Active and passive compliant fusion robot grinding and polishing machining system and method

Abstract

The invention relates to the technical field of robot intelligent manufacturing, in particular to an active and passive compliance fusion robot grinding and polishing machining system and method, and aims to solve the problems that parameter debugging is tedious and difficult to master, and the contact force control precision is low due to rigidity changes of different parts of a workpiece. The machining system comprises a robot, a belt sander, a six-dimensional force sensor and a one-dimensional force sensor. The machining method comprises the steps that a robot grinding and polishing machining track program is generated; acquiring a contact force F10 of a grinding part at the end of the six-dimensional force sensor; acquiring the contact force F20 of the grinding part of the one-dimensional force sensor end; the F10 and the F20 are fused, and the accurate contact force fe of the grinding part is obtained; comparing fe with a set expected contact force fd to obtain a force deviation value delta f, and performing feedback control on the robot grinding and polishing processing system through the delta f; designing a robot active compliance control system; designing a belt sander passive compliance control system; and actual grinding machining is conducted.

Description

Technical Field

[0001] The present invention relates to the field of robot intelligent manufacturing technology, and in particular to an active, passive and compliant fusion robot grinding and polishing system and method. Background Art

[0002] Robotic belt grinding has a series of advantages such as strong applicability, good flexibility, high processing quality and high degree of automation. It has been widely used in 3C electronics, automobile parts manufacturing, aerospace and weapon equipment and other fields. Usually, there are a series of difficulties such as poor consistency of parts in the same batch, errors in clamping and positioning, and easy deformation of curved workpieces. If a fixed robot grinding program is used, there will be "over-grinding" and "under-grinding" in the local processing parts of the parts, which will seriously affect the contour accuracy and surface quality of robot belt grinding; especially the normal contact force control accuracy of the robot's flexible grinding process will also affect the consistency of surface quality, seriously limiting the promotion and application of robot belt grinding equipment and technology in the field of grinding.

[0003] In order to improve the large-scale application of robots in the field of precision grinding and improve the precision grinding quality of robot belt grinding equipment, domestic patent document CN107962480A discloses a robot constant force control method, which sets different control parameters for different grinding objects to improve the force control accuracy of the grinding process; CN 110561237 B discloses a robot belt grinding system and method for active and passive force control based on force-position hybrid control, which can improve the control accuracy of grinding contact force and optimize the problem of over-grinding and under-grinding in the grinding process.

[0004] Patent documents CN107962480A and CN 110561237 B respectively use active force control and active-passive combined force-position hybrid control to improve the contact force control accuracy of robot grinding. The parameter debugging in patent CN107962480A is cumbersome and difficult to master, and it is difficult to debug appropriate parameters for different processing scenarios. CN 110561237 B is based on the active and passive combination of force-position hybrid control, in which the stiffness of the passive force control of the belt sander is difficult to adjust. The stiffness of different parts of the complex curved workpiece changes during grinding, and the force control effect of the fixed parameter force-position hybrid control model is poor. The change in stiffness of different parts of the workpiece leads to low contact force control accuracy. Problems still exist. Summary of the invention

[0005] In view of this, the purpose of the present invention is to propose an active-passive compliant fusion robot grinding and polishing system and method to solve the problem of cumbersome and difficult to master parameter debugging and low contact force control accuracy caused by changes in stiffness of different parts of the workpiece.

[0006] Based on the above objectives, the present invention provides an active-passive compliant fusion robot grinding and polishing system, comprising a robot, a belt sander, a six-dimensional force sensor arranged at the end of the robot, and a one-dimensional force sensor arranged on the belt sander.

[0007] A robot grinding and polishing method combining active, passive and compliant processing, using the processing system,

[0008] S1: Generate robot grinding and polishing trajectory program;

[0009] S2: Obtain the contact force F of the grinding part at the end of the six-dimensional force sensor 10 ; Obtain the contact force F of the grinding part at the one-dimensional force sensor end 20 ;

[0010] S3: Measuring contact force F with six-axis force sensor 10 The contact force F is measured with a one-dimensional force sensor 20 Fusion is performed to obtain accurate contact force f of the grinding part e ;

[0011] Use the grinding part to accurately measure the contact force f e With the expected contact force f d By comparison, a force deviation value Δf is obtained, and feedback control is performed on the robot grinding and polishing system through the force deviation value Δf;

[0012] S4: Design an active compliance control system for a robot;

[0013] S5: Design a passive compliance control system for a sanding machine;

[0014] S6: Perform actual grinding process.

