TWI664823B - Instant correction method for encoder and system thereof - Google Patents

Abstract

An instant calibration method for an encoder includes the following steps. The movement of the device under test is sensed to obtain a first sine wave signal and a second sine wave signal, wherein the phase of the first sine wave signal and the second sine wave signal are 90 degrees out of phase. Sampling the first sine wave signal and the second sine wave signal to obtain N first digital signal values and N second digital signal values. According to the N first digital signal values and the N second digital signal values, N positioning positions are generated. Add N positioning positions to the calculation group. Regression analysis is performed on the positioning positions in the calculation group to obtain a regression curve. According to the regression curve, the N + 1th predicted position is predicted. According to the ideal position curve, at the time point of the N + 1th predicted position, the ideal position of the device under test is obtained, and the error value between the N + 1th predicted position and the ideal position is used to correct the device to be tested.

Description

Instant correction method and system for encoder

The invention relates to an instant correction method and system for an encoder.

The encoder is mainly used to provide the precise position of the rotor (mover) of the servo motor to meet the requirements of stable speed control and precise positioning of the servo drive device. However, the error caused by the mechanism during assembly will affect the accuracy of the position output of the encoder. In addition, after continuous use for a period of time, the accuracy of the encoder's position output deteriorates due to changes in the relative position of the mechanism or other external environmental pollution. Therefore, how to calculate the position error of the encoder and correct the position output in real time is a problem that needs to be solved at present.

In order to solve the above problems, the present invention provides an instant correction method for an encoder, so as to improve positioning accuracy and extend the service life of the encoder, thereby increasing convenience in use.

The invention provides an instant correction method for an encoder, which includes the following steps. Sensing the motion of the DUT to obtain the first sine wave signal and the first A two-sine wave signal, where the phase of the first sine wave signal and the second sine wave signal differ by 90 degrees. Sampling the first sine wave signal and the second sine wave signal to obtain N first digital signal values and N second digital signal values. According to the N first digital signal values and the N second digital signal values, N positioning positions are generated. Add N positioning positions to the calculation group. Regression analysis is performed on the positioning positions in the calculation group to obtain a regression curve. According to the regression curve, the N + 1th predicted position is predicted. According to the ideal position curve, at the time point of the N + 1th predicted position, an ideal position of the device under test is obtained, and an error value between the N + 1th predicted position and the ideal position is used to correct the device to be tested.

Another embodiment of the present invention provides an instant correction system for an encoder, including a sensing unit, a sampling unit, and a processing unit. The sensing unit is used for sensing the movement of the device under test to obtain a first sine wave signal and a second sine wave signal, wherein the phase of the first sine wave signal and the second sine wave signal are 90 degrees out of phase. The sampling unit samples the first sine wave signal and the second sine wave signal to obtain N first digital signal values and N second digital signal values. The processing unit generates N positioning positions according to the N first digital signal values and the N second digital signal values, adds the N positioning positions to the calculation group, and performs regression analysis on the positioning positions in the calculation group to obtain Regression curve, predict the N + 1th predicted position based on the regression curve, and obtain the ideal position of the component to be tested at the time point of the N + 1th predicted position based on the ideal position curve, and use the N + 1th prediction The error value between the position and the ideal position is corrected for the component under test.

The real-time correction method and system for an encoder disclosed in the embodiment of the present invention obtains a regression curve by obtaining N positioning positions corresponding to the component to be tested, and performing regression analysis on the N positioning positions, thereby predicting the Nth position. +1 Based on the ideal position curve, at the time point corresponding to the N + 1th predicted position, obtain the ideal position of the component under test, and use the error value between the N + 1th predicted position and the corresponding ideal position Correct the component to be tested. In this way, the accuracy of the positioning position of the encoder can be effectively maintained within a certain range, and the service life of the encoder can be extended, so as to increase convenience in use.

