Method for correcting measurement signals

Abstract

To correct measurement signals that are provided by at least one sensor unit.SOLUTION: Two processed measurement signals are generated on the basis of at least two currently provided measurement signals, from which two corrected measurement signals are generated using angle-independent arithmetic operations and at least one correction coefficient, from which a corrected angle is calculated and output. A plurality of at least two measurement signals is provided in advance in order to determine at least one correction coefficient, from which the two processed measurement signals are generated. A corresponding angular error is calculated on the basis of the two processed measurement signals and a reference angle, and the angular error is subjected to a discrete Fourier transform. At least one correction coefficient is determined and stored on the basis of coefficients of the discrete Fourier transform, and at least one correction coefficient is determined such that a remaining angular error in the corrected angle is smaller than an angular error in an angle based on the two processed measurement signals.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a method for correcting measurement signals, which may represent, for example, the rotational movement of an object captured by a sensor device or the current angular position of a moving body. The subject of the invention is also a sensor device configured to perform such a method. [Background technology]

[0002] The prior art discloses sensor devices used as rotational motion sensors for detecting the rotational motion of a moving object or as linear distance sensors for detecting the linear motion of a moving object. In this case, the current angular position of a moving object during rotational motion or the current position of a moving object during linear motion is not measured directly but is encoded into two orthogonal signals, commonly referred to as sine and cosine channels, which form a vector in the complex plane. The actual angular position of the moving object or the current position of the moving object is then calculated using an arctangent function, in which case Cartesian coordinates can essentially be converted into polar angles. The sine and / or cosine signals provided by the sensor device may contain errors, such as offsets, amplitude mismatches, orthogonality errors, nonlinearities, etc. These errors can be corrected, for example, analogically by an evaluation and control unit or digitally by a corresponding evaluation program. To minimize the angle error, the corresponding correction coefficients should be calculated or determined as accurately as possible. For this purpose, it is known to calculate the correction coefficients using, for example, a Fourier transform of the sine and / or cosine signals provided by the sensor. A distinction is made between the signal domain and the angle domain because the current angular position or current position is calculated and not directly measured. In this case, harmonic disturbances of a certain order in the signal domain will also lead to overtones in the angle domain, but of a different order. The causes of this type of error depend on the measurement principle and are most frequently caused by manufacturing tolerances or imperfections in the sensor design, for example, non-ideal magnetization or magnetic flux distribution in magnetic sensors.

[0003] German Patent Application No. DE 10260862 A1 discloses a method and a circuit arrangement for correcting an angle and / or distance measuring sensor device in which measurement signals with sine and cosine waveforms are evaluated. These measurement signals are obtained by scanning a moving measurement object. An angle error or a phase error of the measurement signal is corrected by deriving a constant for estimating and correcting the angle error or phase error and / or amplitude of the measurement signal from a plurality of measurement signals. The method is corrected by deriving a constant for estimating and correcting the angle error or phase error and / or amplitude of the measurement signal from a plurality of measurement signals.

[0004] German Patent Application No. DE 10 2004 029 815 A1 discloses a method and an arrangement for correcting an angle and / or distance measuring sensor device, in which sine- and cosine-shaped measurement signals obtained by scanning a moving measurement object are evaluated. For correcting angle or phase errors in the measurement signals, the method comprises a balancing method followed by a correction method. In the balancing method, correction parameters are provided, and in the correction method, corrected measurement value pairs are determined from each measurement value pair. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] DE 10260862 [Patent Document 2] German Patent Application Publication No. 102004029815 Summary of the Invention [Problem to be solved by the invention]

[0006] Disclosure of the Invention The method for correcting a measurement signal having the features of independent patent claim 1 has the advantage that correction factors can be calculated in such a way that the angular error is reduced, preferably minimized, after correction. Embodiments of the method can also be used with "difficult" measurement signals in which angular errors are caused by harmonics of sine and / or cosine signals. Correction factors calculated using conventional methods are corrupted by such harmonics and do not result in the smallest achievable value for the angular error.