[0015] First, the parameters of the robot active compliance control system and the sanding machine passive compliance control system are set; the robot is moved to the starting position of the robot grinding and polishing trajectory, the robot grinding and polishing trajectory program is run, and the robot active compliance control system and the sanding machine passive compliance control system are turned on at the same time.

[0016] As a further improvement of the present application, the design of the robot active compliance control system includes equating the robot-environment contact model to a second-order mass-spring-damper model, and establishing a dynamic model of the grinding and polishing force on the robot end workpiece and the environmental position:

[0017]

[0018] Where m is the inertia coefficient of the robot, b is the damping coefficient, k is the stiffness coefficient, and Δf is the desired force f d and f e The difference, e = xm -x r is the robot reference position error;

[0019] When the robot comes into contact with the environment, the force on the end of the robot is:

[0020] f e =k e (x m -x e )

[0021] where k e is the environmental stiffness, x e is the actual position of the environment, x m is the actual position of the robot end;

[0022] By estimating the stiffness of the environment and environmental location So that the f at the end of the robot e Approximation to the expected contact force f d , and adaptively adjust the robot's expected position, the robot's expected position x r for:

[0023]

[0024] in For x e The estimated value of k e The estimated value of

[0025] The estimated environment stiffness and environmental location To design an adaptive control algorithm to estimate the environment position x e and environmental stiffness k e :

[0026]

[0027] where λ 1 ,λ 2 is a given positive coefficient.

[0028] As a further improvement of the present application, the passive compliance control system of the sanding machine is designed to include:

[0029] The contact force of the belt grinding machine during grinding is detected by a one-dimensional force sensor in the horizontal motion component. The measured data is processed and fed back to the controller. The belt grinding machine compliance control module is designed based on the impedance control theory of position.

[0030]

[0031] Where md is the belt sander quality coefficient, b d is the damping coefficient of the belt sander, Δf is the deviation between the expected force and the actual contact force, e is the deviation between the expected position and the actual position of the belt sander, is the second-order derivative of the position error, is the first-order derivative of the position deviation;

[0032] By f e With the expected force f d The error Δf is converted into the forward and backward position movement of the belt sander, thereby achieving a passive compliance control effect. The expression of the offset obtained by the control model is:

[0033]

[0034] Wherein, the sampling period is time t, x(t) is the speed of the motion control component moving forward and backward at time t, Δf is the force deviation, and m d is the belt sander quality coefficient, b d is the damping coefficient of the belt sander, which is adjusted by m d and b d The value of is used to change the flexibility of the belt sander, x(0) is the initial position speed, e is the deviation between the expected position and the actual position of the belt sander, and τ is the time delay of the system;

[0035] Discretize it, set the sampling time Δt to 4ms, within the sampling time m d , b d All are constants

[0036]

[0037] Δx(k)=v n (k)Δt

[0038] where v n (k) is the speed of the motor at the current moment, Δx is the offset that the horizontal motion component of the belt sander needs to compensate at the current moment, and the offset is sent to the motor to cause the slide to move.

[0039] As a further improvement of the present application, the contact force F of the grinding part of the six-dimensional force sensor end is obtained. 10 , including preprocessing the six-dimensional force sensor and performing Kalman filtering and denoising on the data collected by the six-dimensional force sensor, so as to obtain the contact force F of the grinding part 10 .

[0040] As a further improvement of the present application, the preprocessing of the six-dimensional force sensor includes gravity compensation for the robot end tool and the six-dimensional force sensor, including collecting the six-dimensional force sensor data of the robot in different postures and calculating the force offset F of the three coordinate axes of the sensor. x0 , F y0 , F z0 And the gravity F of the end fixture and sensor on the coordinate axis x1 , F y1 , F z1 And remove the influence of gravity in the measurement data.

[0041] As a further improvement of the present application, the contact force F of the grinding part at the one-dimensional force sensor end is obtained. 20 ,include:

[0042] The data fitting method is used, F 20 =A*U+a;

[0043] Among them, F 20 is the measured contact force, U is the voltage value of the sensor, a is the coefficient, and A is the sensor calibration fitting coefficient.

