A non-contact magnetic force measurement robot end position system and method

Abstract

This invention discloses a non-contact magnetic force measurement system and method for robot end-effector pose measurement. The system includes a ring magnet A, a ring magnet B, a six-dimensional force sensor, and a data processing system. Ring magnet A is fixedly attached to the robot end-effector, and the six-dimensional force sensor is fixedly attached to ring magnet A. Ring magnet B is fixedly installed at a position opposite to the robot end-effector. The six-dimensional force sensor measures the magnetic force between ring magnets A and B. The data processing system includes a mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them, determining the pose of ring magnet B relative to the robot's base coordinate system. The magnetic force data between ring magnets A and B detected by the six-dimensional force sensor is input into the data processing system. The pose data of ring magnet A is obtained through the mathematical model and coordinate transformation. The pose data of the robot end-effector is further obtained from the pose data of ring magnet A. This invention is low-cost and easier to deploy.

Description

Technical Field

[0001] This invention relates to the field of robot end-effector pose measurement, and in particular to a system and method for measuring robot end-effector pose based on non-contact magnetic force. Background Technology

[0002] Currently, in many industrial robot applications such as milling, polishing, and stacking, it is necessary to measure the robot's end-effector pose in real time in order to ensure the control accuracy of the robot's end-effector pose.

[0003] Common methods for obtaining robot end-effector pose include vision measurement schemes based on industrial cameras and measurement schemes based on laser rangefinders. However, both currently face challenges such as high cost and difficult deployment, and optical measurement devices cannot be used in situations with obstructions.

[0004] In vision measurement solutions based on industrial cameras, lighting stability has the greatest impact on measurement accuracy in machine vision applications. Even a slight change in illumination can result in a difference of 1 to 2 pixels in the measurement results. Unstable lighting can affect the position of image acquisition edges. Therefore, in the design of machine vision systems, it is essential to minimize the influence of ambient light while ensuring the stability of the supporting active light source.

[0005] In measurement schemes based on laser rangefinders, the laser rangefinder is a highly precise device. However, due to the influence of various factors, its accuracy cannot reach an ideal state. These factors include laser wavelength, laser power, the sensor's own accuracy, ambient temperature, atmospheric pressure, and humidity. Even with adjustments to these factors, the accuracy of a laser rangefinder is still difficult to achieve at the millimeter level, which is insufficient for some high-precision measurements. Laser rangefinders are quite sensitive to environmental conditions; if exposed to strong light interference or other external environmental factors, the measurement results will be inaccurate. Furthermore, in environments with significant variations in ambient temperature, atmospheric pressure, and humidity, the performance of laser rangefinders will also be affected. Summary of the Invention

[0006] This invention provides a non-contact magnetic force measurement robot end-effector pose system and method to solve the technical problems existing in the prior art.

[0007] The technical solution adopted by this invention to solve the technical problems existing in the prior art is as follows:

[0008] A non-contact magnetic force measurement robot end-effector pose system includes a ring magnet A, a ring magnet B, a six-dimensional force sensor, and a data processing system. The six-dimensional force sensor is fixedly attached to the robot end-effector, and the ring magnet A is fixedly attached to the six-dimensional force sensor. The ring magnet B is fixedly installed at a position opposite to the robot end-effector. The six-dimensional force sensor is used to measure the magnetic force between the ring magnets A and B. The data processing system includes a mathematical model representing the relationship between the relative poses of the ring magnets A and B and the magnetic force between them, determining the pose of the ring magnet B relative to the robot's base coordinate system. The magnetic force data between the ring magnets A and B detected by the six-dimensional force sensor is input into the data processing system. The pose data of the ring magnet A is obtained through the mathematical model and coordinate transformation. The pose data of the robot end-effector is further obtained from the pose data of the ring magnet A.

[0009] Furthermore, the axis of the ring magnet A coincides with the axis of the robot's end effector.

[0010] Furthermore, the structure and magnetic properties of ring magnet A and ring magnet B are identical.

[0011] This invention also provides a method for non-contact magnetic force measurement of robot end-effector pose. The method includes: setting up a ring magnet A, a ring magnet B, a six-dimensional force sensor, and a data processing system; fixing the six-dimensional force sensor to the robot end-effector, fixing the ring magnet A to the six-dimensional force sensor, and fixing the ring magnet B at a position opposite to the robot end-effector; measuring the magnetic force between the ring magnets A and B using the six-dimensional force sensor; establishing a mathematical model within the data processing system representing the relationship between the relative poses of the ring magnets A and B and the magnetic force between them, and determining the pose of the ring magnet B relative to the robot's base coordinate system; inputting the magnetic force data between the ring magnets A and B detected by the six-dimensional force sensor into the data processing system, obtaining the pose data of the ring magnet A through the mathematical model and coordinate transformation, and further obtaining the pose data of the robot end-effector from the pose data of the ring magnet A.