[0007] Embodiments of the present invention allow the calculation or determination of correction factors that allow compensation of first and / or second electrical harmonic oscillations in the angle error, even if they are not caused by or solely caused by offset, amplitude mismatch, or orthogonality errors, but are instead caused, for example, by second or third electrical harmonic oscillations or overtones of the oscillator signal or the sine and / or cosine signals, by a discrete Fourier transform of the angle error. In this case, the measurement or oscillator signal may also be provided by a polyphase system with three or more measurement or oscillator signals, which can be transformed into the complex plane after a corresponding transformation (e.g., Clarke transform). If the periodicity is greater than the value "1", a distinction is made between a mechanical angle and an electrical angle that differs from the mechanical angle by a factor of the periodicity. [Means for solving the problem]

[0008] An embodiment of the present invention provides a method for correcting measurement signals, in which at least two measurement signals are provided by at least one sensor unit. Two processed measurement signals are generated based on the currently provided at least two measurement signals, from which two corrected measurement signals are generated using angle-independent arithmetic operations and at least one correction factor, from which a corrected angle is calculated and output. To determine the at least one correction factor, a plurality of at least two measurement signals are provided in advance. Two processed measurement signals are generated based on the previously provided plurality of at least two measurement signals. A corresponding angle error is calculated based on the two processed measurement signals and a reference angle, and a discrete Fourier transform is performed on the angle error. At least one correction factor is determined and stored based on the coefficients of the discrete Fourier transform. In this case, the at least one correction factor is determined so that the remaining angle error in the corrected angle is smaller than the angle error in the angle based on the two processed measurement signals.

[0009] Also proposed is a sensor device comprising at least one sensor unit and at least one evaluation and control unit, which is adapted to implement the method as described above.

[0010] Embodiments of the present invention introduce first and second electrical harmonic oscillations into the angle error, counteracting and canceling the corresponding harmonic oscillations present in the angle error. The amplitude and phase of the generated harmonic oscillations can be controlled by selecting the corresponding correction values ​​for offset, amplitude mismatch, and orthogonality. It is particularly advantageous that the calculated correction values ​​can be applied angle-independently, which avoids the evaluation of complex trigonometric function calculations at the time of correction, in contrast to known harmonic compensation methods. By calculating the correction coefficients from the discrete Fourier transform of the angle error, it is nevertheless possible to achieve a good quality of correction for first and second harmonic oscillations of the angle error, i.e., a small residual error, as is possible with more complex harmonic compensation. Angle-independent application here refers to the independence of arithmetic operations when applying the correction values, for example, from a corrected angle or from an angle based on two processed measurement signals. However, the use of trigonometric functions for correction is also possible. Angle-dependent, eg, harmonic, corrections can be performed additionally, eg, before or after the described angle-independent methods, or in parallel.

[0011] The evaluation and control unit in this example can be understood to mean an electrical assembly or circuit or an electrical device, such as a control device, that prepares, processes, or evaluates the provided sensor signals. The evaluation and control unit can thereby include, for example, an ASIC module (ASIC: Application Specific Integrated Circuit) or a microcontroller. The evaluation and control unit can have at least one interface that can be configured based on hardware and / or software. In the case of a hardware-based configuration, the interface can be, for example, part of the ASIC module. However, the interface can also be a dedicated integrated circuit or at least partially consist of discrete components. In the case of a software-based configuration, the interface can be, for example, a software module that resides adjacent to other software modules on a microcontroller. A computer program product having program code stored on a machine-readable carrier, such as a semiconductor memory, a hard disk memory, or an optical memory, and that is used to perform the evaluation and determine at least one correction factor when the program is executed by the evaluation and control unit is also advantageous.

[0012] A sensor unit is understood in the present case to mean a structural unit comprising at least one sensor element which directly or indirectly provides a physical variable or a change in a physical variable and preferably converts it into an electrical sensor signal, whereby, for example, magnetic and / or inductive sensor elements can be used.

[0013] Advantageous refinements of the method for correcting a measurement signal presented in independent patent claim 1 are possible by means of the measures and developments set out in the dependent claims.

[0014] Particularly advantageously, when processing the plurality of at least two previously provided measurement signals and / or the at least two currently provided measurement signals, a respective conversion and / or filtering of the at least two previously provided measurement signals and / or the at least two currently provided measurement signals can be performed. For example, a first processed measurement signal of the at least two previously provided measurement signals and a first processed measurement signal of the at least two currently provided measurement signals can be assigned to a sine channel based on a periodic sine function having a predetermined period. A second processed measurement signal of the at least two previously provided measurement signals and a second processed measurement signal of the at least two currently provided measurement signals can be assigned to a cosine channel based on a periodic cosine function having a predetermined period. Such a conversion can also include, for example, a "rudimentary" compensation for a known offset of the measurement signals, especially if it is determined due to the structure. The structural offset can be, for example, caused by an offset in the analog-to-digital conversion, especially in the case of non-differential signal transmission.