[0044] The six-dimensional force sensor measures the contact force F 10 The contact force F is measured with a one-dimensional force sensor 20 To integrate, including

[0045] The weights W of the six-dimensional force sensor and one-dimensional force sensor measurements are calculated using similarity theory. i (k):

[0046] W i (k)=[W 1 (k), W 2 (k), ....W i (k)]

[0047] Wherein, k represents the time, i represents the number of sensors, and in the present invention, there are 2 sensors, so K=2, and the fusion calculation process expression is as follows:

[0048]

[0049] The beneficial effects of the present invention are as follows: the contact force of the grinding part is precisely measured by the six-dimensional force sensor at the end of the robot and the one-dimensional force sensor on the belt sander, and the measurement data is filtered and processed. The multi-sensor information fusion technology is used to improve the measurement accuracy and precision of the contact force of the grinding part. The active-passive compliance control technology is combined to ensure that the stiffness of the robot belt grinding system remains constant when grinding different parts of the workpiece, reduce the impact on the stability performance of the active compliance control system, and reduce the difficulty of adjusting the parameters of the active compliance control system. At the same time, the contact force control accuracy of the grinding process is also significantly improved, ensuring the removal of product grinding materials and the consistency of surface quality, and improving the grinding quality and efficiency of the robot belt grinding system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 It is a schematic structural diagram of an active, passive and compliant fusion robot grinding and polishing system according to an embodiment of the present invention;

[0052] Figure 2 A schematic flow chart of an active, passive and compliant fusion robot grinding and polishing method according to an embodiment of the present invention;

[0053] Figure 3 The present invention is a schematic diagram of the force fusion process of an active, passive and compliant fusion robot grinding and polishing method according to an embodiment of the present invention.

[0054] The markings in the figure are:

[0055] 1. Robot; 2. Belt sander. DETAILED DESCRIPTION

[0056] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0057] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0058] like Figure 1-3 As shown, an active-passive compliant fusion robot grinding and polishing system includes a robot 1, a belt sander 2, a six-dimensional force sensor arranged at the end of the robot 1, and a one-dimensional force sensor arranged on the belt sander.

[0059] A robot grinding and polishing method combining active, passive and compliant processing, using the processing system,

[0060] S1: Generate robot grinding and polishing trajectory program;

[0061] S2: Obtain the contact force F of the grinding part at the end of the six-dimensional force sensor 10 , including preprocessing the six-dimensional force sensor and performing Kalman filtering and denoising on the data collected by the six-dimensional force sensor, so as to obtain the contact force F of the grinding part 10 .

[0062] The preprocessing of the six-dimensional force sensor includes zero drift of the contact force signal collected by the six-dimensional force sensor at the end of the robot in real time, gravity compensation of the robot end tool and the six-dimensional force sensor, including collecting the six-dimensional force sensor data under different postures of the robot and calculating the force offset F of the three coordinate axes of the sensor. x0 , F y0 , F z0 And the gravity F of the end fixture and sensor on the coordinate axis x1 , F y1 , F z1 And remove the influence of gravity in the measurement data.

[0063] Obtain the contact force F of the grinding part at the end of the one-dimensional force sensor 20 ,include:

[0064] The data fitting method is used, F20 =A*U+a;

[0065] Among them, F 20 is the measured contact force, U is the voltage value of the sensor, a is the coefficient, and A is the sensor calibration fitting coefficient.

[0066] S3: Measuring contact force F with six-axis force sensor 10 The contact force F is measured with a one-dimensional force sensor 20 Fusion is performed to obtain accurate contact force f of the grinding part e ;include

[0067] The weights W of the six-dimensional force sensor and one-dimensional force sensor measurements are calculated using similarity theory. i (k):

[0068] W i (k)=[W 1 (k), W 2 (k), ....W i (k)]

[0069] Wherein, k represents the time, i represents the number of sensors, and in the present invention, there are 2 sensors, so K=2, and the fusion calculation process expression is as follows:

[0070]

[0071] Use the accurate contact force fe of the grinding part and set the expected contact force f d By comparison, a force deviation value Δf is obtained, and feedback control is performed on the robot grinding and polishing system through the force deviation value Δf;

[0072] S4: Design an active compliance control system for the robot, including:

[0073] The contact model between the robot and the environment is equivalent to a second-order mass-spring-damper model, and a dynamic model of the grinding and polishing force on the robot end workpiece and the environmental position is established:

[0074]

[0075] Where m is the inertia coefficient of the robot, b is the damping coefficient, k is the stiffness coefficient, and Δf is the desired force f d and f e The difference, e = x m -x r is the robot reference position error;

[0076] When the robot comes into contact with the environment, the force on the end of the robot is:

