参考数据4做比对，以调整打磨策略。
重复步骤S6，直到待打磨的焊缝特征数据3i和理论参考数据4相差在允许范围以内，判断打磨结束。
本实施例中，步骤S3和S5中，机器人打磨设备可以在打磨工作的同时采集机器人坐标数据和位移传感器数据。
上述实施例1和2中，为实现打磨工具变形的自动补偿，更优选的是无论是扫描还是打磨时浮动打磨结构为恒力浮动，即向浮动部施加恒定的压力，使浮动部能相对固定部轴向或/和径向柔顺浮动；为更进一步实现打磨目标变形的自动补偿，更优选的，在打磨工具或浮动打磨结构进行扫描和/或打磨时，浮动打磨结构都采用相同的恒力，以保证打磨目标变形量相同，位移传感器获得的数据实现自动补偿。
本实施例进一步提供了优选的浮动部能相对固定部轴向浮动的浮动打磨结构。如图3所示，其包括有浮动轴1、恒压外缸体2、恒压内缸体3和浮动活塞4：恒压内缸体3和浮动活塞4分别套设于浮动轴1的外部，恒压外缸体2套设于恒压内缸体3和浮动活塞4的外部，恒压外缸体2和恒压内缸体3之间形成一腔室，浮动活塞4一端与浮动轴1固定连接，另一端位于腔室内并将腔室分隔为第一腔室5和第二腔室6，恒压外缸体2的侧壁上设置有第一进气口7和第二进气口8，第一进气口7与第一腔室5相连通，第二进气口8与第二腔室6相连通。
本实施例的浮动打磨结构处于工作状态时，通过第一进气口7或者第二进气口8向第一腔室5或者第二腔室6内充气。当通过第一进气口7向第一腔室5内充气时，浮动活塞4受气流作用向第二腔室6方向移动，由于浮动活塞4与浮动轴1固定连接，因此浮动活塞4带动浮动轴1一起运动，浮动轴1处于伸出的状态。当第一腔室5输入的气流一定时，浮动活塞4受力一定，因此通过浮动活塞4传递给浮动轴1的力是恒定的，即所说的轴向恒力浮动。当通过第二进气口8向第二腔室6内充气时，浮动活塞4受气流作用向第一腔室5方向移动，浮动活塞4带动浮动轴1一起运动，浮动轴1处于缩入的状态。当第二腔室6输入的气流一定时，浮动活塞4受力一定，因此通过浮动活塞4传递给浮动轴1的力也是恒定的。值得注意的是，由于浮动活塞4可在恒压外缸体2和恒压内缸体3形成的腔室内活动，因此第一腔室5和第二腔室6的界线并没有明显的界定。另外，还值得注意的是，浮动活塞4与恒压外缸体2并不一定是完全绝密的，因此当第一进气口7向第一腔室5内充气时，气体也可通过浮动活塞4与恒压外缸体2之间的缝隙以及第二进气口8排出；当第二进气口8向第二腔室6内充气时，气体也可通过浮动活塞4与恒压外缸体2之间的缝隙以及第一进气口7排出。通过上述的设置，可实现在轴向恒力浮动，这个恒力，可以是工作时对工件恒定下压的力，也可以是对工件恒定上拉的力，不管是哪种恒力，都能实现对加工器件的轴向恒力作用，从而起到恒力作用、保护器件、提高精度的目的。
