Planar motor and mover position detection method

Abstract

The application discloses a planar motor, wherein a first sensor group and a second sensor group are arranged on a stator assembly of the planar motor to jointly cover a working plane of the planar motor, and the second sensor group is arranged in a staggered manner with a first offset in a first direction and a second offset in a second direction relative to the first sensor group. When detection is performed, a processing unit acquires first sensing information of the first sensor group and second sensing information of the second sensor group, and constructs a magnetic field distribution map in a unified space coordinate system according to the first sensing information, the second sensing information and a set of offset coordinates. Then, the center position coordinates of a mover assembly are calculated according to a preset magnetic field distribution-position mapping relationship and the magnetic field distribution map. Therefore, the number of sensors can be significantly reduced under the premise of ensuring the position detection accuracy, and the hardware requirements and costs are reduced, and meanwhile, the difficulty of data acquisition and processing is reduced, and the real-time control performance is improved. The application further discloses a mover position detection method based on the planar motor.

Description

Technical Field

[0001] This invention relates to the field of planar motor position detection technology, and in particular to a planar motor and a method for detecting the position of the mover based on a double-set staggered Hall sensor array. Background Technology

[0002] In a planar magnetic levitation motor system, it is necessary to detect the three-dimensional position information of the mover within the stator plane region in real time (position in the X and Y planes and levitation height in the Z direction). Existing technologies typically use a 3D Hall sensor array to sample the spatial magnetic field generated by the permanent magnet array at the bottom of the mover, and then calculate the mover's three-dimensional position coordinates through the correspondence between magnetic field strength and position. Therefore, the sensor density directly affects the accuracy and resolution of position detection. To obtain high-precision position information, traditional methods typically arrange sensors uniformly at intervals of λ / 2 (the half-cycle length of the magnetic field) to ensure effective sampling of magnetic field changes at any position. However, this approach has the following shortcomings: (1) Too many sensors; taking a stator size of 240mm×240mm and a magnetic field period of λ=40mm as an example, if the sensors are evenly arranged in the X and Y directions on the stator surface with an interval of λ / 2=20mm, the entire array requires 13×13=169 sensors. Figure 1 As shown, this results in higher hardware costs, more complex wiring, and higher power consumption. (2) Heavy signal processing burden; the acquisition and processing of data from a large number of sensors requires higher hardware resources and longer processing time, which affects real-time control performance; (3) PCB layout is difficult; integrating 169 sensors and their signal traces on a multi-layer PCB stator board places high demands on the area and stack-up structure of the PCB stator board, which may conflict with the wiring of the motor coil, making the PCB layout extremely difficult. Summary of the Invention

[0003] The technical solution of this invention is as follows: A planar motor is provided, comprising a stator assembly, a mover assembly, a position detection assembly, and a processing unit; the stator assembly includes multiple stator units, which are spliced ​​together to form a working plane; each stator unit includes a first coil group and a second coil group, the arrangement directions of the first coil group and the second coil group being staggered; the mover assembly can move along the working plane under the drive of the stator units; the mover assembly includes a first magnet array and a second magnet array, the array directions of the first magnet array and the second magnet array being staggered; the position detection assembly includes a first sensor group and a second sensor group disposed on the stator assembly, the first sensor group including multiple first sensors, and the multiple... The first sensor group comprises a plurality of second sensors arranged in an array along the first direction and the second direction at a first spacing; the second sensor group comprises a plurality of second sensors arranged in an array along the first direction and the second direction at a second spacing; the second sensor group is staggered relative to the first sensor group in the first direction by a first offset, and the second sensor group is staggered relative to the first sensor group in the second direction by a second offset, the first sensor group and the second sensor group together covering the working plane of the stator assembly; wherein, the first sensor group is used to detect the magnetic field strength of the mover assembly to obtain first sensing information, and the second sensor group is used to detect the magnetic field strength of the mover assembly to obtain second sensing information. The processing unit is communicatively connected to the first sensor group and the second sensor group, and the processing unit is used for: Acquire the first sensing information of the first sensor group and the second sensing information of the second sensor group; Based on the first sensing information, the second sensing information, and the offset coordinate set, a magnetic field distribution map in a unified spatial coordinate system is constructed; wherein, the offset coordinate set is the physical coordinate information of all the first sensors and all the second sensors in a unified spatial coordinate system constructed based on the first offset and the second offset; Based on the magnetic field distribution map and the preset magnetic field distribution-position mapping relationship, the center position coordinates of the moving component are calculated.

[0004] Preferably, the first spacing and the second spacing are equal and both are λ, and the first offset and the second offset are equal and both are λ / 2; where λ=L / N, λ is the magnetic field period length, L is the physical length of the first magnet array and the second magnet array, and N is the number of magnetic field periods contained in the first magnet array and the second magnet array.

[0005] Preferably, the total number of the first sensors is N.11 *N 12 N 11 = L1 / λ+1,N 12= L2 / λ+1, the total number of the second sensors is N 21 *N 22 N 21 = L1 / λ,N 12= L2 / λ; where N 11 N represents the number of the first sensors in the first direction. 12 N represents the number of the first sensors in the second direction. 21 N represents the number of the second sensors in the first direction. 22 L1 represents the number of the second sensors in the second direction, L2 represents the first length of the stator assembly in the first direction, and L3 represents the second length of the stator assembly in the second direction.

[0006] Preferably, after "constructing a magnetic field distribution map in a unified spatial coordinate system based on the first sensing information, the second sensing information, and the offset coordinate set," the processing unit is further configured to: Define a high-density grid map that is completely independent of the magnetic field distribution map; wherein the high-density grid map and the deviation coordinate set share the same coordinate system, and the grid spacing of the high-density grid map is much smaller than the first offset and the second offset; Spatial interpolation is performed on each grid point of the high-density grid to construct a high-resolution magnetic field distribution grid map.

[0007] Preferably, the spatial interpolation includes bilinear interpolation or bicubic interpolation; wherein, the bilinear interpolation takes the four nearest original sampling points around each grid point in the high-density grid map and performs step-by-step linear interpolation, and the bicubic interpolation takes the sixteen original sampling points around each grid point in the high-density grid map and performs cubic polynomial surface fitting.

[0008] Preferably, the preset magnetic field distribution-position mapping relationship includes a Halbach array magnetic field analytical model or a pre-calibrated magnetic field-position lookup table; wherein, the Halbach array magnetic field analytical model is used to substitute the actually detected magnetic field vectors in the magnetic field distribution map into the analytical equation, and calculate the center position coordinates of the moving component through iteration or numerical solution; the pre-calibrated magnetic field-position lookup table is used by the processing unit to look up the magnetic field vectors actually detected in the magnetic field distribution map to obtain the center position coordinates of the moving component; wherein, the magnetic field-position lookup table is established by offline control of the moving component to traverse all positions within the entire stroke and recording the magnetic field data corresponding to the coordinates of each moving component.

[0009] Preferably, the processing unit is further configured to: perform a fusion process on the coordinates of the center position of the mover component calculated by each of the multiple grid points in the magnetic field distribution map, the fusion process including weighted averaging or Kalman filtering, to output the final coordinates of the center position of the mover component.

[0010] Preferably, in the weighted average, the weights are assigned according to the signal-to-noise ratio and magnetic field gradient of each grid point, and then the summation is normalized. Grid points that are close to the first sensor or the second sensor and have a large magnetic field gradient and a high signal-to-noise ratio are assigned high weights, while grid points at the edge of the stator assembly, with weak magnetic fields, and with high noise are assigned low weights.

[0011] Preferably, in the Kalman filter, the center position coordinates of the stator assembly calculated for each grid point are used as observations to establish a motion state model of the stator assembly, and all observations are fused through a prediction-correction loop to output smooth center position coordinates.