[0012] Furthermore, let: the fixed coordinate system of the robot end effector be {E}; the force measurement coordinate system of the six-dimensional force sensor be {S}; the fixed coordinate system of the ring magnet A be {M}; and the fixed coordinate system of the ring magnet B be {M}. r}; This represents the relative pose matrix of the ring magnets A and B. Let A represent the pose matrix of the ring magnet A relative to the robot's end effector. Indicates the robot's end-effector pose. This represents the pose matrix of the six-dimensional force sensor's force measurement coordinate system relative to the ring magnet A. express The displacement vector, express The pose submatrix, Represents the pose matrix displacement vector The antisymmetric matrix, F m T represents the magnetic force between ring magnets A and B detected by a six-dimensional force sensor in coordinate system {M}. m F represents the torque detected by the six-dimensional force sensor in coordinate system {M}. s T represents the magnetic force between ring magnets A and B detected by a six-dimensional force sensor in coordinate system {S}. s Let the torque detected by the six-dimensional force sensor in coordinate system {S} be represented by:

[0013]

[0014]

[0015] Let the mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them be:

[0016]

[0017] Further obtain the robot's end-effector pose

[0018]

[0019] In the formula, G(F) m ,T m ) indicates about F m ,T m The function.

[0020] Furthermore, a method for establishing a mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them includes the following steps: First, construct a finite element simulation model of ring magnets A and B; then, adjust different magnet parameters within a certain relative pose space for simulation; next, perform curve fitting on the simulation results to obtain the optimal polynomial order, thereby obtaining a parametric linear polynomial representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them; use a laser tracker to collect data on the relative pose changes of ring magnets A and B, and simultaneously collect the corresponding magnetic force data between them when their relative poses change using a six-dimensional force sensor; substitute the collected magnetic force data into the obtained parametric linear polynomial for parameter regression calculation, and finally obtain the mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them.

[0021] Furthermore, the method for constructing finite element simulation models of toroidal magnets A and B includes the following steps:

[0022] Step A1: Using the power electronics simulation sub-software in ANSYS finite element simulation software, draw the models of ring magnet A and ring magnet B, define their relative motion range and motion step size, and mesh them.

[0023] Step A2: Determine the magnetic performance parameters and parameter value ranges of ring magnet A and ring magnet B, and divide the magnetic performance parameters into multiple groups;

[0024] Step A3: Take a set of magnet performance parameters, run the simulation, and output the different relative pose data of the ring magnets A and B and the corresponding magnetic force data between them.

[0025] Step A4: Import the different relative pose data of the ring magnets A and B and the corresponding magnetic force data between them into Matlab software, and use polynomial fitting to obtain the highest order of the polynomial with the best fitting effect.

[0026] Step A5: Update the magnet performance parameters. Repeat steps A3 and A4 until the simulation and polynomial fitting of all groups of magnet performance parameters are completed. Take the highest order of the polynomial that fits all groups of magnet performance parameters well as the best polynomial order.

[0027] Step A6, let the highest order of the optimal polynomial be n. The parametric linear polynomial representing the relationship between the relative poses of the ring magnets A and B and the magnetic force between them is as follows:

[0028] G(f,t)=A0+A1f+A2f 2 +…+A n-1 f n-1 +A n f n +B0+B1t+B2t 2 +…+B n-1 t n-1 +B n t n ;

[0029] Where: A0, A1, A2, ..., A n-1 A n Corresponding to f 0 f, f 2 ... f n-1 f n The coefficient matrix; B0, B1, B2, ..., B n-1 B n Corresponding to t 0 , t, t 2 、…、t n-1 t nThe coefficient matrix; f represents the magnetic force between the ring magnets A and B detected by the six-dimensional force sensor; t represents the torque detected by the six-dimensional force sensor; G(f,t) represents a function of f and t.

[0030] Furthermore, the performance parameters of a magnet include coercivity and remanence.

[0031] Furthermore, the method for collecting the corresponding magnetic force data between ring magnets A and B when their relative poses change, and for obtaining the mathematical model of the relationship between the relative poses of ring magnets A and B and the magnetic force between them, includes the following steps:

[0032] Step B1: Fix the six-dimensional force sensor and the ring magnet A to each other and install them on the robot end effector. Fix the ring magnet B at a position opposite to the robot end effector. Fix the laser tracker on the robot end effector. Based on the robot base coordinate system, calibrate the fixed coordinate system of the ring magnet B and the reference coordinate system of the laser tracker. Set them into the laser tracker operating software so that it can directly output the relative pose of the ring magnets A and B.