[0077] f e =k e(x m -x e )

[0078] where k e is the environmental stiffness, x e is the actual position of the environment, x m is the actual position of the robot end;

[0079] In order to reduce the force deviation value Δf; by estimating the stiffness of the environment and environmental location So that the f at the end of the robot e Approximation to the expected contact force f d , and adaptively adjust the robot's expected position, the robot's expected position x r for:

[0080]

[0081] in For x e The estimated value of k e The estimated value of

[0082] The estimated environment stiffness and environmental location To design an adaptive control algorithm to estimate the environment position x e and environmental stiffness k e :

[0083]

[0084] where λ 1 ,λ 2 As a given positive coefficient, the adaptive impedance algorithm is helpful to improve the tracking accuracy of the robot grinding force.

[0085] S5: Design a passive compliance control system for a sanding machine, including:

[0086] The contact force of the belt grinding machine during grinding is detected by a one-dimensional force sensor in the horizontal motion component, and the measured data is fed back to the controller after processing. The passive compliance control module of the belt grinding machine is designed based on the impedance control theory of position.

[0087]

[0088] Among them, m d is the belt sander quality coefficient, b d is the damping coefficient of the belt sander, Δf is the deviation between the expected force and the actual contact force, e is the deviation between the expected position and the actual position of the belt sander, is the second-order derivative of the position error, is the first-order derivative of the position deviation;

[0089] By f e With the expected force f d The error Δf is converted into the forward and backward position movement of the belt sander, thereby achieving a passive and compliant control effect.

[0090]

[0091] The expression of the offset obtained from the control model is:

[0092]

[0093] Wherein, the sampling period is time t, x(t) is the speed of the motion control component moving forward and backward at time t, Δf is the force deviation, and m d is the belt sander quality coefficient, b d is the damping coefficient of the belt sander, which is adjusted by m d and b d The value of is used to change the flexibility of the belt sander, x(0) is the initial position speed, e is the deviation between the expected position and the actual position of the belt sander, is the second-order derivative of the position error, is the first-order derivative of the position error, τ is the time delay of the system;

[0094] Discretize it, set the sampling time Δt to 4ms, within the sampling time m d , b d All are constants

[0095]

[0096] Δx(k)=v n (k)Δt

[0097] where v n (k) is the speed of the motor at the current moment, Δx is the offset that needs to be compensated for the horizontal motion component of the belt sander at the current moment, and the offset is sent to the motor to cause the slide to move, which makes the horizontal motion component have a passive compliance function, ensuring the consistent stiffness of the robot grinding system during processing.

[0098] S6: Perform actual grinding process.

[0099] First, before the robot belt grinding system works, the robot active compliance system control parameter m 1 , b 1 , k 1 and λ 1 ,λ 2 The parameters of the belt sander are set to control the control parameters m in the belt sander compliance control system. d , b dSet up;

[0100] The robot is moved to the starting position of the robot grinding and polishing machining trajectory, the robot grinding and polishing machining trajectory program is run, and the robot active compliance control system and the belt machine compliance control system are started at the same time.

[0101] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0102] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A robot grinding and polishing system integrating active, passive and compliant processing, characterized in that: It comprises a robot (1), a sanding machine (2), a six-dimensional force sensor arranged at the end of the robot (1), and a one-dimensional force sensor arranged on the sanding machine.

2. A robot grinding and polishing method combining active and passive compliance, using the processing system described in claim 1, characterized in that: S1: Generate robot grinding and polishing trajectory program; S2: Obtain the contact force F of the grinding part at the end of the six-dimensional force sensor 10 ; Obtain the contact force F of the grinding part at the one-dimensional force sensor end 20 ; S3: Measuring contact force F with six-axis force sensor 10 The contact force F is measured with a one-dimensional force sensor 20 Fusion is performed to obtain accurate contact force f of the grinding part e ; Use the grinding part to accurately determine the contact force f e With the expected contact force f d By comparison, a force deviation value Δf is obtained, and feedback control is performed on the robot grinding and polishing system through the force deviation value Δf; S4: Design an active compliance control system for a robot; S5: Design a passive compliance control system for a sanding machine; S6: Perform actual grinding process. First, the parameters of the robot active compliance control system and the sanding machine passive compliance control system are set; the robot is moved to the starting position of the robot grinding and polishing trajectory, the robot grinding and polishing trajectory program is run, and the robot active compliance control system and the sanding machine passive compliance control system are turned on at the same time.