100‧‧‧ Real-time correction system for encoder

110‧‧‧sensing unit

120‧‧‧Sampling unit

130‧‧‧processing unit

131‧‧‧ Computing Unit

132‧‧‧ Computing Unit

133‧‧‧correction unit

134‧‧‧Filter

210‧‧‧DUT

220‧‧‧Drive unit

S502 ~ S514, S602, S604, S606‧‧‧ steps

FIG. 1 is a schematic diagram of an instant correction system for an encoder according to an embodiment of the present invention.

FIG. 2A is a schematic diagram of an ideal position output of an encoder according to an embodiment of the present invention.

FIG. 2B is a schematic diagram of the actual position output of the encoder according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of a correspondence relationship between a positioning position, a predicted position, and a regression curve according to an embodiment of the present invention.

FIG. 4 is a comparison diagram of a corrected position signal and an uncorrected position signal according to an embodiment of the present invention.

FIG. 5 is a flowchart of an instant correction method for an encoder according to an embodiment of the present invention.

FIG. 6 is a flowchart of an instant correction method for an encoder according to another embodiment of the present invention.

Other areas of applicability of the device and method of the present invention will be The details provided below are clear and easy to see. It must be understood that the following detailed description and specific embodiments, when presenting an exemplary embodiment of an instant correction system and method for an encoder, are only for the purpose of description and are not intended to limit the scope of the invention.

FIG. 1 shows a schematic diagram of an instant correction system for an encoder according to an embodiment of the present invention. The instant calibration system 100 for an encoder in this embodiment is suitable for calibrating the DUT 210, and the DUT 210, such as a motor, is a variety of different DUTs. Please refer to FIG. 1. The instant calibration system 100 for an encoder includes a sensing unit 110, a sampling unit 120, a processing unit 130, and a storage unit 140.

The sensing unit 110 is connected to the device under test 150 to sense the movement of the device under test 210 to obtain a first sine wave signal and a second sine wave signal. In this embodiment, the phase of the first sine wave signal and the second sine wave signal differ by 90 degrees. The first sine wave signal is, for example, a sine signal, and the second sine wave signal is, for example, a cosine (cos) signal. In addition, the first sine wave signal and a second sine wave signal may be obtained when the DUT 210 starts to operate, or after the DUT 210 is operated for a period of time.

The sampling unit 120 is connected to the sensing unit 110 and samples the first sine wave signal and the second sine wave signal to obtain N first digital signal values and N second digital signal values, where N is a positive integer greater than 1. . In this embodiment, the sampling unit 120 is, for example, a high-speed signal sampler.

The processing unit 130 is, for example, a digital signal processor (DSP) or a Field Programmable Gate Array (FPGA). The processing unit 130 is connected to the sampling unit 120: Generate N positioning positions according to the N first digital signal values and the N second digital signal values, where N is a positive integer greater than 1. In this embodiment, the processing unit 130 may use a Coordinate Rotation Digital Computer (CORDIC) algorithm or an inverse trigonometric function algorithm according to the N first digital signal values and the N second digital signal values. Generate N positioning positions. Next, the processing unit 130 adds N positioning positions to the calculation group, and performs regression analysis on the positioning positions in the calculation group to obtain a regression curve.

After that, the processing unit 130 predicts the N + 1th prediction position according to the regression curve. That is, the N + 1th prediction position corresponds to the prediction position at the next sampling time point. Next, the processing unit 130 obtains the ideal position of the device under test 210 at the time point of the N + 1th predicted position according to the ideal position curve. In this embodiment, the ideal position curve may be generated according to at least a part of the N positioning positions in the calculation group.

Next, the processing unit 130 may use the error value between the N + 1th predicted position and the above-mentioned ideal position to correct the component to be tested 210. For example, the processing unit 130 may generate an output driving signal to the driving unit 220 according to the above error value. In this way, the driving unit 220 can correct the component to be tested 210 according to the driving signal described above, so as to improve the positioning accuracy of the encoder.

Further, after the component to be tested 210 is corrected by using the above error value, the processing unit 130 can further obtain an N + 1th positioning position. Then, the processing unit 130 may delete the first positioning position in the calculation group, and add the N + 1th positioning position to the calculation group to update the calculation group. At this time, the calculation group includes the second positioning position to the N + 1th positioning position.