[0015] In an advantageous configuration of the method, the discrete Fourier transform can be performed on the cumulative sum of the individual angular errors calculated from two processed measurement signals, which are based on a previously provided plurality of at least two measurement signals. This has the advantage that the cumulative sum requires less memory capacity. Alternatively, the discrete Fourier transform can be applied to the sum of each angular error calculated from two processed measurement signals, which are based on a previously provided plurality of at least two measurement signals. This means that first all angular errors are calculated for the provided and processed plurality of at least two measurement signals, and then the discrete Fourier transform is performed.

[0016] In a further advantageous configuration of the method, the first coefficient of the discrete Fourier transform can be based on a fundamental oscillation. The second coefficient of the discrete Fourier transform can be based on a harmonic oscillation of order "p". The third coefficient of the discrete Fourier transform can be based on a harmonic oscillation of order "2p". In this case, the value "p" corresponds to the period of the first and second processed measurement signals.

[0017] In a further advantageous configuration of the method, the first correction value can be calculated as the average value of the angular errors of a plurality of at least two previously provided measurement signals corresponding to the real part of the first coefficient of the discrete Fourier transform.

[0018] In a further advantageous configuration of the method, second and third correction values ​​can be determined based on a second coefficient of the discrete Fourier transform, and these correction values ​​are suitable for compensating for a component of the angle error due to harmonic vibrations of order "p." Here, the second correction value can correspond to a calculated or estimated signal offset of the sine channel. The third correction value can correspond to a calculated or estimated signal offset of the cosine channel. Here, the second correction value can be additionally scaled using a first scaling factor based on the determined amplitude of the sine channel, and the third correction value can be additionally scaled using a second scaling factor based on the determined amplitude of the cosine channel.

[0019] In a further advantageous configuration of the method, fourth and fifth correction values ​​can be determined based on the third coefficient of the discrete Fourier transform. These correction values ​​are suitable for compensating for a first component of the angular error due to harmonic oscillations of order 2p. This first component can correspond, for example, to the imaginary part of the Fourier-transformed angular error of order 2p. In this case, the fourth correction value can correspond to the equivalent relative amplitude of the sine channel, and the fifth correction value can correspond to the equivalent relative amplitude of the cosine channel. These fourth and fifth correction values ​​can be simplified and combined into a common correction value representing the ratio of the two correction values. Furthermore, a sixth correction value can be determined based on the third coefficient of the discrete Fourier transform. This sixth correction value is suitable for compensating for a second component of the angular error due to harmonic oscillations of order 2p. This second component can correspond, for example, to the real part of the Fourier-transformed angular error of order 2p. The sixth correction value may represent the calculated orthogonality error.

[0020] In a particularly advantageous configuration of the method, the second and third correction values ​​can be taken into account when calculating the fourth, fifth, and / or sixth correction values. This allows components of the angular error caused by secondary disturbances of the signal offset to be compensated for when calculating the fourth, fifth, and sixth correction values, thereby further improving the reduction of the angular error. This can be particularly advantageous when the influence of higher-order harmonic vibrations in large offset cases becomes significant and may lead to incomplete compensation of the second-order harmonic vibrations. Because the angular error lacks information about the absolute signal amplitude, the second and third correction values ​​are each related to the signal offset by unit amplitude. For use in correcting the processed measurement signal, they can be scaled by the amplitude of the sine or cosine channel. For similar reasons, only the fourth correction value can be calculated or estimated, whereas the cosine channel can be arbitrarily selected as a reference, and the fifth correction value can be assigned the value "1." The scaling factors required for scaling by the amplitude of the sine or cosine channel can be estimated, for example, from the discrete Fourier transform of the sine or cosine signal itself. Alternatively, the signal amplitude can be estimated, for example, from the average value of the vector length calculated based on the vectors calculated from the sine and cosine signals of the sine channel.