[0033] Step B2: Set the relative movement space between ring magnets A and B; move the robot end effector relative to ring magnet B within the movement space, and simultaneously collect the relative pose data of ring magnets A and B output by the laser tracker and the magnetic force data between ring magnets A and B output by the six-dimensional force sensor; compile them into a dataset.

[0034] Step B3: Input the dataset into Matlab. Based on the optimal polynomial order of the mathematical model of the relationship between the relative poses of the ring magnets A and B and the magnetic force between them, call the least squares regression function to obtain the coefficient matrix of the polynomial.

[0035] Furthermore, a method for establishing a mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them includes the following steps: using Biot-Saffar's law, surface integrals are performed on the cylindrical surface and the upper and lower base surfaces of ring magnets A and B to calculate the magnetic field generated by each of ring magnets A and B in space. Then, the magnetic field expression is substituted into the relative pose matrix of ring magnets A and B, and the expression is transformed into the robot's base coordinate system. The magnetic force generated by each surface of the cylindrical surface and the upper and lower base surfaces of ring magnet A or ring magnet B is calculated again by surface integral and added together to obtain the expression of the magnetic force with respect to the relative pose matrix. This expression is used to represent the relationship between the relative poses of ring magnets A and B and the magnetic force between them.

[0036] The advantages and positive effects of this invention are:

[0037] In situations where there are no magnetic materials or external magnetic fields within the robot's end effector's movement space, this invention offers lower cost and easier deployment compared to vision measurement solutions based on industrial cameras and measurement solutions based on laser rangefinders. It can be manufactured as a plug-and-play module that is standard with the robot's end effector to adapt to different needs. Attached Figure Description

[0038] Figure 1 This is a schematic diagram illustrating the structure and working principle of a robot end-effector pose system based on non-contact magnetic force measurement according to the present invention.

[0039] In the diagram: 1. Robot end effector; 2. Six-dimensional force sensor; 3. Ring magnet A; 4. Ring magnet B; E represents the origin of the fixed coordinate system of the robot end effector; S represents the origin of the force measurement coordinate system of the six-dimensional force sensor; M represents the origin of the fixed coordinate system of ring magnet A; M r The coordinate system origin of the fixed coordinate system is represented by ring magnet B; F represents the magnetic force between ring magnets A and B; F1 represents the magnetic force between ring magnets A and B when ring magnet A moves to position 1. n The diagram represents the magnetic force between ring magnets A and B when ring magnet A moves to position n. X, Y, and Z represent the coordinate axes. The dashed line diagram of the robot's end effector, the six-dimensional force sensor, and ring magnet A represents the different positions reached. Detailed Implementation

[0040] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0041] The Chinese definitions of the following English words, phrases, and abbreviations in this application are as follows:

[0042] ANSYS: ANSYS software is a large-scale general-purpose finite element analysis (FEA) software developed by ANSYS Inc. in the United States. It is the fastest growing computer-aided engineering (CAE) software in the world and can interface with most computer-aided design (CAD) software to achieve data sharing and exchange.

[0043] Ansys Electronics Desktop: A simulation modeling software for creating complex electronic components and systems, simulating and analyzing various electromagnetic fields, including high-frequency electromagnetic fields, radio frequency electromagnetic fields, and microwave electromagnetic fields. Through electromagnetic field simulation analysis, it can help users predict and solve various electromagnetic interference problems.

[0044] Mathworks: A software developer and supplier that provides mathematical computation and model-based design for engineers and scientists, headquartered in Natick, Massachusetts, USA.

[0045] Matlab: Commercial mathematical software produced by MathWorks, Inc. in the United States, used in fields such as data analysis, wireless communication, deep learning, image processing and computer vision, signal processing, quantitative finance and risk management, robotics, and control systems.

[0046] Maxwell: Maxwell's low-frequency electromagnetic field simulation software, used for designing and analyzing electric motors, actuators, sensors, transformers, and other electromagnetic and electromechanical devices.

[0047] Please see Figure 1 A non-contact magnetic force measurement robot end-effector pose system includes a ring magnet A3, a ring magnet B4, a six-dimensional force sensor 2, and a data processing system. The six-dimensional force sensor 2 is fixedly attached to the robot end-effector 1, and the ring magnet A3 is fixedly attached to the six-dimensional force sensor 2. The ring magnet B4 is fixedly installed at a position opposite to the robot end-effector 1. The six-dimensional force sensor 2 is used to measure the magnetic force between the ring magnets A3 and B4. The data processing system includes a mathematical model representing the relationship between the relative poses of the ring magnets A3 and B4 and the magnetic force between them, determining the pose of the ring magnet B4 relative to the robot's base coordinate system. The magnetic force data between the ring magnets A3 and B4 detected by the six-dimensional force sensor 2 is input into the data processing system. The pose data of the ring magnet A3 is obtained through the mathematical model and coordinate transformation. The pose data of the robot end-effector 1 is further obtained from the pose data of the ring magnet A3.