[0012] Preferably, the first sensor group is arranged on the surface of the stator assembly, and the second sensor group is arranged inside the stator assembly, with a height difference between the two in a third direction of [missing information]. Furthermore, the projections of both directions onto the plane perpendicular to the third direction are offset by a distance of λ / 2 along both the first and second directions; wherein the first direction, the second direction, and the third direction are mutually perpendicular; the processing unit is further configured to: Spatial interpolation is performed on the first sensing information and the second sensing information respectively to construct the lower-level magnetic field distribution map under the unified spatial coordinate system. and upper magnetic field distribution map Among them, the lower magnetic field distribution map and upper magnetic field distribution map It has vertically paired grid points; The lower magnetic field distribution map is calculated according to the following formula (1). and the upper magnetic field distribution map The magnetic field ratio of all vertically paired grid points; (1) According to the magnetic field ratio Determine whether the moving part component is tilted; If the moving part is tilted, the attitude of the moving part is calculated and corrected. If the moving component is not tilted, the local suspension height H of the moving component corresponding to each grid point is calculated in reverse according to the formula (1); The average suspension height H of all local suspension heights is obtained by averaging all the local suspension heights H. .

[0013] Preferably, when "calculating the attitude of the moving part", the processing unit is further configured to: The magnetic field ratio of all grid points Construct as a two-dimensional matrix ; Calculate the gradient Gx in the X direction and the gradient Gy in the Y direction respectively, wherein the gradient Gx in the X direction corresponds to the pitch angle of the mover component in the first direction, and the gradient Gy in the Y direction corresponds to the roll angle of the mover component in the second direction. Calculate the mean of the gradient Gx in the X direction for all grid points. The mean of the gradient Gy in the Y direction ; Substitute all the mean values ​​into a preset gradient-tilt calibration model to calculate the pitch and roll angles of the mover assembly. The gradient-tilt calibration model is established by offline control of the mover assembly to perform pure pitch and pure roll standard motions at a fixed overall height. The gradient-tilt calibration model includes the mapping relationship between different pitch angles and the mean value of the gradient in the X direction of the entire field, and the mapping relationship between different roll angles and the mean value of the gradient in the Y direction of the entire field. Based on the center position coordinates of the moving component and the average suspension height The pitch angle (Pitch) and roll angle (Roll) are calculated, and the position and attitude of the moving part are output.

[0014] Correspondingly, the present invention also provides a method for detecting the position of a mover based on a planar motor as described above, which includes the following steps: S1. Obtain the magnetic field strength of the moving part detected by the first sensor group as the first sensing information, and obtain the magnetic field strength of the moving part detected by the second sensor group as the second sensing information. S2. Based on the first sensing information, the second sensing information, and the offset coordinate set, construct a magnetic field distribution map in a unified spatial coordinate system; wherein, the offset coordinate set is the physical coordinate information of all the first sensors and all the second sensors in a unified spatial coordinate system constructed based on the first offset and the second offset; S3. Based on the magnetic field distribution map and the preset magnetic field distribution-position mapping relationship, calculate the center position coordinates of the moving component.

[0015] Preferably, in the mover position detection method, after step S2, the following steps are further included: Define a high-density grid map that is completely independent of the magnetic field distribution map; wherein the high-density grid map and the deviation coordinate set share the same coordinate system, and the grid spacing of the high-density grid map is much smaller than the first offset and the second offset; Spatial interpolation is performed on each grid point of the high-density grid to construct a high-resolution magnetic field distribution grid map.

[0016] Preferably, in the mover position detection method, the spatial interpolation includes: Bilinear interpolation, wherein the bilinear interpolation is performed by taking the four nearest original sampling points around each grid point in the high-density grid map and performing step-by-step linear interpolation; Alternatively, bicubic interpolation can be used, where bicubic interpolation takes sixteen original sampling points around each grid point in the high-density grid map and performs cubic polynomial surface fitting.

[0017] Preferably, in the mover position detection method, in step S3, the preset magnetic field distribution-position mapping relationship includes: The Halbach array magnetic field analytical model is used to substitute the actual detected magnetic field vector in the magnetic field distribution map into the analytical equation, and calculate the center position coordinates of the moving component in reverse through iterative or numerical solution. Alternatively, a pre-calibrated magnetic field-position lookup table can be used, with the actual detected magnetic field vector in the magnetic field distribution map as an index, to query the magnetic field-position lookup table to obtain the center position coordinates of the moving part; wherein, the magnetic field-position lookup table is established by offline control of the moving part to traverse all positions within the entire stroke and recording the stator magnetic field data corresponding to each coordinate of the moving part.

[0018] Preferably, the mover position detection method further includes the following steps: S4. The coordinates of the center position of the moving component calculated by each of the multiple grid points in the magnetic field distribution map are fused. The fusion process includes weighted averaging or Kalman filtering to output the final coordinates of the center position of the moving component. In the weighted average, weights are assigned based on the signal-to-noise ratio and magnetic field gradient of each grid point, and then the summation is normalized. Grid points with large magnetic field gradients and high signal-to-noise ratios that are close to the first or second sensor are assigned high weights, while grid points at the edge of the stator assembly, with weak magnetic fields and high noise are assigned low weights. In the Kalman filter, the center position coordinates of the stator assembly calculated for each grid point are used as observations to establish a motion state model of the stator assembly. All observations are fused through a prediction-correction loop to output smooth center position coordinates.

[0019] Preferably, in the mover position detection method, when the first sensor group is arranged on the surface of the stator assembly and the second sensor group is arranged inside the stator assembly, the height difference between them along the third direction is... When the projections of both objects onto a plane perpendicular to the third direction are offset by λ / 2 along both the first and second directions, the mover position detection method further includes: Spatial interpolation is performed on the first sensing information and the second sensing information respectively to construct the lower-level magnetic field distribution map under the unified spatial coordinate system. and upper magnetic field distribution map Among them, the lower magnetic field distribution map and upper magnetic field distribution map It has vertically paired grid points; The lower magnetic field distribution map is calculated according to the following formula (1). and the upper magnetic field distribution map The magnetic field ratio of all vertically paired grid points; (1) According to the magnetic field ratio Determine whether the moving part component is tilted; If the moving part is tilted, the attitude of the moving part is calculated and corrected. If the moving component is not tilted, the local suspension height H of the moving component corresponding to each grid point is calculated in reverse according to the formula (1); The average suspension height H of all local suspension heights is obtained by averaging all the local suspension heights H. .

[0020] Preferably, in the method of the present invention, "calculating the attitude of the mover assembly" includes the following steps: The magnetic field ratio of all grid points Construct as a two-dimensional matrix ; Calculate the gradient Gx in the X direction and the gradient Gy in the Y direction respectively, wherein the gradient Gx in the X direction corresponds to the pitch angle of the mover component in the first direction, and the gradient Gy in the Y direction corresponds to the roll angle of the mover component in the second direction. Calculate the mean of the gradient Gx in the X direction for all grid points. The mean of the gradient Gy in the Y direction ; Substitute all the mean values ​​into a preset gradient-tilt calibration model to calculate the pitch and roll angles of the mover assembly. The gradient-tilt calibration model is established by offline control of the mover assembly to perform pure pitch and pure roll standard motions at a fixed overall height. The gradient-tilt calibration model includes the mapping relationship between different pitch angles and the mean value of the gradient in the X direction of the entire field, and the mapping relationship between different roll angles and the mean value of the gradient in the Y direction of the entire field. Based on the center position coordinates of the moving component and the average suspension height The pitch angle (Pitch) and roll angle (Roll) are calculated, and the position and attitude of the moving part are output.

[0021] Compared with the prior art, the planar motor of the present invention has a first sensor group and a second sensor group arranged on the stator assembly to jointly cover its working plane. The first sensor group has multiple first sensors arranged in an array along a first direction and a second direction with a first spacing, and the second sensor group has multiple second sensors arranged in an array along the first direction and a second direction with a second spacing. Furthermore, the second sensor group is staggered relative to the first sensor group in the first direction with a first offset and in the second direction with a second offset. The first sensor group is used to detect the magnetic field strength of the mover assembly to obtain first sensing information, and the second sensor group is used to detect the magnetic field strength of the mover assembly to obtain second sensing information. During detection, the processing unit acquires the first sensing information of the first sensor group and the second sensing information of the second sensor group, and constructs a magnetic field distribution map in a unified spatial coordinate system based on the first sensing information, the second sensing information and the offset coordinate set. Then, based on the magnetic field distribution map and the preset magnetic field distribution-position mapping relationship, the center position coordinates of the mover assembly are calculated. This application adopts a staggered arrangement of the first sensor group and the second sensor group, which can significantly reduce the number of sensors while ensuring position detection accuracy. This reduces the wiring difficulty of the sensors and PCB stator board, thereby reducing hardware requirements and hardware costs. At the same time, it reduces the difficulty of data acquisition and processing, thereby improving real-time control performance and greatly reducing system operation and power consumption.