After that, the processing unit 130 re-positions the positioning bits in the calculation group. Position (that is, the second positioning position to the N + 1th positioning position) for regression analysis to obtain a regression curve, and based on this regression curve, predict the N + 1th prediction position (in this case, the N + 1th prediction position Position is the N + 2th predicted position). Then, the processing unit 130 obtains the ideal position of the device under test 210 at the time point of the N + 1th prediction position (ie, the Nth + 2 prediction position) according to the ideal position curve. After that, the processing unit 130 may use the error value between the N + 1th prediction position (ie, the N + 2th prediction position) and the ideal position to correct the component to be tested 210.

Then, the processing unit 130 can further obtain the N + 1th positioning position (in this case, the N + 1th positioning position is the N + 2th positioning position). Then, the processing unit 130 may delete the first positioning position in the calculation group, and add the N + 1th positioning position to the calculation group to update the calculation group. At this time, the calculation group includes the third positioning position to the N + 2th positioning position.

After that, the processing unit 130 performs a regression analysis on the positioning positions in the calculation group (ie, the third positioning position to the N + 2 positioning position) to obtain a regression curve, and predicts the N + th position based on the regression curve. 1 prediction position (in this case, the N + 1th prediction position is the N + 3th prediction position). Then, the processing unit 130 obtains the ideal position of the device under test 210 at the time point of the N + 1th prediction position (ie, the Nth + 3th prediction position) according to the ideal position curve. After that, the processing unit 130 may use the error value between the N + 1th prediction position (ie, the N + 3th prediction position) and the ideal position to correct the component to be tested 210. The rest is analogous.

The storage unit 140 may be a non-volatile memory. The storage unit 140 is used to store the N positions corresponding to the above, the calculation formulas required to obtain various parameter values of the present invention, and the error values and error tables used to correct the positioning positions of the component 210 to be tested.

Further, the processing unit 130 includes calculation units 131 and 132 and a correction unit 133. The calculation unit 131 is connected to the sampling unit 120, and generates N positioning positions based on the N first digital signal values and the N second digital signal values, and adds the N positioning positions to the calculation group. The calculation unit 131 generates N positioning positions based on the N first digital signal values and the N second digital signal values through a coordinate rotation digital calculator algorithm or an inverse trigonometric function algorithm. The calculation unit 132 is connected to the calculation unit 131, receives the calculation group, and performs regression analysis on the positioning positions in the calculation group to obtain a regression curve.

The correction unit 133 is connected to the calculation unit 132 and the storage unit 140, and predicts the N + 1 prediction position based on the regression curve, and according to the ideal position curve, obtains the ideal of the DUT 210 at the time point of the N + 1 prediction position. Position, and use the error value between the N + 1th predicted position and the ideal position to correct the component to be tested 210. Then, the correction unit 133 also stores the error value between the N + 1th predicted position and the ideal position in the storage unit 140 to update the data stored in the storage unit 140, such as the error value and the error table.

Further, the processing unit 130 further includes a filter 134. The filter 134 is connected between the calculation unit 131 and the calculation unit 132, and is used for filtering the N positioning positions to filter out the noise of the N positioning positions and output the filtered N positioning positions to the computing module. Group 132. In this way, the regression curve generated by the calculation group 132 can be avoided from being disturbed by noise and causing an excessively large calculation error. In this embodiment, the filter 134 may be a low-pass filter (Low Pass Filter, LPF).

FIG. 2A is a schematic diagram of an ideal positioning position output of an encoder according to an embodiment of the present invention, and FIG. 2B is a diagram of an ideal positioning position output according to an embodiment of the present invention. The schematic diagram of the actual positioning position output of the encoder is described. When the DUT 210 (such as a motor) is running at a constant speed, the position change of the DUT 210 in a short time should show a linear change due to the inertia of the mechanism of the DUT 210, as shown in FIG. 2A.