为减少摩擦力，浮动轴1和恒压内缸体3之间设有直线导向机构，直线导向机构使得浮动轴1和恒压内缸体3之间只能进行轴向的直线运动。直线导向机构包括：设于浮动轴1的外表面上的第一滑槽、设于恒压内缸体3的内表面且与第一滑槽位置相对的第二滑槽以及设于第一滑槽和第二滑槽内的多个滚动体9。通过这种优选的设计，既能进一步限制浮动轴1只能进行轴 向的直线运动，且保证了轴向的直线运动的流畅性。
本实施例的浮动部即为浮动轴1和/或浮动活塞4，固定部即为恒压外缸体2和/或恒压内缸体3，为检测浮动部和固定部的相对位移，在浮动部和固定部上安装能测量他们之间相对位移的位移传感器。
本实施例所说的恒力，是工程意义上的近似恒力。
上述实施例1和2中，机器人、浮动打磨结构、打磨目标和打磨工具的安装方式也可以替换为其他形式，例如：(1)机器人前端连接浮动打磨结构，浮动打磨结构前端连接打磨目标，打磨工具固定在其他地方；(2)机器人前端连接打磨工具，打磨目标和浮动打磨结构连接并固定在其他地方；(3)机器人连接打磨目标，打磨工具和浮动打磨结构连接并固定在其他地方。
上述实施例的打磨方法，具有以下优点：
(1)适应性好：本发明采用了浮动打磨结构，浮动部能相对固定部浮动。在打磨和位置测量时，浮动功能能够自动的适应工件本身的尺寸差异和工件的定位精度误差。且采用打磨工具或浮动打磨结构和被测物体接触的方法来测量位置和尺寸数据，因此粉尘和振动对测量的影响小。
(2)准确的位置感知：浮动打磨结构内部有位移传感器，能测量浮动部相对固定部的相对位置变化，为智能打磨系统提供了打磨目标的尺寸信息信息，结合位移传感器能有效的获取的位置信息，即可以准确获取打磨目标具体部位的待打磨量和打磨后的剩余量。
(3)实时性好：在打磨和位置测量步骤中，本方法能综合采集位移传感器数据和机器人运动轨迹坐标数据，两者有较好的对应关系，可实现实时的同步。
(4)系统相对简单：通过上述方法获取打磨目标的实际特征和打磨目标的理论特征数据，两者相减就能得到待打磨量，进而可制定针对性的打磨策略，无须经过复杂的力学数据转换分析处理，处理效率高，打磨效果好，过程相对简单。
(5)具有自动补偿功能：浮动打磨结构包含有位移传感器，使得整个打磨系统(包括机器人、浮动打磨结构、打磨工具、位移传感器、待磨工件及其它夹具)都是位置感知系统的一部分，浮动打磨结构中任意部分的变形都最终反应在位移传感的输出上；又进一步由于浮动打磨结构恒力浮动，且机器人打磨设备判断打磨的结果是通过浮动部与固定部之间的相对位移实现的，那么无论是打磨目标发生形变，或是磨具发生了磨损，位移传感器的位移数据都能获得自动补偿，从而大大提高了打磨精度。