[0022] Correspondingly, the planar motor mover position detection method based on this application also has the above-mentioned technical effects. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of a sensor arrangement scheme in the existing technology.

[0024] Figure 2 This is a schematic diagram of the stator assembly in one embodiment of the present invention.

[0025] Figure 3 This is a schematic diagram of the sensor array arrangement on the stator assembly in one embodiment of the present invention.

[0026] Figure 4 This is a flowchart of a mover position detection method according to one embodiment of the present invention. Detailed Implementation

[0027] Embodiments of the present invention will now be described with reference to the accompanying drawings, in which similar element reference numerals represent similar elements. It should be noted that the directional descriptions involved in the present invention, such as up, down, left, right, front, and rear, indicating directions or positional relationships, are based on the directions or positional relationships shown in the drawings and are only for the convenience of describing the technical solutions of this application or / and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. The terms "first," "second," etc., described are only used to distinguish technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the sequential relationship of the indicated technical features.

[0028] Combination Figures 2-3 As shown, in one embodiment of the present invention, the provided planar motor 100 includes a stator assembly 110, a mover assembly 120, a position detection assembly 130, and a processing unit. The stator assembly includes multiple stator units 111, which are assembled to form a working plane. Each stator unit 111 includes a first coil group and a second coil group, the arrangement directions of the first and second coil groups being staggered. The mover assembly 120 can move along the working plane under the drive of the stator units 111. The mover assembly 120 includes a first magnet array and a second magnet array, the array directions of the first and second magnet arrays being staggered. For example, in one specific embodiment, the mover assembly 120 is provided with two first magnet arrays and two second magnet arrays. The two first magnet arrays are parallel to each other along a first direction and are offset from the center of the mover assembly 120. The two second magnet arrays are parallel to each other along a second direction and are also offset from the center of the mover assembly 120. This makes the array directions of the first and second magnet arrays perpendicular to each other, and the first and second magnet arrays are arranged sequentially and at intervals around the center of the mover assembly 120. Of course, the number of magnet arrays in the mover assembly 120 is not limited to four, and the arrangement is not limited to this specific embodiment.

[0029] In one embodiment, both the first magnet array and the second magnet array are preferably Halbach permanent magnet arrays. Since the Halbach array magnetic field has a nonlinear but monotonic variation characteristic in space, there is a reversible mapping relationship between the magnetic field vector and the position of the mover component 120 in a local region. Based on this, the center position coordinates of the mover component 120 can be calculated in reverse using mathematical means, as detailed below. Of course, it is not limited to using a Halbach array magnetic field.

[0030] See below. Figure 3As shown, in this embodiment, the position detection component 130 includes a first sensor group 131 and a second sensor group 132 disposed on the stator assembly. The second sensor group 132 is offset relative to the first sensor group 131 in a first direction by a first offset, and is also offset relative to the first sensor group 131 in a second direction by a second offset. The first sensor group 131 and the second sensor group 132 together cover the working plane of the stator assembly 110. Specifically, the first sensor group 131 includes a plurality of first sensors, which are arranged in an array along the first and second directions with a first spacing. The second sensor group 132 includes a plurality of second sensors, which are arranged in an array along the first and second directions with a second spacing. The second sensors are offset relative to the first sensors along the first direction by a first offset and along the second direction by a second offset. The first sensor group 131 is used to detect the magnetic field strength of the mover assembly 120 to obtain first sensing information, and the second sensor group 132 is used to detect the magnetic field strength of the mover assembly 120 to obtain second sensing information. In this embodiment, by using a double staggered arrangement of the first sensor group 131 and the second sensor group 132, the number of sensors is greatly reduced while maintaining the same spatial sampling density as the traditional dense arrangement scheme, thereby reducing hardware costs and the difficulty of data acquisition and processing.

[0031] Continue reading Figure 3 As shown, in this embodiment, the processing unit is communicatively connected to the first sensor group 131 and the second sensor group 132, respectively. The processing unit is used to: acquire the first sensing information of the first sensor group 131 and the second sensing information of the second sensor group 132; then, construct a magnetic field distribution map in a unified spatial coordinate system based on the first sensing information, the second sensing information and the offset coordinate set; wherein, the offset coordinate set is the physical coordinate information of all the first sensors and all the second sensors in the unified spatial coordinate system constructed based on the first offset and the second offset; finally, calculate the center position coordinates of the actuator component 120 according to the magnetic field distribution map and the preset magnetic field distribution-position mapping relationship.

[0032] In this embodiment, the processing unit pre-constructs and stores a set of deviation coordinates. Specifically, the processing unit first defines a global coordinate system covering the working plane of the stator assembly 110 and outputs a unified spatial coordinate system framework. Then, it acquires the identification information of all first sensors and the physical coordinates of each first sensor to form a first set of coordinates, and acquires the identification information of all second sensors and the physical coordinates of each second sensor to form a second set of coordinates. Next, the first set of coordinates and the second set of coordinates are integrated into the unified spatial coordinate system framework, thereby constructing the set of deviation coordinates. This set of deviation coordinates contains the mapping relationship between the identification information of each first sensor and each second sensor and its physical coordinates.

[0033] In this embodiment, when constructing the magnetic field distribution map, the processing unit, after acquiring the first induction information, the second induction information, and the offset coordinate set, merges the first induction information and the second induction information according to the offset coordinate set to construct a magnetic field distribution map located in a unified spatial coordinate system. This magnetic field distribution map serves as the reference for subsequent position calculations or spatial interpolation, as detailed below.

[0034] The following is combined Figure 1 , Figure 3 As shown, in one embodiment of the present invention, the first spacing and the second spacing are equal and both are λ, and the first offset and the second offset are equal and both are λ / 2. Where λ = L / N, λ is the length of the magnetic field period, L is the physical length of the first magnet array and the second magnet array, and N is the number of magnetic field periods contained in the first magnet array and the second magnet array. This results in the first sensor group 131 and the second sensor group 132 forming a bidirectional λ / 2 staggered arrangement, that is, the equivalent sampling distance between the two sensor groups is λ / 2. Figure 1 The traditional dense arrangement scheme shown has the same spatial sampling density, but the number of sensors in this scheme is greatly reduced.

[0035] Continue reading Figure 3 As shown, in this embodiment, the first sensor starts from the edge starting position (0,0) of the stator assembly 110 and is arranged at a distance λ in both the first and second directions. The second sensor starts from the edge starting position (0,0) of the stator assembly 110 at an offset of λ / 2, and is also arranged at a distance λ in both the first and second directions. Therefore, the total number of the first sensors is N. 11 *N 12 N 11 = L1 / λ+1,N 12= L2 / λ+1, the total number of the second sensors is N 21 *N 22 N 21 = L1 / λ,N 12=L2 / λ; where N 11 N represents the number of first sensors in the first direction. 12 N represents the number of the first sensors in the second direction. 21 N represents the number of second sensors in the first direction. 22 L1 represents the number of second sensors in the second direction, L2 represents the first length of stator assembly 110 in the first direction, and L3 represents the second length of stator assembly 110 in the second direction.