[0048] The data processing system may include a processor and may employ existing mathematical modeling methods to construct a mathematical model of the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them, such as analytical methods, data fitting, parameter estimation, and neural networks. Existing software may also be used to construct the mathematical model.

[0049] A six-dimensional force sensor is a device capable of simultaneously measuring force and torque in a Cartesian coordinate system and converting them into electrical signals. It can simultaneously measure force and torque in six directions and output the external force information as a voltage signal through a special conversion and adjustment device. It boasts advantages such as high accuracy, fast frequency response, high reliability, and long lifespan. As an important device for acquiring external information, the six-dimensional force sensor can be applied in manufacturing technology, aerospace, and biomedicine.

[0050] Ring magnet A3, ring magnet B4, and the six-dimensional force sensor can all be ring cylinders.

[0051] Preferably, the axis of the annular magnet A3 can coincide with the axis of the robot end effector 1.

[0052] Preferably, the annular magnet A3 and the annular magnet B4 have the same structure and magnetic properties. The size of the toroidal surface of the six-dimensional force sensor can be greater than or equal to the size of the toroidal surface of the annular magnet A3, so that the toroidal surface of the annular magnet A3 is located within the toroidal surface of the six-dimensional force sensor.

[0053] This invention also provides a method for non-contact magnetic force measurement of robot end-effector pose. The method comprises: setting up a ring magnet A3, a ring magnet B4, a six-dimensional force sensor 2, and a data processing system; fixing the ring magnet A3 to the robot end-effector 1, fixing the six-dimensional force sensor 2 to the ring magnet A3, and fixing the ring magnet B4 at a position opposite to the robot end-effector 1; using the six-dimensional force sensor 2 to measure the magnetic force between the ring magnet A3 and the ring magnet B4; establishing a mathematical model within the data processing system representing the relationship between the relative poses of the ring magnets A3 and B4 and the magnetic force between them, and determining the pose of the ring magnet B4 relative to the robot's base coordinate system; inputting the magnetic force data between the ring magnets A3 and B4 detected by the six-dimensional force sensor 2 into the data processing system, obtaining the pose data of the ring magnet A3 through the mathematical model and coordinate transformation, and further obtaining the pose data of the robot end-effector 1 from the pose data of the ring magnet A3.

[0054] Preferably, the following coordinate systems can be defined: the fixed coordinate system of the actuator of the robot end effector 1 is {E}; the force measurement coordinate system of the six-dimensional force sensor 2 is {S}; the fixed coordinate system of the annular magnet A3 is {M}; and the fixed coordinate system of the annular magnet B4 is {M}. r}; This represents the relative pose matrix of ring magnets A3 and B4. This represents the pose matrix of the ring magnet A3 relative to the robot's end effector 1. This indicates the pose of the robot's end effector at position 1. This represents the pose matrix of the six-dimensional force sensor 2 relative to the ring magnet A3 in the force measurement coordinate system. express The displacement vector, express The pose submatrix, Represents the pose matrix displacement vector The antisymmetric matrix, F m T represents the magnetic force between ring magnets A3 and B4 detected by the six-dimensional force sensor 2 in coordinate system {M}. m F represents the torque detected by the six-dimensional force sensor 2 in coordinate system {M}. s T represents the magnetic force between ring magnets A3 and B4 detected by the six-dimensional force sensor 2 in coordinate system {S}. s Let the torque detected by the six-dimensional force sensor 2 in coordinate system {S} represent the torque. Then we have:

[0055]

[0056]

[0057] Let the mathematical model representing the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them be:

[0058]

[0059] The robot's end-effector pose can then be obtained.

[0060]

[0061] In the formula, G(F) m ,T m ) indicates about F m ,T m A function. G(X,Y) represents a function of X and Y.

[0062] Preferably, the method for establishing a mathematical model representing the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them may include the following steps: First, construct a finite element simulation model of ring magnets A3 and B4; then, adjust different magnet parameters within a certain relative pose space for simulation; then, perform curve fitting on the simulation results to obtain the optimal polynomial order, and further obtain a parametric linear polynomial representing the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them; use a laser tracker to collect data on the relative pose changes of ring magnets A3 and B4, and simultaneously collect the corresponding magnetic force data between them when their relative poses change through a six-dimensional force sensor 2; substitute the collected magnetic force data into the obtained parametric linear polynomial, perform parameter regression calculation, and finally obtain the mathematical model representing the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them.