Claims (9)

  1. 一种智能打磨方法，其特征在于包括以下步骤：
    S1.机器人打磨设备包括机器人、浮动打磨结构和打磨工具，其中所述浮动打磨结构包括固定部、能相对固定部移动的浮动部以及测量固定部和浮动部相对位移的位移传感器；按照打磨工作的需求安装机器人、浮动打磨结构、打磨目标和打磨工具；
    S2.机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标表面并相对于打磨目标运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；所述特定坐标系为大地坐标系或基坐标系或工具坐标系或用户坐标系；
    S3.获取特定坐标系下打磨目标的理论参考数据4，数据4的获取方法包括以下方式的一种或多种：(1)打磨目标自身的理论数据；(2)通过线性拟合方法分析数据3获得理论参考数据4；(3)按步骤S2的方法获取打磨目标周边位置或类似位置的实际特征数据5，通过线性拟合方法计算获得理论参考数据4；
    S4.数据3减去数据4，获得打磨目标的待打磨量数据6，并根据数据6的特征，制定打磨策略；
    S5.采用与步骤S1相同的机器人、浮动打磨结构、打磨目标和打磨工具的安装方式对打磨目标进行打磨。
  2. 根据权利要求1所述的智能打磨方法，其特征在于：
    所述S2和S3的步骤替换为：
    S2.获取特定坐标系下打磨目标的理论参考数据4，数据4的获取方法包括以下方式的一种或两种：(1)打磨目标自身的理论数据；(2)机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标周边位置或类似位置的表面并相对于打磨目标周边位置或类似位置运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1a和位移传感器数据2a；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1a和数据2a在特定坐标系下进行叠加，获得打磨目标周边位置或类似位置的实际特征数据5，再通过线性拟合方法计算获得理论参考数据4；
    S3.机器人按照预定轨迹运动，使打磨工具或浮动打磨结构接触到打磨目标表面并相对 于打磨目标运动，进行扫描和/或打磨，得到机器人运动轨迹坐标数据1和位移传感器数据2；通过空间点的平移关系转换矩阵和旋转关系转换矩阵运算，将数据1和数据2在特定坐标系下进行叠加，获得打磨目标的实际特征数据3；
    所述特定坐标系为大地坐标系或基坐标系或工具坐标系或用户坐标系。
  3. 根据权利要求1或2所述的智能打磨方法，其特征在于：
    所述步骤S1中，所述浮动部能相对固定部轴向或/和径向浮动。
  4. 根据权利要求1或2所述的智能打磨方法，其特征在于：
    所述打磨策略包括：(1)调整机器人运动轨迹；(2)调整打磨时间、打磨压力和打磨次数；(3)更换打磨工具和/或磨料；(4)调整打磨工具转速。
  5. 根据权利要求1或2所述的智能打磨方法，其特征在于：
    还包括步骤S6：S6.机器人打磨设备打磨的同时获取机器人运动轨迹坐标数据和位移传感器数据，以与步骤S2或S3相同的扫描方法获得打磨后打磨目标的实际特征数据3i，并与打磨目标的理论参考数据4做比对，以调整打磨策略或判断打磨是否结束。
  6. 根据权利要求1所述的智能打磨方法，其特征在于：
    所述步骤S5中，机器人打磨设备在打磨工作的同时采集机器人坐标数据和位移传感器数据。
  7. 根据权利要求1或2所述的智能打磨方法，其特征在于：
    所述浮动打磨结构恒力浮动。
  8. 根据权利要求1或2所述的智能打磨方法，其特征在于：
    所述打磨工具或浮动打磨结构接触到打磨目标及打磨目标周边位置或类似位置表面进行扫描和/或打磨时，浮动打磨结构采用相同的恒力。
  9. 根据权利要求1所述的智能打磨方法，其特征在于：
    所述步骤S1中，机器人、浮动打磨结构、打磨目标和打磨工具的安装方式为：(1)机器人前端连接浮动打磨结构，浮动打磨结构前端连接打磨工具，打磨目标固定在其他地方；(2) 机器人前端连接浮动打磨结构，浮动打磨结构前端连接打磨目标，打磨工具固定在其他地方；(3)机器人前端连接打磨工具，打磨目标和浮动打磨结构连接并固定在其他地方；(4)机器人连接打磨目标，打磨工具和浮动打磨结构连接并固定在其他地方。
PCT/CN2018/100476 2017-09-11 2018-08-14 一种智能打磨方法 WO2019047683A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710812010.9 2017-09-11
CN201710812010.9A CN107756145B (zh) 2017-09-11 2017-09-11 一种智能打磨方法

Publications (1)

Publication Number Publication Date
WO2019047683A1 true WO2019047683A1 (zh) 2019-03-14

Family

ID=61265703

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/100476 WO2019047683A1 (zh) 2017-09-11 2018-08-14 一种智能打磨方法

Country Status (2)