[0021] In a further advantageous configuration of the method, the first corrected measurement signal can be generated on the basis of the processed first measurement signal, the processed second measurement signal, the second correction value, the third correction value, the fourth correction value, the fifth correction value and the sixth correction value, and the second corrected measurement signal can be generated on the basis of the processed second measurement signal, the third correction value and the fifth correction value.

[0022] Embodiments of the invention are illustrated in the drawings and explained in more detail in the following specification, in which identical reference numbers represent components or elements that perform the same or similar functions. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a schematic flow chart illustrating an embodiment of a method for correcting a measurement signal according to the present invention. [Figure 2] 2 is a schematic diagram of an embodiment of a sensor arrangement according to the invention when determining correction values ​​during execution of the method according to the invention of FIG. 1; [Figure 3] 2 is a schematic view of the sensor arrangement according to the invention when correcting the measurement signal during the execution of the method according to the invention of FIG. 1; DETAILED DESCRIPTION OF THE INVENTION

[0024] Embodiments of the invention 1 to 3, the illustrated embodiment of the method 100 according to the present invention for correcting measurement signals MS1, MS2, MS3 includes a step S200 in which at least two measurement signals MS1, MS2, MS3 are provided by at least one sensor unit 3. Based on the currently provided at least two measurement signals MS1, MS2, MS3, two processed measurement signals a1, b1 are generated in step S210, from which two corrected measurement signals ac, bc are generated in step S220 using angle-independent arithmetic operations and at least one correction factor O, K. In step S230, a corrected angle WK is calculated from the two corrected measurement signals ac, bc, and this corrected angle WK is output in step S240. Subsequently, the method returns to step S200.

[0025] To determine at least one correction factor O, K, in step S100, a plurality N of at least two measurement signals vMS1, vMS2, vMS3 are provided in advance. Based on the plurality N of at least two measurement signals vMS1, vMS2, vMS3 provided in advance, two processed measurement signals a, b are generated in step S110. Based on the two processed measurement signals a, b and a reference angle W, a corresponding angle error dW is calculated in step S120, and a discrete Fourier transform (DFT) is performed on the angle error dW in step S130. Based on the coefficients X[0], X[p], and X[2p] of the discrete Fourier transform (DFT), at least one correction factor O, K is determined and stored in step S140. In this case, the at least one correction factor O, K is determined so that the remaining angle error dW in the corrected angle WK is smaller than the angle error dW in the angle W based on the two processed measurement signals a1, b1.

[0026] As is further apparent from Figures 2 and 3, the illustrated embodiment of the sensor device 1 according to the invention comprises at least one sensor unit 3 and at least one evaluation and control unit 10 and is adapted to perform the method 100 according to the invention. In the illustrated embodiment, the sensor device 1 comprises only one sensor unit 3 and only one evaluation and control unit 10 with multiple function blocks for performing the method 100.

[0027] 2, the sensor unit 3 in the illustrated embodiment provides three measurement signals vMS1, vMS2, vMS3 for determining at least one correction factor O, K for a plurality of N samples or measurements. When processing a plurality N of the three previously provided measurement signals vMS1, vMS2, vMS3, a Clarke transformation is performed in the first transformation block 12 on each of the three previously provided measurement signals vMS1, vMS2, vMS3 into a first processed measurement signal a based on a periodic sine function with a preset period p and a second processed measurement signal b based on a periodic cosine function with a preset period p. In this case, the first processed measurement signal a is assigned to the sine channel 12.1, and the second processed measurement signal b is assigned to the cosine channel 12.2.

[0028] For example, by ignoring noise and considering only periodic disturbances, the processed measurement signals a and b can be written as Fourier series according to the following equations (1) and (2), respectively:

number

number

[0029] In this case, A0 is the signal offset and A p is the amplitude of the fundamental of order p of the sine channel 12.1. B0 is the signal offset, and B p is the amplitude of the fundamental of order p in the cosine channel 12.2. p is the phase angle of the fundamental wave of sine channel 12.1, and U p is the phase angle of the fundamental of the cosine channel 12.2. The following equation (3) defines the orthogonality error OF: OF p =V p -U p (3)