[0063] Preferably, the method for constructing finite element simulation models of ring magnet A3 and ring magnet B4 may include the following steps:

[0064] Step A1: Using the power electronics simulation sub-software in ANSYS finite element simulation software, draw the models of ring magnet A3 and ring magnet B4, define their relative motion range and motion step size, and generate meshes.

[0065] Step A2: Determine the magnetic performance parameters and parameter value ranges of ring magnet A3 and ring magnet B4, and divide the magnetic performance parameters into multiple groups.

[0066] Step A3: Take a set of magnet performance parameters, run the simulation, and output the different relative pose data of ring magnet A3 and ring magnet B4 and the corresponding magnetic force data between them.

[0067] Step A4: Import the different relative pose data of ring magnets A3 and B4 and the corresponding magnetic force data between them into Matlab software, and use polynomial fitting to obtain the highest order of the polynomial with the best fitting effect.

[0068] Step A5: Update the magnet performance parameters. Repeat steps A3 and A4 until the simulation and polynomial fitting of all groups of magnet performance parameters are completed. Take the highest order of the polynomial that can fit all groups of magnet performance parameters well as the optimal polynomial order.

[0069] Step A6, let the highest order of the optimal polynomial be n. The parametric linear polynomial representing the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them is as follows:

[0070] G(f,t)=A0+A1f+A2f 2 +…+A n-1 f n-1 +A n f n +B0+B1t+B2t 2 +…+B n-1 t n-1 +B n t n .

[0071] Where: A0, A1, A2, ..., A n-1 A n Corresponding to f 0 f, f 2 ... f n-1 f n The coefficient matrix; B0, B1, B2, ..., B n-1 B n Corresponding to t 0 , t, t 2 、…、t n-1 t n The coefficient matrix; f represents the magnetic force between the ring magnets A3 and B4 detected by the six-dimensional force sensor 2; t represents the torque detected by the six-dimensional force sensor 2; G() represents the mathematical model of the relationship between the relative pose of the ring magnets A3 and B4 and the magnetic force between them.

[0072] A parameterized linear polynomial is a polynomial in which all or some of its coefficients are unknowns.

[0073] If all or some of the coefficients of a polynomial are unknowns, we can use the principle that when two polynomials are identities, the coefficients of like terms are equal to determine these coefficients, and thus obtain the value to be found.

[0074] Preferably, the magnet performance parameters may include coercivity and remanence.

[0075] Preferably, the method for collecting the magnetic force data between ring magnets A and B when their relative poses change, and for obtaining a mathematical model of the relationship between the relative poses of ring magnets A and B and their magnetic force, may include the following steps:

[0076] Step B1: Fix the six-dimensional force sensor 2 and the ring magnet A3 together and install them on the robot end effector 1. Fix the ring magnet B4 at a position opposite to the robot end effector 1. Fix a laser tracker on the robot end effector 1. Based on the robot base coordinate system, calibrate the fixed coordinate system of the ring magnet B4 and the reference coordinate system of the laser tracker. Set them into the laser tracker operating software so that it can directly output the relative pose of the ring magnet A3 and the ring magnet B4.

[0077] The laser tracker operating software can be the existing laser tracker operating software, which directly converts the contour detection data of the object under test into the pose data of the object under test.

[0078] Step B2: Set the relative movement space between ring magnet A3 and ring magnet B4; make the robot end effector 1 move relative to ring magnet B4 within the movement space, and simultaneously collect the relative pose data of ring magnet A3 and ring magnet B4 output by the laser tracker and the magnetic force data between ring magnet A3 and ring magnet B4 output by the six-dimensional force sensor 2; compile them into a dataset.

[0079] Step B3: Input the dataset into Matlab. Based on the optimal polynomial order of the mathematical model of the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them, call the least squares regression function to obtain the coefficient matrix of the polynomial.

[0080] Preferably, the method for establishing a mathematical model representing the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them may include the following steps: using Biot-Saffar's law, performing surface integrals on the cylindrical surface and the upper and lower base surfaces of ring magnets A3 and B4 to calculate the magnetic field generated by each of ring magnets A3 and B4 in space; then substituting this into the relative pose matrix of ring magnets A3 and B4, transforming the magnetic field expression into the robot's base coordinate system; again calculating the magnetic force generated by each surface of the cylindrical surface and the upper and lower base surfaces of ring magnets A3 or B4 through surface integrals and adding them together to obtain an expression for the magnetic force with respect to the relative pose matrix; and using this expression to represent the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them.