Country Link
CN (1) CN107756145B (zh)
WO (1) WO2019047683A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107756145B (zh) * 2017-09-11 2020-05-22 南宁宇立仪器有限公司 一种智能打磨方法
CN108818160B (zh) * 2018-06-29 2020-10-16 贵州筑信达创科技有限公司 一种平面石板打磨抛光加工控制方法
CN110774102B (zh) * 2019-11-04 2021-07-27 广东博智林机器人有限公司 一种打磨点定位系统及定位方法
CN110842782B (zh) * 2019-12-25 2021-04-06 东风汽车有限公司 一种浮动感知打磨头装置、恒力打磨方法
CN112123989A (zh) * 2020-09-22 2020-12-25 西京学院 一种绿色环保无污染工艺美术品制作方法及智能系统
CN113836110B (zh) * 2021-08-23 2024-01-19 中车青岛四方机车车辆股份有限公司 一种建立平面腻子机器人打磨数据库的方法
CN115056043B (zh) * 2022-06-14 2023-06-06 惠州市盈泰欣精密金属科技有限公司 五金件打磨智能自动启停方法、装置、计算机和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110223836A1 (en) * 2010-03-12 2011-09-15 Duescher Wayne O Three-point fixed-spindle floating-platen abrasive system
CN103862380A (zh) * 2014-03-26 2014-06-18 上海交通大学 光学镜面少轴磨削倾斜弧面砂轮误差分析与补偿系统
CN103885334A (zh) * 2013-12-30 2014-06-25 深圳先进技术研究院 一种抛光打磨力-位混合控制方法和系统
CN107756145A (zh) * 2017-09-11 2018-03-06 南宁宇立仪器有限公司 一种智能打磨方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3221758A1 (de) * 1982-06-09 1983-12-15 Hartmann & Lämmle GmbH & Co KG, 7255 Rutesheim Hydraulische antriebsvorrichtung
CN102501172B (zh) * 2011-11-22 2013-10-23 清华大学 用于机器人修磨系统的面向空间曲面加工的在位测量方法
CN103786082B (zh) * 2013-12-26 2016-02-10 吉林大学 恒力磨削抛光工具系统
CN105382697B (zh) * 2015-12-09 2017-11-07 浙江工业大学 一种液动压力、浮力及上浮距离的检测方法
CN106378704B (zh) * 2016-10-28 2018-08-31 南宁宇立仪器有限公司 一种柔性打磨装置
CN106808357A (zh) * 2017-01-20 2017-06-09 广东省智能制造研究所 胚体表面打磨装置与方法
CN106881648B (zh) * 2017-03-27 2022-12-27 江苏科技大学 一种调距桨桨叶表面打磨装置和打磨方法
CN106938423A (zh) * 2017-03-29 2017-07-11 苏州亚思科精密数控有限公司 桨叶表面打磨方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110223836A1 (en) * 2010-03-12 2011-09-15 Duescher Wayne O Three-point fixed-spindle floating-platen abrasive system
CN103885334A (zh) * 2013-12-30 2014-06-25 深圳先进技术研究院 一种抛光打磨力-位混合控制方法和系统
CN103862380A (zh) * 2014-03-26 2014-06-18 上海交通大学 光学镜面少轴磨削倾斜弧面砂轮误差分析与补偿系统
CN107756145A (zh) * 2017-09-11 2018-03-06 南宁宇立仪器有限公司 一种智能打磨方法

Also Published As

Publication number Publication date
CN107756145A (zh) 2018-03-06
CN107756145B (zh) 2020-05-22

Similar Documents

Publication Publication Date Title
WO2019047683A1 (zh) 一种智能打磨方法
CN109489580B (zh) 一种航空发动机叶片表面加工的在机点云检测及补偿方法
JP7171114B2 (ja) シャフトワークのインサイチュ非接触検出方法
Xu et al. TCP-based calibration in robot-assisted belt grinding of aero-engine blades using scanner measurements
US9448552B2 (en) Numerically-controlled machine tool and spindle error compensating method thereof
CN110977962B (zh) 一种基于3d视觉的机器人打磨路径自动纠偏方法
CN110640585A (zh) 一种叶片磨抛加工三维非接触式测量装置及方法
US11992962B2 (en) Robot and robot system
CN107498467B (zh) 一种恒力浮动系统
Guo et al. A robotic grinding motion planning methodology for a novel automatic seam bead grinding robot manipulator
TW202032093A (zh) 自動研磨系統
Zhang et al. Online adaptive measurement and adjustment for flexible part during high precision drilling process
US20210402542A1 (en) Machining device for controlling tool position considering tool wear and method for controlling tool position using the same
CN116060852A (zh) 一种管道内焊机
CN111928776A (zh) 基于多传感器的数控机床非接触式在线测量系统与方法
US20190321967A1 (en) Work robot system and work robot
KR20230051579A (ko) 결함 연삭 시스템, 결함 연삭 방법 및 이것을 이용한 강 제품의 제조 방법
Prabhu et al. Dynamic alignment control using depth imagery for automated wheel assembly
Wang et al. Development and evaluation of non-contact automatic tool setting method for grinding internal screw threads
CN111872796A (zh) 一种用于铸管承口的自动化打磨设备及使用方法
TWM590084U (zh) 用以輔助表面加工裝置的恆力加工設備
TWM538446U (zh) 一種壓力感測補償系統
CN112223102A (zh) 基于视觉定位的珩磨方法和装置
Mikołajczyk Robot application to surface finish machining
CN216967440U (zh) 一种用于打磨抛光的浮动装置

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: 18854489

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: 18854489

Country of ref document: EP

Kind code of ref document: A1