[0036] Combination Figure 1 , Figure 3 As shown, the example is given with a stator assembly 110 having a first length of 240mm in the first direction and a second length of 40mm in the second direction, with λ=40mm. The physical length L of the first and second magnet arrays is 80mm, and the number of magnetic field cycles N contained in the first and second magnet arrays is 2. Therefore, the length of the magnetic field cycle λ=80 / 2=40mm. In this embodiment, when arranging the first sensors, starting from the edge starting position (0,0) of the stator assembly 110, one first sensor is placed every λ=40mm along the first direction (X direction) and the second direction (Y direction). Therefore, the number of first sensors in the first direction (X direction) is 240mm / 40mm+1=7, and the number in the second direction is also 7, for a total of 49 first sensors. Figure 3 (As shown by the blue dot). When arranging the second sensors, starting from the edge starting position (0,0) of the stator assembly 110, offset by λ / 2 = 20mm, a second sensor is placed every λ = 40mm along the first direction (X direction) and the second direction (Y direction), such that the number of second sensors in the first direction (X direction) is 240mm / 40mm = 6, and the number in the second direction (Y direction) is also 6, for a total of 36 second sensors. Figure 3 (As shown by the orange dot). The two sensor groups have a total of 85 sensors, compared to... Figure 1 The traditional method, which requires 169 sensors with a spacing of λ / 2=20mm, is shown in the figure. This method saves about 49% of the number of sensors while maintaining the equivalent sampling density of λ / 2. This greatly reduces the hardware cost, data materials, and computational load.

[0037] See below for further details. Figure 3As shown, in one embodiment of the present invention, after the processing unit "constructs a magnetic field distribution map in a unified spatial coordinate system based on the first sensing information, the second sensing information, and the offset coordinate set," it is further used to: perform spatial interpolation on the magnetic field distribution map to construct a high-resolution magnetic field distribution grid map. The purpose of constructing the high-resolution magnetic field distribution grid map is to reconstruct the discrete sampling points of the first sensor and the second sensor into a continuous high-resolution magnetic field distribution through mathematical interpolation, providing a more accurate basis for subsequent calculation of the center position coordinates of the actuator component 120, thereby improving the position accuracy.

[0038] In one specific implementation, when constructing the high-resolution magnetic field distribution grid map, the processing unit first defines a high-density grid map completely independent of the magnetic field distribution map. This high-density grid map shares the same coordinate system as the deviation coordinate set, and the grid spacing of the high-density grid map is much smaller than the first offset and the second offset. For example, a preset grid spacing of 2mm is much smaller than the 20mm offset, thus generating 121×121=14641 grid points within a 240mm×240mm working plane. Of course, the grid spacing is not limited to this. Then, spatial interpolation is performed on each grid point in the high-density grid map; that is, the corresponding magnetic field vector is calculated for each grid point through interpolation, and then the calculated magnetic field vector is filled into each grid point accordingly, thereby constructing a complete high-resolution magnetic field distribution grid map. In subsequent steps, the processing unit uses this high-resolution magnetic field distribution grid map to reverse-calculate the center position coordinates of the actuator component 120, which can effectively improve positional accuracy.

[0039] In one embodiment of the present invention, spatial interpolation includes bilinear interpolation or bicubic interpolation. Bilinear interpolation uses the four nearest original sampling points around each grid point in the high-density grid map for step-by-step linear interpolation, which has the advantages of simple calculation and fast computation speed. For bicubic interpolation, sixteen original sampling points around each grid point in the high-density grid map are used for cubic polynomial surface fitting. This method has a higher computational cost than bilinear interpolation, but the first derivative of the interpolation surface is continuous, the magnetic field transition is more natural and smooth, it can better restore the sine and cosine variation characteristics of the Halbach array magnetic field, and can effectively suppress high-frequency noise introduced by interpolation, thus resulting in higher accuracy.

[0040] Continue to combine Figure 2-3As shown, in one embodiment of the present invention, the preset magnetic field distribution-position mapping relationship includes a Halbach array magnetic field analytical model or a pre-calibrated magnetic field-position lookup table. The Halbach array magnetic field analytical model is used to substitute the actually detected magnetic field vectors in the magnetic field distribution map into the analytical equations, and to calculate the center position coordinates of the mover component 120 through iterative or numerical solutions. In this method, there is no need for offline calibration of the mover component 120's position, and the calculation requires less storage resources; however, the iterative computation is large, and each grid point requires nonlinear solutions.

[0041] In another implementation, a pre-calibrated magnetic field-position lookup table is used for calculation. Specifically, the processing unit uses the actually detected magnetic field vector in the magnetic field distribution map as an index to query the magnetic field-position lookup table to obtain the center position coordinates of the moving component 120. This method requires constructing the magnetic field-position lookup table offline. Specifically, the moving component 120 is controlled to traverse all positions within its full travel range across the working plane, and the magnetic field data corresponding to the coordinates of each moving component 120 is recorded, thereby constructing and storing a magnetic field-position mapping table. Thus, during online reverse calculation, the measured magnetic field vector is used as an index to search for the corresponding entry in the lookup table, obtaining the center position coordinates of the moving component 120, which are then directly output. This method has no requirements for the magnetic field model and offers fast online calculation speed, but suffers from the problem of large offline calibration workload and high time and effort consumption.

[0042] Recombined Figure 2-3 As shown, in one embodiment of the present invention, the processing unit is further configured to: perform a fusion process on the center position coordinates of the mover assembly 120 calculated from multiple grid points in the magnetic field distribution map, the fusion process including weighted averaging or Kalman filtering, to output the final center position coordinates of the mover assembly 120, so as to power the motor control system for closed-loop position control. Specifically, after the processing unit calculates the positions of multiple grid points, it obtains multiple sets of estimated center position coordinates. These multiple sets of estimated values ​​may contain noise and small deviations, so a fusion process is required to make the final output structure more accurate.

[0043] In one implementation, a weighted average is used, that is, weights are assigned to each grid point based on its signal-to-noise ratio and magnetic field gradient magnitude, and then the average is summed and normalized. Grid points that are close to the first or second sensor and have a large magnetic field gradient and a high signal-to-noise ratio are assigned high weights, while grid points at the edge of the stator assembly 110, with weak magnetic fields, and high noise are assigned low weights.

[0044] In another implementation, Kalman filtering is employed. Specifically, the center position coordinates of the stator assembly 110 calculated from each grid point are used as observations to establish a motion state model of the stator assembly 110. All observations are then fused through a prediction-correction loop to suppress sensor random noise, interpolation errors, magnetic field model errors, etc., thereby outputting smooth center position coordinates. This method can balance steady-state accuracy and dynamic response in dynamic scenarios.

[0045] Recombined Figure 2-3 As shown, in one embodiment of the present invention, a first sensor group 131 is arranged on the surface of the stator assembly 110, and a second sensor group 132 is arranged inside the stator assembly 110, with a height difference between the two in a third direction. Furthermore, the projections of both objects onto a plane perpendicular to the third direction are offset by a distance of λ / 2 along both the first direction (X direction) and the second direction (Y direction), see [reference]. Figure 3 As shown. The first direction (X direction), the second direction (Y direction), and the third direction are all perpendicular to each other. This arrangement of the two sets of sensors eliminates common-mode interference when calculating the levitation height of the actuator assembly 120, resulting in higher stability and accuracy of the final calculated levitation height.

[0046] In this embodiment, the processing unit is further configured to: perform spatial interpolation on the first sensing information and the second sensing information respectively, so as to construct a lower-level magnetic field distribution map under a unified spatial coordinate system. and upper magnetic field distribution map Among them, the lower magnetic field distribution map and upper magnetic field distribution map It has vertically paired grid points; The lower magnetic field distribution map is calculated according to the following formula (1). and upper magnetic field distribution map The ratio of the magnetic fields of all vertically paired grid points ; (1) In formula (1), H is the local levitation height of the moving part 120; The fixed height difference between the first sensor group 131 and the second sensor group 132, which are arranged vertically. According to the magnetic field ratio Determine whether the moving part 120 is tilted; If the moving part 120 is tilted, the attitude of the moving part 120 is calculated and corrected. If the moving part 120 is not tilted, the local suspension height H of the moving part 120 corresponding to each grid point is calculated in reverse according to formula (1); The average suspension height H of all local suspension heights is obtained by averaging all the local suspension heights H. .

[0047] Specifically, since the vertical magnetic field strength of the Halbach permanent magnet array decreases inversely with the square of the levitation height, the magnetic field strength detected by the second sensor group 132 in the lower layer is... The magnetic field strength detected by the first sensor group 131 in the upper layer .