[0081] The working principle of the present invention will be further explained below with reference to a preferred embodiment:

[0082] Please see Figure 1 A non-contact magnetic force measurement robot end-effector pose system includes a ring magnet A3, a ring magnet B4, a six-dimensional force sensor 2, and a data processing system. The ring magnet A3 is fixedly attached to the robot end-effector 1, and the six-dimensional force sensor 2 is fixedly attached to the ring magnet A3. The ring magnet B4 is fixedly installed at a position opposite to the robot end-effector 1. The six-dimensional force sensor 2 is used to measure the magnetic force between the ring magnets A3 and B4. The data processing system includes a mathematical model representing the relationship between the relative poses of the ring magnets A3 and B4 and the magnetic force between them, determining the pose of the ring magnet B4 relative to the robot's base coordinate system. The magnetic force data between the ring magnets A3 and B4 detected by the six-dimensional force sensor 2 is input into the data processing system. The pose data of the ring magnet A3 is obtained through the mathematical model and coordinate transformation, and the pose data of the robot end-effector 1 is further obtained from the pose data of the ring magnet A3.

[0083] Ring magnet A3, ring magnet B4, and the six-dimensional force sensor are all ring cylinders.

[0084] Ring magnet A3 and ring magnet B4 have the same structure and magnetic properties.

[0085] The toroidal surface of the six-dimensional force sensor is larger than that of the toroidal surface of the ring magnet A3, so that the toroidal surface of the ring magnet A3 is located inside the toroidal surface of the six-dimensional force sensor. The toroidal surface of the six-dimensional force sensor is 1.1 to 1.5 times the size of the toroidal surface of the ring magnet A3.

[0086] The axes of the six-dimensional force sensor 2, the ring magnet A3, and the robot end effector 1 are aligned.

[0087] The above-mentioned non-contact magnetic force measurement robot end-effector pose method, which employs the aforementioned non-contact magnetic force measurement robot end-effector pose system, specifically includes the following steps:

[0088] S1: Using the power electronics simulation sub-software Ansys ElectronicsDesktop (formerly Maxwell electromagnetic simulation software) of the finite element simulation software ANSYS2023R1, draw the models of the ring magnet A3 and ring magnet B4, define their relative motion range and motion step size H, generate the mesh and make other necessary settings.

[0089] S2: Within the range of magnetic performance parameters corresponding to the grade of magnet used, take several sets of magnetic performance parameters at a certain step size h. The magnetic performance parameters include coercivity H. c,i and remanence B r,i Where i = 1, 2, ... n are parameter groups;

[0090] S3.: Take a set of magnetic performance parameters, run the simulation, and output the different relative pose data of ring magnet A3 and ring magnet B4 and the corresponding magnetic force data between them;

[0091] S4: Import the magnetic force data tables under different relative poses into the Matlab mathematical software of Mathworks, USA, and use polynomial fitting to find the lowest order that provides the best fitting effect.

[0092] S5: Return to S3, take a new set of parameters, and repeat the simulation and fitting process of S3 and S4 until all sets of parameters are fitted. Find the polynomial order that can be accurately fitted for all parameters, and the best polynomial order is 4.

[0093] S6: This yields a parametric fourth-order polynomial model of magnetic force to relative pose in the defined relative pose space, namely G(f,t)=A0+A1f+A2f 2 +A3f 3 +A4f 4 +B0+B1t+B2t 2 +B3t 3 +B4t 4 ;

[0094] Where: A0, A1, A2, A3, A n Corresponding to f 0 f, f 2 f 3 f 4 The coefficient matrix; B0, B1, B2, B3, B4 correspond to t 0 , t, t 2 t 3 t4 The coefficient matrix; f represents the magnetic force between the ring magnet A3 and the ring magnet B4 detected by the six-dimensional force sensor 2; t represents the torque detected by the six-dimensional force sensor 2; G(f,t) represents a function of f and t.

[0095] The relative pose of ring magnets A3 and B4 can be adjusted according to different magnetic property parameters, or different magnetic property parameters of ring magnets A3 and B4 can be selected based on their relative pose. The above method can be used for optimization and iteration to construct a mathematical model of the relationship between the relative pose of ring magnets A and B and the magnetic force between them.

[0096] S7: Install the six-dimensional force sensor 2 and the ring magnet A3 onto the end effector 1 of a robot, and fix the ring magnet B4 to the worktable. Prepare the laser tracker assembly, which may include the laser tracker body, a magnetic target ball mount, and a target ball for fixing the laser tracker body. Attach the magnetic target ball mount of the laser tracker to the end effector 1 of the robot, and attach the target ball for fixing the laser tracker body to it. Calibrate the fixed coordinate system of the ring magnet B4 and the reference coordinate system of the laser tracker; set them into the laser tracker operating software so that it can directly output the relative pose of the ring magnet A3 and the ring magnet B4.