However, as mentioned above, due to the error caused when the encoder is assembled with the device under test 210 or the external environment, the actual positioning position output of the encoder will have errors, as shown in Figure 2B. In order to improve the positioning position output accuracy of the encoder, the embodiment of the present invention will predict the output position of the encoder, and obtain an error between the predicted position and the corresponding ideal position according to an ideal position curve, and the component to be tested 210 is corrected to improve positioning accuracy.

FIG. 3 is a schematic diagram of a correspondence relationship between a positioning position, a predicted position, and a regression curve according to an embodiment of the present invention. In this embodiment, N is taken as an example of 18, for example. As shown in Fig. 3, "x" indicates the positioning position corresponding to the sampling time T1 to T18, and "o" indicates the predicted position at the fixed sampling time point T19.

Please refer to Figure 1 and Figure 3 together. In this embodiment, after obtaining 18 (ie N) positioning positions corresponding to the sampling time points T1 to T18 (ie, T1 to TN), the processing unit 130 further adds the above 18 (ie N) positioning positions to the calculation group. In the group. Next, the processing unit 130 performs regression analysis on 18 (ie, N) positioning positions in the calculation group to obtain a polynomial corresponding to the 18 (ie, N) positions (such as the regression curve 301 in FIG. 3). After that, the processing unit 130 may predict the predicted position corresponding to the sampling time point T19 (ie, TN + 1) according to the regression curve 301 as described above.

In one embodiment, the sampling time range of the N positioning positions is at least greater than one cycle of the first sine wave signal or the second sine wave signal. In addition, The order of the above polynomial is determined according to the movement state of the DUT 210 (motor). For example, when the operation of the DUT 210 is at a fixed speed or the speed variation is less than a predetermined range, the polynomial used to predict the positioning position of the next sampling time point TN + 1 may be a first-order polynomial.

In addition, when the acceleration of the operation of the DUT 210 occurs or the positioning position of the DUT 210 at each sampling time point is significantly different, the positioning of the predicted sampling time point TN + 1 used by the processing unit 130 The polynomial of the position is a quadratic polynomial. In this embodiment, the coefficient of the above polynomial can be obtained by a least squares algorithm. That is, the processing unit 130 may perform a regression analysis on the positioning positions in the calculation group through a least square algorithm to obtain a regression curve.

For example, since the change of the positioning position corresponding to each two sampling time points in FIG. 3 is not fixed, the polynomial used by the processing device 130 is a quadratic polynomial to obtain a regression curve. In this embodiment, the sampling time range of the regression curve 301 may be T1 ~ T18 (ie, TN), or only a predetermined time range corresponding to one cycle period. Then, after the regression curve 301 is obtained, the processing unit 130 may further obtain at least a part of the 18 (ie, N) positioning positions in the calculation group to obtain an ideal position curve.

For example, taking the positioning positions corresponding to the sampling time points T1 to T6 (that is, calculating at least a part of the N positioning positions of the group) as an example. The processing unit 130 may calculate changes in the positioning positions corresponding to the sampling time points T1 to T2, T2 to T3, T3 to T4, T4 to T5, and T5 to T6 respectively, and then divide the obtained positioning positions by 5 Time intervals to calculate the average distance moved by the device under test 210 in each time interval. Among them, the above 5 time intervals are sampling time points T1 to T6 The sampling time interval. Then, the processing unit 130 may obtain an ideal position curve according to the above average distance. In one embodiment, the speed of the device under test 210 remains unchanged when the device under test 210 operates under the same environmental conditions, such as a sampling time corresponding to an ideal position curve.

In addition, in another embodiment, the ideal position curve may be generated by a user-defined equation, that is, the ideal position curve is not directly related to the operating state of the device under test 210. Then, the processing unit 130 may obtain the corresponding sampling time point T19 (TN + 1) at the sampling time point T19 (TN + 1) corresponding to the 19th (ie, N + 1) predicted position according to the ideal position curve. The ideal location of the device under test 210.