3. The active and passive compliant fusion robot grinding and polishing method according to claim 2 is characterized in that: The design of the robot active compliance control system includes equating the contact model between the robot and the environment to a second-order mass-spring-damper model, and establishing a dynamic model of the grinding and polishing force on the robot end workpiece and the environmental position: Where m is the inertia coefficient of the robot, b is the damping coefficient, k is the stiffness coefficient, and Δf is the desired force f d and f e The difference, e = x m -x r is the robot reference position error; When the robot comes into contact with the environment, the force on the end of the robot is: f e =k e (x m -x e ) where k e is the environmental stiffness, x e is the actual position of the environment, x m is the actual position of the robot end; By estimating the stiffness of the environment and environmental location So that the f at the end of the robot e Approximation to the expected contact force f d , and adaptively adjust the robot's expected position, the robot's expected position x r for: in For x e The estimated value of k e The estimated value of .

4. The active and passive compliant fusion robot grinding and polishing method according to claim 3 is characterized in that: The estimated environment stiffness and environmental location To design an adaptive control algorithm to estimate the environment position x e and environmental stiffness k e : Where λ1 and λ2 are given positive coefficients.

5. The active and passive compliant fusion robot grinding and polishing method according to claim 2 is characterized in that: The designed passive compliance control system of the sanding belt machine includes: The contact force of the belt grinding machine during grinding is detected by a one-dimensional force sensor in the horizontal motion component, and the measured data is fed back to the controller after processing, and the belt grinding machine compliance control module is designed based on the impedance control theory of position; Among them, m d is the belt sander quality coefficient, b d is the damping coefficient of the belt sander, Δf is the deviation between the expected force and the actual contact force, e is the deviation between the expected position and the actual position of the belt sander, is the second-order derivative of the position error, is the first-order derivative of the position deviation; By f e With the expected force f d The error Δf is converted into the forward and backward position movement of the belt sander, thereby achieving a passive compliance control effect. The expression of the offset obtained by the control model is: The sampling period is t, x(t) is the speed of the motion control component moving forward and backward at time t, Δf is the force deviation, and m d is the belt sander quality coefficient, b d is the damping coefficient of the belt sander, which is adjusted by m d and b d The value of is used to change the flexibility of the belt sander, x(0) is the initial position speed, e is the deviation between the expected position and the actual position of the belt sander, and τ is the time delay of the system; Discretize it, set the sampling time Δt to 4ms, within the sampling time m d 、b d All are constants Δx(k)=v n (k)△t where v n (k) is the speed of the motor at the current moment, Δx is the offset that the horizontal motion component of the belt sander needs to compensate at the current moment, and the offset is sent to the motor to cause the slide to move.

6. The active and passive compliant robot grinding and polishing method according to claim 2 is characterized in that: The contact force F of the grinding part at the six-dimensional force sensor end is obtained. 10 , including preprocessing the six-dimensional force sensor and performing Kalman filtering and denoising on the data collected by the six-dimensional force sensor, so as to obtain the contact force F of the grinding part 10 .

7. The active and passive compliant robot grinding and polishing method according to claim 2 is characterized in that: The preprocessing of the six-dimensional force sensor includes gravity compensation for the robot end tool and the six-dimensional force sensor, including collecting the six-dimensional force sensor data of the robot in different postures and calculating the force offset F of the three coordinate axes of the sensor. x0 , F y0 , F z0 And the gravity F of the end fixture and sensor on the coordinate axis x1 , F y1 , F z1 And remove the influence of gravity in the measurement data.

8. The active and passive compliant fusion robot grinding and polishing method according to claim 2 is characterized in that: The contact force F of the grinding part at the one-dimensional force sensor end is obtained. 20 ,include: The data fitting method is used, F 20 =A*U+a; Among them, F 20 is the measured contact force, U is the voltage value of the sensor, a is the coefficient, and A is the sensor calibration fitting coefficient.

9. The active and passive compliant fusion robot grinding and polishing method according to claim 2 is characterized in that: The six-dimensional force sensor measures the contact force F 10 The contact force F is measured with a one-dimensional force sensor 20 To integrate, including The weights W of the six-dimensional force sensor and one-dimensional force sensor measurements are calculated using similarity theory. i (k): W i (k)=[W1(k),W2(k),...W i (k)] Wherein, k represents the time, i represents the number of sensors, and in the present invention, there are 2 sensors, so K=2, and the fusion calculation process expression is as follows:

Images