[0048] Within the aforementioned magnetic field strength, common-mode interferences such as temperature drift, Hall effect device sensitivity deviation, and overall permanent magnet magnetic attenuation will scale synchronously. , This causes both sensors to increase or decrease synchronously, making it impossible to distinguish between temperature changes and altitude changes. Therefore, it is necessary to eliminate common-mode interference. In this embodiment, the magnetic field strength detected by the first sensor group 131 and the magnetic field strength detected by the second sensor group 132 are first spatially interpolated, and then the magnetic field ratio is calculated using the above formula (1). magnetic field ratio It is only uniquely related to the local suspension height H. The larger the value, the smaller the value of H, meaning the closer the mover assembly 120 is to the stator assembly 110, and vice versa. The smaller the value, the larger the value of H, meaning the higher the mover assembly 120 levitates, thus eliminating common-mode interference from factors such as temperature, components, and magnet strength. This further eliminates the impact of common-mode interference on height calculation, resulting in a more accurate calculated levitation height.

[0049] In this embodiment, "based on the magnetic field ratio" When determining whether the moving part 120 is tilted, the magnetic field ratio is used as a reference. Whether a continuous gradient distribution appears can be used to determine this. Specifically, when the mover assembly 120 tilts, the local suspension height H of the grid points in different regions of the mover assembly 120 is different. When one side of the mover assembly 120 is depressed, the magnetic field ratio of the grid points in the corresponding region changes. The magnetic field ratio of the grid points in the corresponding region is too large when the other side of the mover assembly 120 is raised. The smaller value results in a lower magnetic field ratio. A smooth gradient field changes from one side to the other, forming a two-dimensional continuous gradient field. This is achieved through the magnetic field ratio... It eliminates common-mode interference and simultaneously determines whether the mover assembly 120 is tilted, making tilt detection more convenient and accurate.

[0050] When the moving component 120 is not tilted, the local suspension height H of the moving component 120 corresponding to each grid point can be calculated in reverse according to formula (1). The local suspension height H of all grid points in the entire working plane is basically the same. Then, by performing a weighted average or mean filtering on all the local suspension heights H, the overall average suspension height of the moving component 120 can be obtained. .

[0051] In one embodiment of the present invention, the processing unit "calculates the attitude of the mover assembly 120" specifically performs the following operation: The ratio of the magnetic fields of all grid points Construct as a two-dimensional matrix ; Calculate the gradient Gx in the X direction and the gradient Gy in the Y direction respectively. The gradient Gx in the X direction corresponds to the pitch angle of the mover component 120 in the first direction (X direction) at the front-to-back height, and the gradient Gy in the Y direction corresponds to the roll angle of the mover component 120 in the second direction (Y direction) at the left-to-right height. Calculate the mean of the gradient Gx in the X direction for all grid points. The mean of the gradient Gy in the Y direction ; Substitute all the mean values ​​into the preset gradient-tilt calibration model to calculate the pitch and roll angles of the mover assembly 120. The gradient-tilt calibration model is established by offline control of the mover assembly 120 to perform pure pitch and pure roll standard motions at a fixed overall height. The gradient-tilt calibration model includes the mapping relationship between different pitch angles and the mean values ​​of the gradients in the X direction of the entire field, and the mapping relationship between different roll angles and the mean values ​​of the gradients in the Y direction of the entire field. Based on the center position coordinates of the mover assembly 120 and the average suspension height The pitch and roll angles are calculated and output to determine the position and attitude of the mover component 120.

[0052] In this embodiment, when constructing the gradient-tilt calibration model offline, the motion element 120 is controlled to perform pure pitch and pure roll standard movements at a fixed overall height, and the entire allowable tilt angle range is traversed. Then, the entire field under different pitch angles is recorded. average Full court under different roll angles average Next, establish the mapping relationship: , This process generates a lookup table, thereby establishing a gradient-tilt calibration model. This model includes the mean gradients of different pitch angles and the gradient in the first direction (X-direction) across the entire field. The mapping relationship between them, the mean gradient of different tilt angles and the second direction (Y direction) of the whole field The mapping relationship between them.

[0053] When the processing unit calculates the attitude of the mover assembly 120, it takes the gradient of all grid points in the first direction (X direction). Calculate the global mean Substituting these values ​​into the gradient-tilt calibration model, the pitch angle can be found and directly output. Correspondingly, the gradient of all grid points in the entire field in the second direction (Y direction) is taken. Calculate the global mean Substituting these values ​​into the gradient-tilt calibration model, the roll angle can be found and directly output. For example, when the front end of the mover assembly 120 is depressed and the rear end is raised, the ratio of the magnetic field from left to right in the X direction is... Continuously decreasing A negative value corresponds to a negative pitch angle; when the left side of the mover assembly is depressed and the right side is raised, the ratio of the magnetic field from top to bottom in the Y direction is... Continuously decreasing A negative value corresponds to a negative tilt angle; when the mover assembly 120° is completely horizontal, the ratio of the magnetic fields across the entire field is... Almost equal, Approaching 0, both pitch and roll angles are approximately 0.

[0054] In this embodiment, due to the magnetic field ratio Common-mode interference has been completely eliminated, therefore, attitude calculation is unaffected by the overall strength of the magnetic field. Furthermore, by averaging the gradients of several interpolation grid points, single-point sensor noise and local magnetic defects do not interfere with the attitude results, thus improving the accuracy of the output. In addition, the pitch and roll angles of the mover assembly 120 are independent of its center position coordinates. Therefore, the center position coordinates and dual-axis tilt (Pitch, Roll) of the mover assembly 120 can be output simultaneously. In other words, the processing unit can calculate and output the center position coordinates and dual-axis tilt angles of the mover assembly 120 based on the magnetic field strength detection, decoupling the tilt angles of the mover assembly 120 from its planar position.

[0055] The following is combined Figures 2-4 As shown, in one embodiment of the present invention, a method for detecting the position of a moving part based on the above-described planar motor is also provided, the method comprising the following steps: S11. Obtain the magnetic field strength of the moving component detected by the first sensor group as the first sensing information, and obtain the magnetic field strength of the moving component detected by the second sensor group as the second sensing information. S12. Based on the first sensing information, the second sensing information, and the offset coordinate set, construct a magnetic field distribution map in a unified spatial coordinate system; wherein, the offset coordinate set is the physical coordinate information of all first sensors and all second sensors in a unified spatial coordinate system constructed based on the first offset and the second offset. S13. Based on the magnetic field distribution diagram and the preset magnetic field distribution-position mapping relationship, calculate the center position coordinates of the moving component.

[0056] In one embodiment of the mover position detection method of the present invention, the following steps are included before step S11: Construct and store the set of deviation coordinates.

[0057] Specifically, "constructing and storing the set of deviation coordinates" includes the following steps: Define a global coordinate system that covers the working plane of the stator assembly and output a unified spatial coordinate system framework; Obtain the identity information of all first sensors and the physical coordinates of each first sensor to form a first set of coordinates, and obtain the identity information of all second sensors and the physical coordinates of each second sensor to form a second set of coordinates; The first set of coordinates and the second set of coordinates are integrated into a unified spatial coordinate system framework, thereby constructing and storing the deviation coordinate set.

[0058] This set of deviation coordinates contains the mapping relationship between the identity information of each first sensor and each second sensor and their physical coordinates.

[0059] In one embodiment of the mover position detection method of the present invention, step S12, "constructing a magnetic field distribution map," specifically includes: Acquire the first sensing information, the second sensing information, and the offset coordinate set respectively; The first and second induction information are merged according to the deviation coordinate set to construct a magnetic field distribution map under a unified spatial coordinate system.

[0060] In one embodiment of the mover position detection method of the present invention, step S12 further includes the following steps: Spatial interpolation is performed on the magnetic field distribution map to construct a high-resolution magnetic field distribution grid map.

[0061] The purpose of constructing a high-resolution magnetic field distribution grid map is to reconstruct the discrete sampling points of the first and second sensors into a continuous high-resolution magnetic field distribution through mathematical interpolation, providing a more accurate basis for subsequent calculation of the center position coordinates of the actuator component 120, thereby improving the position accuracy.