Next, the processing unit 130 performs the 19th (N + 1) prediction based on the 19th (N + 1) th prediction position corresponding to the sampling time point T19 (TN + 1) (ie, "o" shown in FIG. 3). Each predicted position is subtracted from the corresponding ideal position to obtain the corresponding error value.

After that, the processing unit 130 can use this error value to correct the component to be tested 210. For example, the processing unit 130 can output a driving signal to the driving unit 220 according to the error value, so that the driving unit 220 can correct the operating speed of the device to be tested 210 according to the driving signal corresponding to the error value to improve the positioning accuracy. .

After the processing unit 130 corrects the component to be tested 210 using the calculated error value, the sampling unit 120 may further sample the 19th (N + 1) th positioning position corresponding to the sampling time point T19 (TN + 1), and 19 (N + 1) positioning positions are provided to the processing unit 130. Then, the processing unit 130 may add the 19th (N + 1) th positioning position to the calculation group, and add the first positioning position in the calculation group. Delete to update the calculation group. At this time, the N positioning positions of the calculation group include the 2nd to 19th positioning positions. The second positioning position is used as the first positioning position in the calculation group, and the 19th positioning position is used as the Nth positioning position in the calculation group.

After that, the processing unit 130 performs regression analysis again according to the positioning positions of the calculation group (ie, the 2nd to 19th positioning positions) to obtain a new regression curve, and predicts the 20th predicted position based on the regression curve. Next, the processing unit 130 obtains the ideal position of the device under test 210 at the time point of the 20th predicted position according to the ideal position curve. After that, the processing unit 130 may use the error value between the 20th predicted position and the corresponding ideal position to correct the component to be tested 210. The rest is analogous. In addition, by continuously reacquiring the regression curve, reacquiring the next predicted position, and reacquiring the ideal position at the time point corresponding to the next predicted position, and based on the error between the next predicted position and the corresponding ideal position, 210 performs calibration to make the actual operation status of the device under test 210 closer to the ideal state to improve the accuracy of its operation.

In addition, in an embodiment, the processing unit 130 may also store the regression curve 301 obtained according to the positioning positions corresponding to the sampling time points T1 to T18 in the storage unit 140, and store the regression curves 301 corresponding to the sampling time points T1 to T18. The positioning position is subtracted from the ideal position of the corresponding ideal position curve to obtain corresponding error values, and an error table is established using these error values. Then, the processing unit 130 can predict the next predicted position according to the regression curve 301 stored in advance after the device under test 210 is activated, or directly correct the device under test 210 according to the error value in the error table to reduce the calculation time.

In this embodiment, the processing unit 130 is further obtaining a regression curve. Before 301, the 18 positioning positions corresponding to the sampling time points T1 to T18 are filtered by the filter 134 to filter out noise generated by the positioning positions, so that the regression curve 301 generated by the calculation unit 132 will not receive noise. Impact. For example, as shown in FIG. 3, the positioning position corresponding to the sampling time point T14 is larger than the positioning position of the sampling time point T13. Therefore, the filtering process through the filter 134 can effectively eliminate the sampling time point. The noise of the positioning position corresponding to T14 smoothes the positioning position corresponding to the sampling time point T14, so that the corresponding regression curve 301 shown in FIG. 3 is not affected by the noise, so as to increase the accuracy of position correction.

FIG. 4 is a comparison diagram of a corrected position signal and an uncorrected position signal according to an embodiment of the present invention. As shown in Figure 4, the uncorrected position signal has larger error fluctuations than the corrected position signal, and the corrected position signal is close to the ideal straight line with a slope.

FIG. 5 is a flowchart of an instant correction method for an encoder according to an embodiment of the present invention. In step S502, the movement of the device under test is sensed to obtain a first sine wave signal and a second sine wave signal, wherein the phase of the first sine wave signal and the second sine wave signal are 90 degrees out of phase. In step S504, the first sine wave signal and the second sine wave signal are sampled to obtain N first digital signal values and N second digital signal values.