[0030] The electrical angle error dW is given by the following equation (4) and is calculated in a first calculation block 14 of the evaluation and control unit 10 based on the first processed measurement signal a of the sine channel 12.1 and the second processed measurement signal b of the cosine channel 12.2, taking into account a provided reference angle WR. The reference angle WR can be provided, for example, by the drive of the moving object whose angular position or position is to be determined. The evaluation and control unit 10 also performs an amplitude calculation in a second calculation block 15 based on the first processed measurement signal a of the sine channel 12.1 and the second processed measurement signal b of the cosine channel 12.2. In this case, for example, a vector can be calculated from the first processed measurement signal a of the sine channel and the second processed measurement signal b of the cosine channel. The signal amplitude of the sine channel 12.1 and / or the cosine channel 12.2 can then be calculated from the average value of the corresponding vector lengths. dW=arctan(a(W) / b(W))-pW (4)

[0031] Equation (4) above is only valid for a limited range of electrical angles W because the quadrants are ambiguous and may divide by zero. In practice, this problem is solved by a modified arctangent function with two arguments, known as atan2(a,b). In this case, the arctangent result is expanded (solved) to remove the effects of discontinuities.

[0032] The second transformation block 15 performs a Discrete Fourier Transform (DFT) on the angular error dW and calculates the coefficients X[0], X[p], X[2p] that are used to determine at least one correction coefficient O, K. From the amplitude-normalized Discrete Fourier Transform (DFT) of the angular error dW of the measurement over 360° mechanically, the Discrete Fourier Transform (DFT) coefficients X[0], X[p], X[2p] are calculated according to the following equations (5) and (6): X=DFT{dW} (5)

number

[0033] In this case, the real part of "X[0]" is equal to the mean value of the angular error dW. "X[k]" is equal to the amplitude of the sinusoid of the kth harmonic oscillation, where "i" is the imaginary unit, provided that the angular error dW is real. To calculate at least one correction coefficient O,K, the first coefficient X[0] of the discrete Fourier transform DFT based on the fundamental oscillation, the second coefficient X[p] of the discrete Fourier transform DFT based on the harmonic oscillation of order "p", and the third coefficient X[2p] of the discrete Fourier transform DFT based on the harmonic oscillation of order "2p" are used. In this case, the value "p" corresponds to the period p of the first and second processed measurement signals a, b.

[0034] In the illustrated embodiment of the method 100, a discrete Fourier transform DFT is performed on the cumulative sum of the individual angle errors dW calculated from the two processed measurement signals a, b, which are based on a plurality N of the three previously provided measurement signals vMS1, vMS2, vMS3.

[0035] In an alternative embodiment of the method 100, not shown, a discrete Fourier transform DFT is applied to the total of each angular error dW calculated from the two processed measurement signals a, b, which are based on a plurality N of the three previously provided measurement signals vMS1, vMS2, vMS3.

[0036] As further apparent from FIG. 2, the coefficients X[0] and X[p] of the discrete Fourier transform DFT and the result of the amplitude calculation in the second calculation block 15 are provided to a third calculation block 17. The amplitude calculation in the second calculation block 15 is based on a vector calculated from the first processed measurement signal a of the sine channel and the second processed measurement signal b of the cosine channel. The third calculation block 17 calculates a first correction value O1 as the average value of the angle error dW corresponding to the real part RE of the first coefficient X[0] of the discrete Fourier transform DFT according to the following equation (7): Each individual angle error dW is determined based on an angle W determined from a plurality N of at least two previously provided measurement signals vMS1, vMS2, vMS3 and a reference angle WR. The first correction value O1 can be interpreted as the average angular deviation between the measured angle W and the reference angle WR, which may be caused, for example, by the sensor unit 3. O1=Re{X[0]} (7)

[0037] Based on the second coefficient X[p] of the discrete Fourier transform DFT, the third calculation block 17 calculates the second correction value O2 according to the following equation (8) and the third correction value O3 according to the following equation (9). Since the angular error dW does not contain information about the absolute signal amplitude, the second correction value O2 and the third correction value O3 are each related to a signal offset by unit amplitude. For use in correcting the processed measurement signals a1, b1, the scaled second correction value sO2 is additionally scaled by a first scaling factor S1 in the illustrated embodiment according to the following equation 8A, which is based on the amplitude determined for the sine channel 12.1 in the second calculation block 15. The scaled third correction value sO3 is, in the illustrated embodiment, additionally scaled using a second scaling factor S2 according to the following equation 9A, which is based on the amplitude determined for the cosine channel 12.2 in the second calculation block 15: O2=Re{X[p]e -i(O1)} (8) O3=Im{X[p]e -i(O1)} (9) sO2=(Re{X[p]e -i(O1)})*S1 (8A) sO3=(Im{X[p]e -i(O1)})*S2 (9A)

[0038] The second correction value O2 or the scaled second correction value sO2 and the third correction value O3 or the scaled third correction value sO3 are suitable for compensating for the component of the angle error dW due to harmonic vibrations of order “p”.