[0062] In one specific implementation, the step of "spatial interpolating the magnetic field distribution map to construct a high-resolution magnetic field distribution grid map" includes the following steps: Define a high-density grid map that is completely independent of the magnetic field distribution map; wherein, the high-density grid map and the deviation coordinate set share the same coordinate system, and the grid spacing of the high-density grid map is much smaller than the first offset and the second offset; for example, the preset grid spacing is 2mm, which is much smaller than the offset of 20mm, thus, 121×121=14641 grid points will be generated in a 240mm×240mm working plane; Spatial interpolation is performed on each grid point of the high-density grid to construct a high-resolution magnetic field distribution grid map. That is, for each grid point, the corresponding magnetic field vector is calculated through interpolation, and then the calculated magnetic field vector is filled into each grid point to construct a complete high-resolution magnetic field distribution grid map.

[0063] In one embodiment of the present invention, spatial interpolation includes bilinear interpolation or bicubic interpolation. Bilinear interpolation uses the four nearest original sampling points around each grid point in the high-density grid map for step-by-step linear interpolation, which has the advantages of simple calculation and fast computation speed. For bicubic interpolation, sixteen original sampling points around each grid point in the high-density grid map are used for cubic polynomial surface fitting. This method has a higher computational cost than bilinear interpolation, but the first derivative of the interpolation surface is continuous, the magnetic field transition is more natural and smooth, it can better restore the sine and cosine variation characteristics of the Halbach array magnetic field, and can effectively suppress high-frequency noise introduced by interpolation, thus achieving higher accuracy.

[0064] In one embodiment of the mover position detection method of the present invention, in step S13, the preset magnetic field distribution-position mapping relationship includes: The Halbach array magnetic field analytical model is used to substitute the actual detected magnetic field vector in the magnetic field distribution map into the analytical equation, and calculate the center position coordinates of the mover component in reverse through iterative or numerical solution. Alternatively, a pre-calibrated magnetic field-position lookup table can be used, with the actual detected magnetic field vector in the magnetic field distribution map as an index, to query the magnetic field-position lookup table to obtain the center position coordinates of the mover component. The magnetic field-position lookup table is established by offline control of the mover component to traverse all positions within the entire stroke and recording the stator magnetic field data corresponding to the coordinates of each mover component.

[0065] Specifically, in the Halbach array magnetic field analytical model, there is no need for offline calibration of the position of the mover component 120, and the storage resources required during the solution are minimal, but the iterative computation is large, and each grid point requires nonlinear solution. In contrast, the magnetic field-position lookup table method has no requirements for the magnetic field model and the online solution speed is fast, but it suffers from the problem of large offline calibration workload and time-consuming process.

[0066] Furthermore, in the mover position detection method of the present invention, the following steps are included before step S11: Build a magnetic field-location lookup table during the offline phase.

[0067] Specifically, by controlling the moving component 120 to traverse all positions within its full travel range on the working plane and recording the magnetic field data corresponding to the coordinates of each moving component 120, a magnetic field-position mapping table is constructed and stored. Thus, during online reverse calculation, the measured magnetic field vector is used as an index to search for the corresponding entry in the lookup table, thereby obtaining and directly outputting the coordinates of the center position of the corresponding moving component 120.

[0068] In one embodiment of the mover position detection method of the present invention, the method further includes the following steps: S14. The center position coordinates of the mover component calculated by each of the multiple grid points in the magnetic field distribution map are fused. The fusion process includes weighted averaging or Kalman filtering to output the final center position coordinates of the mover component.

[0069] In step S14, multiple sets of estimated center position coordinates are obtained by solving the positions of multiple grid points. These multiple sets of estimated values ​​may contain noise and small deviations, so they need to be fused to eliminate the noise and small deviations in the estimated center position coordinates, making the final output structure more accurate.

[0070] In one implementation, a weighted average is used, that is, weights are assigned to each grid point based on its signal-to-noise ratio and magnetic field gradient magnitude, and then the average is summed and normalized. Grid points that are close to the first or second sensor and have a large magnetic field gradient and a high signal-to-noise ratio are assigned high weights, while grid points at the edge of the stator assembly 110, with weak magnetic fields, and high noise are assigned low weights.

[0071] In another implementation, Kalman filtering is employed. Specifically, the center position coordinates of the stator assembly 110 calculated from each grid point are used as observations to establish a motion state model of the stator assembly 110. All observations are then fused through a prediction-correction loop to suppress sensor random noise, interpolation errors, magnetic field model errors, etc., thereby outputting smooth center position coordinates. This method can balance steady-state accuracy and dynamic response in dynamic scenarios.

[0072] See below again Figure 3 As shown, in one embodiment of the present invention, a first sensor group 131 is arranged on the surface of the stator assembly 110, and a second sensor group 132 is arranged inside the stator assembly 110, with a height difference between the two staggered along a third direction. The projections of both objects onto a plane perpendicular to the third direction are offset by λ / 2 along both the first direction (X direction) and the second direction (Y direction). In this embodiment, the mover position detection method further includes the following steps: S21. Obtain the magnetic field strength of the moving component detected by the first sensor group as the first sensing information, and obtain the magnetic field strength of the moving component detected by the second sensor group as the second sensing information. S22. Spatial interpolation is performed on the first and second induction information respectively to construct a lower-level magnetic field distribution map under a unified spatial coordinate system. and upper magnetic field distribution map Among them, the lower magnetic field distribution map and upper magnetic field distribution map It has vertically paired grid points; S23. Calculate the lower magnetic field distribution map according to the following formula (1). and upper magnetic field distribution map The magnetic field ratio of all vertically paired grid points; (1) In formula (1), H is the local levitation height of the moving part 120; The fixed height difference between the first sensor group 131 and the second sensor group 132, which are arranged vertically. S24. Based on the magnetic field ratio Determine if the moving component is tilted; S25. If the moving part is tilted, the attitude of the moving part is calculated and corrected. S26. If the moving component is not tilted, the local suspension height H of the moving component corresponding to each grid point is calculated in reverse according to formula (1); S27. Take the average value of all local suspension heights H to obtain the average suspension height of the moving part assembly. .

[0073] In this embodiment, since the vertical magnetic field strength of the Halbach permanent magnet array decreases inversely with the square of the levitation height, the magnetic field strength detected by the second sensor group 132 in the lower layer is... The magnetic field strength detected by the first sensor group 131 in the upper layer .

[0074] Within the aforementioned magnetic field strength, common-mode interferences such as temperature drift, Hall effect device sensitivity deviation, and overall permanent magnet magnetic attenuation will scale synchronously. , This causes both sensors to increase or decrease synchronously, making it impossible to distinguish between temperature changes and height changes. Therefore, it is necessary to eliminate common-mode interference. Specifically, in step S22, the magnetic field strength detected by the first sensor group 131 and the second sensor group 132 is spatially interpolated to obtain the upper magnetic field distribution map and the upper magnetic field distribution map. Then, the magnetic field ratio is calculated according to formula (1) in step S23. magnetic field ratio It is only uniquely related to the local suspension height H. The larger the value, the smaller the value of H, meaning the closer the mover assembly 120 is to the stator assembly 110, and vice versa. The smaller the value, the larger the value of H, meaning the higher the mover assembly 120 levitates, thus eliminating common-mode interference from factors such as temperature, components, and magnet strength. This further eliminates the impact of common-mode interference on height calculation, resulting in a more accurate calculated levitation height.

[0075] In this embodiment, when determining whether the moving part assembly is tilted in step S24, the magnetic field ratio is used as the basis. Whether a continuous gradient distribution appears can be used to determine this. Specifically, when the mover assembly 120 tilts, the local suspension height H of the grid points in different regions of the mover assembly 120 is different. When one side of the mover assembly 120 is depressed, the magnetic field ratio of the grid points in the corresponding region changes. The magnetic field ratio of the grid points in the corresponding region is too large when the other side of the mover assembly 120 is raised. The smaller value results in a lower magnetic field ratio. A smooth gradient field changes from one side to the other, forming a two-dimensional continuous gradient field. This is achieved through the magnetic field ratio... It eliminates common-mode interference and simultaneously determines whether the mover assembly 120 is tilted, making tilt detection more convenient and accurate.