In step S506, N positioning positions are generated according to the N first digital signal values and the N second digital signal values. In step S508, N positioning positions are added to the calculation group. In step S510, regression analysis is performed on the positioning positions in the calculation group to obtain a regression curve. In step S512, according to the back Return to the curve and predict the N + 1th predicted position. In step S514, according to the ideal position curve, the ideal position of the device under test is obtained at the time point of the N + 1th predicted position, and the error value between the N + 1 predicted position and the ideal position is used. Make corrections. In this embodiment, the ideal position curve is generated according to at least a part of the positioning positions in the calculation group.

FIG. 6 is a flowchart of an instant correction method for an encoder according to an embodiment of the present invention. In step S502, the movement of the device under test is sensed to obtain a first sine wave signal and a second sine wave signal, wherein the phase of the first sine wave signal and the second sine wave signal are 90 degrees out of phase. In step S504, the first sine wave signal and the second sine wave signal are sampled to obtain N first digital signal values and N second digital signal values.

In step S506, N positioning positions are generated according to the N first digital signal values and the N second digital signal values. In step S508, N positioning positions are added to the calculation group. In step S602, the positioning position of the calculation group is filtered. In step S510, regression analysis is performed on the positioning positions in the calculation group to obtain a regression curve. In step S512, the N + 1th prediction position is predicted according to the regression curve. In step S514, according to the ideal position curve, the ideal position of the device under test is obtained at the time point of the N + 1th predicted position, and the error value between the N + 1 predicted position and the ideal position is used. Make corrections. In this embodiment, the ideal position curve is generated according to at least a part of the positioning positions in the calculation group.

In step S604, the N + 1th positioning position is obtained. In step S606, the first positioning position in the calculation group is deleted and the N + 1th positioning position is added to the calculation group to update the calculation group. Then, it progresses to step S602 In order to filter the positioning positions of the updated calculation group, re-obtain the regression curve, re-predict the N + 1th predicted position, re-acquire the ideal position corresponding to the time point of the predicted N + 1th predicted position, And continue to make corrections to the component under test.

It is worth noting that the order of the steps in FIG. 5 and FIG. 6 is for illustrative purposes only, and is not intended to limit the order of the steps in the embodiment of the present invention, and the order of the above steps can be changed by the user according to his needs. In addition, without departing from the spirit and scope of the present invention, additional steps may be added or fewer steps may be used.

In summary, the real-time correction method and system for an encoder disclosed in the embodiments of the present invention obtain a regression curve by obtaining N positioning positions corresponding to the component under test and performing regression analysis on the N positioning positions. , And then predict the N + 1th predicted position, then according to the ideal position curve, at the time point corresponding to the N + 1th predicted position, obtain the ideal position of the component to be tested, and use the N + 1th predicted position and the corresponding The error value between the ideal positions is corrected for the component under test. In addition, the embodiment of the present invention can further obtain the N + 1th positioning position to reacquire the regression curve, re-predict the N + 1th predicted position, and reacquire the ideal position at the time point corresponding to the predicted N + 1th predicted position. And continue to make corrections to the component under test. In this way, the accuracy of the positioning position of the encoder can be effectively maintained within a certain range, and the service life of the encoder can be extended, so as to increase convenience in use.

Although the present invention is disclosed as above with the examples, it is not intended to limit the scope of the present invention. Any person with ordinary knowledge in the technical field can make some modifications and retouches without departing from the spirit and scope of the present invention. Therefore this The scope of protection of the invention shall be determined by the scope of the attached patent application.