[0039] As further apparent from FIG. 2 , the coefficients X[0] and X[2p] of the discrete Fourier transform DFT are provided to a fourth calculation block 18. In the illustrated embodiment, the fourth calculation block 18 calculates a fourth correction value K4′ based on the third coefficient X[2p] of the discrete Fourier transform DFT according to the following equation (10): In the illustrated embodiment, the fourth calculation block 18 also takes into account the second correction value O2 and the third correction value O3 from the third calculation block 17 when calculating the improved fourth correction value K4 according to the following equation (10A). The fourth calculation block 18 also determines the fifth correction value K5 to be the value “1.”

number

number

[0040] The fourth correction value K4' or the improved fourth correction value K4 and the fifth correction value K5 are suitable for compensating for a first component of the angle error dW due to harmonic vibrations of order "2p", which corresponds to the imaginary part of order "2p" of the Fourier transformed angle error dW. The fourth and fifth correction values ​​K4 and K5 can also be combined into a common correction value KG, which represents the ratio of the two correction values ​​(K4 / K5).

[0041] 2, the fourth calculation block 18 calculates the sixth correction value K6' based on the third coefficient X[2p] of the discrete Fourier transform DFT according to equation 11. In the illustrated embodiment, the fourth calculation block 18 also takes into account the second correction value O2 and the third correction value O3 from the third calculation block 17 when calculating the improved sixth correction value K6 according to equation (11A). K6'=2Re{X[2p]e -i2(O1)} (11) K6=2(Re{X[2p]e -i2(O1)}+(O2)(O3)) (11A) vK6=K6-((1 / 12)(K6) 3 ) (11B)

[0042] The sixth correction value K6' or improved sixth correction value K6 is suitable for compensating for the second component of the angle error dW due to harmonic vibration of order 2p, which corresponds to the real part of the Fourier transformed angle error dW of order 2p. Additionally, the cube of the improved sixth correction value K6 can be used when calculating the further improved sixth correction value vK6 according to equation (11B). This can achieve a smaller residual error in the case of a large orthogonality error.

[0043] In calculating the second, third, fourth, improved fourth, fifth, sixth, improved sixth, and further improved sixth correction values ​​O2, sO2, O3, sO3, K4', K4, K5, K6', K6, and vK6, the first correction value O1 is used in each case to compensate for the influence of the average angular deviation. This can alternatively be done by already compensating the calculation of the angular error dW with a correction value similar to the first correction value O1, for example, determined by an additional preceding calculation of the angular error dW.

[0044] As is further apparent from FIG. 2, the correction values ​​O1, O2, sO2, O3, sO3, K4, K4', K5, K6, K6', and vK6 are stored in the memory unit 19.

[0045] 3, the sensor unit 3 provides, in the illustrated embodiment, three current measurement signals MS1, MS2, MS3 based on a current scan or measurement. When processing these three current measurement signals MS1, MS2, MS3, the first conversion block 12 performs a Clarke transformation of the three currently provided measurement signals MS1, MS2, MS3 and converts them into a first processed measurement signal a1 based on a periodic sine function with a preset period p and a second processed measurement signal b1 based on a periodic cosine function with a preset period p. In this case, the first processed measurement signal a1 is assigned to the sine channel 12.1 and the second processed measurement signal b1 is assigned to the cosine channel 12.2.