[0076] When the moving component 120 is not tilted, in step S26, the local suspension height H of the moving component 120 corresponding to each grid point is calculated in reverse according to formula (1), and the local suspension height H of all grid points in the entire working plane is basically the same. Then, the weighted average or mean filtering of all the local suspension heights H can be used to obtain the overall average suspension height of the moving component 120. .

[0077] In this embodiment, step S25, "calculating the attitude of the moving component," specifically includes the following steps: S251, Ratio of the magnetic fields of all grid points Construct as a two-dimensional matrix ; S252. Calculate the gradient Gx in the X direction and the gradient Gy in the Y direction respectively; where the gradient Gx in the X direction corresponds to the pitch angle of the mover component 120 in the first direction (X direction) and the gradient Gy in the Y direction corresponds to the roll angle of the mover component 120 in the second direction (Y direction). S253. Calculate the mean value of the gradient Gx in the X direction for all grid points. The mean of the gradient Gy in the Y direction ; S254. Substitute all the mean values ​​into the preset gradient-tilt calibration model to calculate the pitch and roll angles of the mover assembly. The gradient-tilt calibration model is established by offline control of the mover assembly 120 to perform pure pitch and pure roll standard motions at a fixed overall height. The gradient-tilt calibration model includes the mapping relationship between different pitch angles and the mean values ​​of the gradients in the X direction of the entire field, and the mapping relationship between different roll angles and the mean values ​​of the gradients in the Y direction of the entire field. S255, Based on the center position coordinates of the moving part and the average suspension height The pitch and roll angles are calculated and output to determine the position and attitude of the moving component.

[0078] In this embodiment, the method further includes the following steps before step S21: Construct a gradient-tilt calibration model in an offline state.

[0079] Specifically, by controlling the mover assembly 120 to perform standard pitch and roll movements at a fixed overall height, and traversing the entire allowable tilt angle range, the entire field at different pitch angles is recorded. average Full court under different roll angles average Next, establish the mapping relationship: , This process generates a lookup table, thereby establishing a gradient-tilt calibration model. This model includes the mean gradients of different pitch angles and the gradient in the first direction (X-direction) across the entire field. The mapping relationship between them, the mean gradient of different tilt angles and the second direction (Y direction) of the whole field The mapping relationship between them.

[0080] When calculating the attitude of the mover assembly 120 in step S254, the mean value from step S253 is used. Substituting the values ​​into the gradient-tilt calibration model, the pitch angle can be found and directly output. Correspondingly, the global mean in step S253 can be used... Substituting the values ​​into the gradient-tilt calibration model, the roll angle can be found and directly output. For example, when the front end of the mover assembly 120 is depressed and the rear end is raised, the ratio of the magnetic field from left to right in the X direction is... Continuously decreasing A negative value corresponds to a negative pitch angle; when the left side of the mover assembly is depressed and the right side is raised, the ratio of the magnetic field from top to bottom in the Y direction is... Continuously decreasing A negative value corresponds to a negative tilt angle; when the mover assembly 120° is completely horizontal, the ratio of the magnetic fields across the entire field is... Almost equal, Approaching 0, both pitch and roll angles are approximately 0.

[0081] It should be noted that, in this embodiment, the position and orientation of the actuator component 120 and its center position coordinates can be calculated separately or simultaneously. The method and steps for calculating the center position coordinates are as described in steps S11-S14 above and will not be repeated. In this embodiment, due to the magnetic field ratio... Common-mode interference has been completely eliminated, therefore, the position and attitude calculation is unaffected by the overall strength of the magnetic field. Moreover, by averaging the gradients of several interpolation grid points, single-point sensor noise and local magnetic defects will not interfere with the attitude results, thus resulting in higher accuracy of the output. In addition, the pitch and roll angles of the mover assembly 120 are independent of its center position coordinates. Therefore, the center position coordinates and dual-axis tilt (Pitch, Roll) of the mover assembly 120 can be output simultaneously, decoupling the tilt angles of the mover assembly 120 from its planar position.

[0082] In summary, the planar motor of the present invention has a first sensor group 131 and a second sensor group 132 arranged in the stator assembly 110 to jointly cover its working plane. The plurality of first sensors in the first sensor group are arranged in an array along a first direction (X direction) and a second direction (Y direction) with a first spacing. The plurality of second sensors in the second sensor group 132 are arranged in an array along the first direction (X direction) and the second direction (Y direction) with a second spacing. Furthermore, the second sensor group 132 is offset relative to the first sensor group 131 in the first direction (X direction) by a first offset and in the second direction (Y direction) by a second offset. The sensors are arranged in a staggered configuration. The first sensor group 131 detects the magnetic field strength of the mover assembly 120 to obtain first sensing information, and the second sensor group 132 detects the magnetic field strength of the mover assembly 120 to obtain second sensing information. During detection, the processing unit acquires the first sensing information from the first sensor group 131 and the second sensing information from the second sensor group 132. Based on the first sensing information, the second sensing information, and the offset coordinate set, a magnetic field distribution map in a unified spatial coordinate system is constructed. Then, based on the magnetic field distribution map and a preset magnetic field distribution-position mapping relationship, the center position coordinates of the mover assembly 120 are calculated. This application uses a staggered arrangement of the first sensor group 131 and the second sensor group 132, which can significantly reduce the number of sensors while ensuring position detection accuracy. This reduces the wiring difficulty of the sensors and the PCB stator board, thereby reducing hardware requirements and costs. It also reduces the difficulty of data acquisition and processing, thus improving real-time control performance and significantly reducing system operating time and power consumption.

[0083] Correspondingly, the planar motor mover position detection method based on this application also has the above-mentioned technical effects.

[0084] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A planar motor, characterized in that, It includes a stator assembly, a mover assembly, a position detection assembly, and a processing unit; the stator assembly includes multiple stator units, which are spliced ​​together to form a working plane, and each stator unit includes a first coil group and a second coil group, wherein the arrangement directions of the first coil group and the arrangement directions of the second coil group are staggered; The moving part assembly can move along the working plane under the drive of the stator unit; the moving part assembly includes a first magnet array and a second magnet array, the array directions of the first magnet array and the array directions of the second magnet array are intersected; The position detection component includes a first sensor group and a second sensor group disposed on the stator component. The first sensor group includes a plurality of first sensors, which are arranged in an array along the first direction and the second direction at a first spacing. The second sensor group includes a plurality of second sensors, which are arranged in an array along the first direction and the second direction at a second spacing. The second sensor group is staggered relative to the first sensor group in the first direction by a first offset, and the second sensor group is staggered relative to the first sensor group in the second direction by a second offset. The first sensor group and the second sensor group together cover the working plane of the stator assembly. The first sensor group is used to detect the magnetic field strength of the mover assembly to obtain first sensing information, and the second sensor group is used to detect the magnetic field strength of the mover assembly to obtain second sensing information. The processing unit is communicatively connected to the first sensor group and the second sensor group, and the processing unit is used for: Acquire the first sensing information of the first sensor group and the second sensing information of the second sensor group; Based on the first sensing information, the second sensing information, and the offset coordinate set, a magnetic field distribution map in a unified spatial coordinate system is constructed; wherein, the offset coordinate set is the physical coordinate information of all the first sensors and all the second sensors in a unified spatial coordinate system constructed based on the first offset and the second offset; Based on the magnetic field distribution map and the preset magnetic field distribution-position mapping relationship, the center position coordinates of the moving component are calculated.

2. The planar motor as described in claim 1, characterized in that, The first spacing and the second spacing are equal and both are λ, and the first offset and the second offset are equal and both are λ / 2; where λ=L / N, λ is the magnetic field period length, L is the physical length of the first magnet array and the second magnet array, and N is the number of magnetic field periods contained in the first magnet array and the second magnet array.

3. The planar motor as described in claim 2, characterized in that, The total number of the first sensors is N 11 *N 12 N 11 =L1 / λ+1,N 12= L2 / λ+1, the total number of the second sensors is N 21 *N 22 N 21 = L1 / λ,N 12= L2 / λ; Where, N 11 N represents the number of the first sensors in the first direction. 12 N represents the number of the first sensors in the second direction. 21 N represents the number of the second sensors in the first direction. 22 L1 represents the number of the second sensors in the second direction, L2 represents the first length of the stator assembly in the first direction, and L3 represents the second length of the stator assembly in the second direction.