[0097] S8: Corresponding to the simulated relative pose movement range, move the robot to the set relative pose space, write the robot motion program to cover as many points in the relative pose space as possible, and simultaneously collect the relative pose data of ring magnet A3 and ring magnet B4 output by the laser tracker and the magnetic force data between ring magnet A3 and ring magnet B4 output by the six-dimensional force sensor 2; obtain 2000 sets of data.

[0098] S9: Input 2000 sets of data into the mathematical software Matlab. Since the optimal order of the polynomial is known, directly call the least squares regression function `polyfit` to obtain the actual parameter matrix, i.e., A0, A1, A2, A3, A... n B0, B1, B2, B3, B4. The resulting A0, A1, A2, A3, A n Substituting B0, B1, B2, B3, and B4 into the polynomial model G(f,t) = A0 + A1f + A2f in S6 2 +A3f 3 +A4f 4 +B0+B1t+B2t 2 +B3t 3 +B4t 4 A mathematical model was established to show the relationship between the relative poses of ring magnets A3 and B4 and the magnetic force between them.

[0099] Practical application has verified that this method can measure the relative pose of the robot end effector 1 and the ring magnet B4 in real time within a certain spatial range.

[0100] The aforementioned components, including ring magnet A3, ring magnet B4, six-dimensional force sensor 2, data processing system, laser tracker, or laser tracker components, can all utilize existing components, devices, and systems, or can be constructed using existing components, devices, and systems and conventional technical means.

[0101] The embodiments described above are only used to illustrate the technical ideas and features of the present invention. Their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The patent scope of the present invention should not be limited by these embodiments. That is, any equivalent changes or modifications made in accordance with the spirit disclosed in the present invention still fall within the patent scope of the present invention.

Claims

1. A method for measuring a pose of a robot end based on non-contact magnetic force, characterized in that, The method is as follows: A ring magnet A, a ring magnet B, a six-dimensional force sensor, and a data processing system are set up; the six-dimensional force sensor is fixedly attached to the robot's end effector, and the ring magnet A is fixedly attached to the six-dimensional force sensor; the ring magnet B is fixedly installed at a position opposite to the robot's end effector; the magnetic force between the ring magnets A and B is measured using the six-dimensional force sensor; a mathematical model representing the relationship between the relative poses of the ring magnets A and B and the magnetic force between them is established within the data processing system to determine the pose of the ring magnet B relative to the robot's base coordinate system; the magnetic force data between the ring magnets A and B detected by the six-dimensional force sensor is input into the data processing system, and the pose data of the ring magnet A is obtained through the mathematical model and coordinate transformation; the pose data of the robot's end effector is further obtained from the pose data of the ring magnet A. A method for establishing a mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them includes the following steps: First, construct a finite element simulation model of ring magnets A and B; then, perform simulations by adjusting different magnet parameters within a certain relative pose space; next, perform curve fitting on the simulation results to obtain the optimal polynomial order, and further obtain a parametric linear polynomial representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them; use a laser tracker to collect data on the relative pose changes of ring magnets A and B, and simultaneously collect the corresponding magnetic force data between them when their relative poses change using a six-dimensional force sensor; substitute the collected magnetic force data into the obtained parametric linear polynomial, perform parameter regression calculations, and finally obtain the mathematical model of the relationship between the relative poses of ring magnets A and B and the magnetic force between them; The method may include the following steps: using Biot-Saffar's law, perform surface integration on the cylindrical surface and the top and bottom surfaces of ring magnets A and B to calculate the magnetic field generated by each of ring magnets A and B in space. Then, substitute the magnetic field expression into the relative pose matrix of ring magnets A and B, transform the magnetic field expression into the robot's base coordinate system, and again calculate the magnetic force generated by each surface of the cylindrical surface and the top and bottom surfaces of ring magnet A or ring magnet B through surface integration and add them together to obtain the expression of the magnetic force with respect to the relative pose matrix. Use this expression to represent the relationship between the relative pose of ring magnets A and B and the magnetic force between them.

2. The method for measuring robot end-effector pose based on non-contact magnetic force according to claim 1, characterized in that, Let the fixed coordinate system of the robot's end effector be: The force measurement coordinate system of the six-dimensional force sensor is... The fixed coordinate system of the ring magnet A is: The fixed coordinate system of the ring magnet B is: ; This represents the relative pose matrix of the ring magnets A and B. Let A represent the pose matrix of the ring magnet A relative to the robot's end effector. Indicates the pose of the robot's end effector. This represents the pose matrix of the six-dimensional force sensor's force measurement coordinate system relative to the ring magnet A. The displacement vector, express The pose submatrix, Represents the pose matrix displacement vector antisymmetric matrix, In coordinate system The magnetic force between ring magnets A and B detected by the six-dimensional force sensor. In coordinate system Torque detected by the lower six-dimensional force sensor In coordinate system The magnetic force between ring magnets A and B detected by the six-dimensional force sensor. In coordinate system The torque detected by the lower six-dimensional force sensor is as follows: ； ； Let the mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them be: ； Further obtain the robot's end-effector pose : ； In the formula, Indicates about The function.