Claims (13)

An instant correction method for an encoder includes: sensing a motion of a device under test to obtain a first sine wave signal and a second sine wave signal, wherein the first sine wave signal and the second sine wave The phase of the signal is 90 degrees apart; the first sine wave signal and the second sine wave signal are sampled to obtain N first digital signal values and N second digital signal values; according to the N first digital signals Value and the N second digital signal values to generate N positioning positions; adding the N positioning positions to a calculation group; performing regression analysis on the positioning positions in the calculation group to obtain a regression curve ; Predicting an N + 1 prediction position based on the regression curve; and obtaining an ideal position of the device under test at the time point of the N + 1 prediction position according to an ideal position curve, and using the first The error value between the N + 1 predicted positions and the ideal position is used to correct the DUT. The instant correction method as described in the first item of the patent application scope, wherein the ideal position curve is generated according to at least a part of the positioning positions in the calculation group. The real-time correction method as described in item 1 of the scope of patent application, further includes: obtaining the N + 1th positioning position; deleting the first positioning position in the calculation group; and removing the N + 1th positioning position. Join the calculation group to update the calculation group; and perform regression analysis on the positioning positions in the calculation group to obtain the regression curve. The real-time correction method according to item 1 of the scope of patent application, wherein the step of generating the N positioning positions according to the N first digital signal values and the N second digital signal values includes: rotating a digital calculation through a standard An algorithm or an inverse trigonometric function algorithm generates the N positioning positions according to the N first digital signal values and the N second digital signal values. The real-time correction method according to item 1 of the scope of patent application, wherein performing regression analysis on the positioning positions in the calculation group to obtain the regression curve includes: performing a least squares algorithm on the calculation group A regression analysis is performed on the positioning positions to obtain the regression curve. The real-time correction method described in item 1 of the scope of the patent application, further includes: filtering the positioning positions before obtaining the regression curve. An instant correction system for an encoder includes: a sensing unit for sensing the movement of a device under test to obtain a first sine wave signal and a second sine wave signal, wherein the first sine wave signal A phase difference from the second sine wave signal by 90 degrees; a sampling unit that samples the first sine wave signal and the second sine wave signal to obtain N first digital signal values and N second digital signals And a processing unit that generates N positioning positions based on the N first digital signal values and the N second digital signal values, adds the N positioning positions to a calculation group, and calculates the calculation group Perform a regression analysis on these positioning positions to obtain a regression curve, predict an N + 1 prediction position based on the regression curve, and according to an ideal position curve, at the time point of the N + 1 prediction position , Obtaining an ideal position of the device under test, and correcting the device under test by using an error value between the N + 1th predicted position and the ideal position. The instant correction system according to item 7 of the scope of patent application, wherein the ideal position curve is generated according to at least a part of the positioning positions in the calculation group. The instant correction system according to item 8 of the patent application scope, wherein the processing unit includes: a first calculation unit that generates N positioning positions based on the N first digital signal values and the N second digital signal values And adding the N positioning positions to the calculation group; a second calculation unit that performs regression analysis on the positioning positions in the calculation group to obtain the regression curve; and a correction unit based on the regression curve , Predict an N + 1th predicted position, and obtain an ideal position of the device under test at the time point of the N + 1th predicted position according to an ideal position curve, and use the N + 1th predicted position The error value between the position and the ideal position corrects the DUT. The instant correction system as described in item 7 of the scope of patent application, wherein the processing unit further obtains the N + 1th positioning position, deletes the 1st positioning position in the calculation group, and deletes the N + 1th positioning position Positioning positions are added to the calculation group to update the calculation group, and regression analysis is performed on the positioning positions in the calculation group to obtain the regression curve, and the Nth prediction is re-predicted based on the regression curve. +1 prediction position, and according to the ideal position curve again, at the time point of the N + 1th prediction position, obtain the ideal position of the device under test, and use the N + 1th prediction position and the ideal The error value between the positions corrects the DUT. The real-time correction system as described in item 7 of the scope of patent application, wherein the processing unit further uses a standard rotation digital calculator algorithm or an inverse trigonometric function algorithm, according to the N first digital signal values and the Nth The two-digit signal value generates the N positioning positions. The real-time correction system described in item 7 of the scope of the patent application, wherein the processing unit further performs a regression analysis on the positioning positions in the calculation group through a least square algorithm to obtain the regression curve. The real-time correction system according to item 7 of the scope of patent application, wherein the processing unit performs filtering processing on the positioning positions before obtaining the regression curve.