4. The planar motor as described in claim 1, characterized in that, After "constructing a magnetic field distribution map in a unified spatial coordinate system based on the first sensing information, the second sensing information, and the offset coordinate set", the processing unit is further used for: Define a high-density grid map that is completely independent of the magnetic field distribution map; wherein the high-density grid map and the deviation coordinate set share the same coordinate system, and the grid spacing of the high-density grid map is much smaller than the first offset and the second offset; Spatial interpolation is performed on each grid point of the high-density grid to construct a high-resolution magnetic field distribution grid map.

5. The planar motor as described in claim 1, characterized in that, The processing unit is further configured to: perform a fusion process on the coordinates of the center position of the moving component calculated by each of the multiple grid points in the magnetic field distribution map, the fusion process including weighted averaging or Kalman filtering, to output the final coordinates of the center position of the moving component; In the weighted average, weights are assigned based on the signal-to-noise ratio and magnetic field gradient magnitude of each grid point, and then the summation is normalized. Grid points that are close to the first sensor or the second sensor and have a large magnetic field gradient and a high signal-to-noise ratio are assigned high weights, while grid points at the edge of the stator assembly, with weak magnetic fields, and with high noise are assigned low weights. In the Kalman filter, the center position coordinates of the stator assembly calculated for each grid point are used as observations to establish a motion state model of the stator assembly. All observations are fused through a prediction-correction loop to output smooth center position coordinates.

6. The planar motor as described in claim 2, characterized in that, The first sensor group is arranged on the surface of the stator assembly, and the second sensor group is arranged inside the stator assembly, with a height difference between them in a third direction. Furthermore, the projections of both directions onto the plane perpendicular to the third direction are offset by a distance of λ / 2 along both the first and second directions; wherein the first direction, the second direction, and the third direction are mutually perpendicular. The processing unit is also used for: Spatial interpolation is performed on the first sensing information and the second sensing information respectively to construct the lower-level magnetic field distribution map under the unified spatial coordinate system. and upper magnetic field distribution map Among them, the lower magnetic field distribution map and upper magnetic field distribution map It has vertically paired grid points; The lower magnetic field distribution map is calculated according to the following formula (1). and the upper magnetic field distribution map The magnetic field ratio of all vertically paired grid points; （1） According to the magnetic field ratio Determine whether the moving part component is tilted; If the moving part is tilted, the attitude of the moving part is calculated and corrected. If the moving component is not tilted, the local suspension height H of the moving component corresponding to each grid point is calculated in reverse according to the formula (1); The average suspension height H of all local suspension heights is obtained by averaging all the local suspension heights H. .

7. The planar motor as described in claim 6, characterized in that, When "calculating the attitude of the moving component", the processing unit is also used to: The magnetic field ratio of all grid points Construct as a two-dimensional matrix ; Calculate the gradient Gx in the X direction and the gradient Gy in the Y direction respectively, wherein the gradient Gx in the X direction corresponds to the pitch angle of the mover component in the first direction, and the gradient Gy in the Y direction corresponds to the roll angle of the mover component in the second direction. Calculate the mean gradient Gx in the X direction for all grid points. The mean of the gradient Gy in the Y direction ; Substitute all the mean values ​​into a preset gradient-tilt calibration model to calculate the pitch and roll angles of the mover assembly. The gradient-tilt calibration model is established by offline control of the mover assembly to perform pure pitch and pure roll standard motions at a fixed overall height. The gradient-tilt calibration model includes the mapping relationship between different pitch angles and the mean value of the gradient in the X direction of the entire field, and the mapping relationship between different roll angles and the mean value of the gradient in the Y direction of the entire field. Based on the center position coordinates and average suspension height of the moving component The pitch angle (Pitch) and roll angle (Roll) are calculated, and the position and attitude of the moving part are output.

8. A method for detecting the position of the mover of a planar motor according to any one of claims 1-7, characterized in that, Includes the following steps: S1. Obtain the magnetic field strength of the moving part detected by the first sensor group as the first sensing information, and obtain the magnetic field strength of the moving part detected by the second sensor group as the second sensing information. S2. Based on the first sensing information, the second sensing information, and the offset coordinate set, construct a magnetic field distribution map in a unified spatial coordinate system; wherein, the offset coordinate set is the physical coordinate information of all the first sensors and all the second sensors in a unified spatial coordinate system constructed based on the first offset and the second offset; S3. Based on the magnetic field distribution map and the preset magnetic field distribution-position mapping relationship, calculate the center position coordinates of the moving component.

9. The mover position detection method as described in claim 8, characterized in that, Step S2 is followed by the following steps: Define a high-density grid map that is completely independent of the magnetic field distribution map; wherein the high-density grid map and the deviation coordinate set share the same coordinate system, and the grid spacing of the high-density grid map is much smaller than the first offset and the second offset; Spatial interpolation is performed on each grid point of the high-density grid to construct a high-resolution magnetic field distribution grid map.

10. The mover position detection method as described in claim 8, characterized in that, It also includes the following steps: S4. The coordinates of the center position of the moving component calculated by each of the multiple grid points in the magnetic field distribution map are fused. The fusion process includes weighted averaging or Kalman filtering to output the final coordinates of the center position of the moving component. In the weighted average, weights are assigned based on the signal-to-noise ratio and magnetic field gradient of each grid point, and then the summation is normalized. Grid points with large magnetic field gradients and high signal-to-noise ratios that are close to the first or second sensor are assigned high weights, while grid points at the edge of the stator assembly, with weak magnetic fields and high noise are assigned low weights. In the Kalman filter, the center position coordinates of the stator assembly calculated for each grid point are used as observations to establish a motion state model of the stator assembly. All observations are fused through a prediction-correction loop to output smooth center position coordinates.

11. The mover position detection method as described in claim 8, characterized in that, When the first sensor group is arranged on the surface of the stator assembly and the second sensor group is arranged inside the stator assembly, the height difference between them along the third direction is: When the projections of both objects onto a plane perpendicular to the third direction are offset by λ / 2 along both the first and second directions, the mover position detection method further includes: Spatial interpolation is performed on the first sensing information and the second sensing information respectively to construct the lower-level magnetic field distribution map under the unified spatial coordinate system. and upper magnetic field distribution map Among them, the lower magnetic field distribution map and upper magnetic field distribution map It has vertically paired grid points; The lower magnetic field distribution map is calculated according to the following formula (1). and the upper magnetic field distribution map The magnetic field ratio of all vertically paired grid points; （1） According to the magnetic field ratio Determine whether the moving part component is tilted; If the moving part is tilted, the attitude of the moving part is calculated and corrected. If the moving component is not tilted, the local suspension height H of the moving component corresponding to each grid point is calculated in reverse according to the formula (1); The average suspension height H of all local suspension heights is obtained by averaging all the local suspension heights H. .

12. The mover position detection method as described in claim 11, characterized in that, "Calculating the attitude of the moving component" includes the following steps: The magnetic field ratio of all grid points Construct as a two-dimensional matrix ; Calculate the gradient Gx in the X direction and the gradient Gy in the Y direction respectively, wherein the gradient Gx in the X direction corresponds to the pitch angle of the mover component in the first direction, and the gradient Gy in the Y direction corresponds to the roll angle of the mover component in the second direction. Calculate the mean gradient Gx in the X direction for all grid points. The mean of the gradient Gy in the Y direction ; Substitute all the mean values ​​into a preset gradient-tilt calibration model to calculate the pitch and roll angles of the mover assembly. The gradient-tilt calibration model is established by offline control of the mover assembly to perform pure pitch and pure roll standard motions at a fixed overall height. The gradient-tilt calibration model includes the mapping relationship between different pitch angles and the mean value of the gradient in the X direction of the entire field, and the mapping relationship between different roll angles and the mean value of the gradient in the Y direction of the entire field. Based on the center position coordinates and average suspension height of the moving component The pitch angle (Pitch) and roll angle (Roll) are calculated, and the position and attitude of the moving part are output.

Images