3. The method for measuring robot end-effector pose based on non-contact magnetic force according to claim 1, characterized in that, The method for constructing finite element simulation models of toroidal magnets A and B includes the following steps: Step A1: Using the power electronics simulation sub-software in ANSYS finite element simulation software, draw the models of ring magnet A and ring magnet B, define their relative motion range and motion step size, and mesh them. Step A2: Determine the magnetic performance parameters and parameter value ranges of ring magnet A and ring magnet B, and divide the magnetic performance parameters into multiple groups; Step A3: Take a set of magnet performance parameters, run the simulation, and output the different relative pose data of the ring magnets A and B and the corresponding magnetic force data between them. Step A4: Import the different relative pose data of the ring magnets A and B and the corresponding magnetic force data between them into Matlab software, and use polynomial fitting to obtain the highest order of the polynomial with the best fitting effect. Step A5: Update the magnet performance parameters. Repeat steps A3 and A4 until the simulation and polynomial fitting of all groups of magnet performance parameters are completed. Take the highest order of the polynomial that fits all groups of magnet performance parameters well as the best polynomial order. Step A6, let the highest order of the optimal polynomial be n. The parametric linear polynomial representing the relationship between the relative poses of the ring magnets A and B and the magnetic force between them is as follows: ； In the formula: , Corresponding to , , , The coefficient matrix; , Corresponding to , , , The coefficient matrix; The magnetic force between ring magnets A and B detected by a six-dimensional force sensor; Torque detected by a six-dimensional force sensor; Indicates about The function.

4. The method for measuring robot end-effector pose based on non-contact magnetic force according to claim 3, characterized in that, The performance parameters of a magnet include coercivity and remanence.

5. The method for measuring robot end-effector pose based on non-contact magnetic force according to claim 1, characterized in that, The method for collecting magnetic force data between ring magnets A and B as their relative poses change, and for obtaining a mathematical model of the relationship between the relative poses of ring magnets A and B and their magnetic force, includes the following steps: Step B1: Fix the six-dimensional force sensor and the ring magnet A to each other and install them on the robot end effector. Fix the ring magnet B at a position opposite to the robot end effector. Fix the laser tracker on the robot end effector. Based on the robot base coordinate system, calibrate the fixed coordinate system of the ring magnet B and the reference coordinate system of the laser tracker. Set them into the laser tracker operating software so that it can directly output the relative pose of the ring magnets A and B. Step B2: Set the relative movement space between ring magnets A and B; move the robot end effector relative to ring magnet B within the movement space, and simultaneously collect the relative pose data of ring magnets A and B output by the laser tracker and the magnetic force data between ring magnets A and B output by the six-dimensional force sensor; compile them into a dataset. Step B3: Input the dataset into Matlab. Based on the optimal polynomial order of the mathematical model of the relationship between the relative poses of the ring magnets A and B and the magnetic force between them, call the least squares regression function to obtain the coefficient matrix of the polynomial.

6. A robot end-effector pose system based on the non-contact magnetic force measurement robot end-effector pose method according to any one of claims 1 to 5, characterized in that, The system includes a ring magnet A, a ring magnet B, a six-dimensional force sensor, and a data processing system. A six-dimensional force sensor is fixed to the robot's end effector, and ring magnet A is fixed to the six-dimensional force sensor. Ring magnet B is fixedly installed at a position opposite to the robot's end effector. The six-dimensional force sensor measures the magnetic force between ring magnets A and B. The data processing system contains a mathematical model representing the relationship between the relative poses of ring magnets A and B and the magnetic force between them, determining the pose of ring magnet B relative to the robot's base coordinate system. The magnetic force data between ring magnets A and B detected by the six-dimensional force sensor is input into the data processing system. Through the mathematical model and coordinate transformation, the pose data of ring magnet A is obtained, and the pose data of the robot's end effector is further obtained from the pose data of ring magnet A.

7. The robot end-effector pose measurement system based on non-contact magnetic force measurement according to claim 6, characterized in that, The axis of the ring magnet A coincides with the axis of the robot's end effector.

8. The robot end-effector pose measurement system based on non-contact magnetic force measurement according to claim 6, characterized in that, Ring magnet A and ring magnet B have the same structure and magnetic properties.

Images