[0046] 3, in the illustrated embodiment, the correction block 20 generates the first corrected measurement signal ac based on the processed first measurement signal a1, the processed second measurement signal b1, the scaled second correction value sO2, the scaled third correction value sO3, the fourth correction value K4, the fifth correction value K5, and the sixth correction value K6 according to the following equation (12):

number

[0047] Based on the processed second current measurement signal b1, the scaled third correction value sO3 and the fifth correction value K5, the correction block 20 generates a second corrected measurement signal bc according to the following equation (13), where the correction values ​​sO3 and K5 are provided by the memory unit 19: bc=(b1-sO3) / K5 (13)

[0048] Of course, the constant components in equation (12) can also be pre-calculated to improve calculation speed and stored in memory unit 19 in addition to or instead of the correction values. This makes it possible to avoid trigonometric functions such as "sin(K6)" or "cos(K6)" during execution. It is further advantageous to first calculate the corrected second signal bc according to equation (13) and then use the component (b1-s03) / K5 in equation (12) directly in equation (12) for the calculation of the first corrected measurement signal ac without having to calculate it again.

[0049] In the output block 22, a corrected angle WK is calculated from these corrected measurement signals a1, b1 and output. In this case, the residual angle error dW in the corrected angle WK is smaller than the angle error dW in the angle based on the two processed but uncorrected measurement signals a1, b1. Additionally, the angular offset in the corrected angle WK can be compensated for by the first correction value O1. Furthermore, the angular offset affected by orthogonality can be additionally compensated for by the sixth correction value, the improved sixth correction value, and the further improved correction values ​​K6', K6, vK6.

Claims

1. A method (100) for correcting measurement signals (MS1, MS2, MS3), comprising: At least two measurement signals (MS1, MS2, MS3) are provided by at least one sensor unit (3), two processed measurement signals (a1, b1) are generated based on at least two currently provided measurement signals (MS1, MS2, MS3), from which two corrected measurement signals (a c, bc) are generated using angle-independent arithmetic operations and at least one correction factor (O, K), from which a corrected angle (W K) is calculated and output; a plurality (N) of at least two measurement signals (vMS1, vMS2, vMS3) are provided in advance to determine the at least one correction factor (O, K), two processed measurement signals (a, b) are generated based on a plurality (N) of at least two measurement signals (vMS1, vMS2, vMS3) provided in advance; Based on the two processed measurement signals (a, b) and the reference angle (WR), a corresponding angle error (dW) is calculated, and a discrete Fourier transform (DFT) is performed on the angle error (dW); The at least one correction coefficient (O, K) is determined and stored based on coefficients (X[0], X[p], X[2p]) of the discrete Fourier transform (DFT); The method (100) wherein the at least one correction coefficient (O, K) is determined such that the residual angle error (dW) in the corrected angle (WK) is smaller than the angle error (dW) in the angle based on the two processed measurement signals (a1, b1).

2. 2. The method (100) according to claim 1, wherein when processing the plurality (N) of at least two previously provided measurement signals (vMS1, vMS2, vMS3) and / or the at least two currently provided measurement signals (MS1, MS2, MS3), a transformation and / or filtering is performed on each of the at least two previously provided measurement signals (vMS1, vMS2, vMS3) and / or the at least two currently provided measurement signals (MS1, MS2, MS3).

3. a first processed measurement signal (a) of the at least two previously provided measurement signals (vMS1, vMS2, vMS3) and a first processed measurement signal (a1) of the at least two currently provided measurement signals (MS1, MS2, MS3) are each assigned to a sine channel (12.1) based on a periodic sine function having a preset period (p); 3. The method (100) according to claim 1 or 2, wherein the second processed measurement signal (b) of the at least two previously provided measurement signals (vMS1, vMS2, vMS3) and the second processed measurement signal (b1) of the at least two currently provided measurement signals (MS1, MS2, MS3) are each assigned to a cosine channel (12.2) based on a periodic cosine function having a preset period (p).

4. 4. The method (100) according to claim 1, wherein the Discrete Fourier Transform (DFT) is performed on a cumulative sum of individual angular errors (dW) calculated from two processed measurement signals (a, b), which are based on the plurality (N) of at least two previously provided measurement signals (vMS1, vMS2, vMS3).

5. 4. The method (100) according to any one of claims 1 to 3, wherein the Discrete Fourier Transform (DFT) is applied to the total of each angular error (dW) calculated from two processed measurement signals (a, b), which are based on the plurality (N) of at least two measurement signals (vMS1, vMS2, vMS3) provided in advance.

6. 6. The method (100) according to claim 4 or 5, wherein a first coefficient (X[0]) of the Discrete Fourier Transform (DFT) is based on a fundamental oscillation, a second coefficient (X[p]) of the Discrete Fourier Transform (DFT) is based on a harmonic oscillation with order (p), and a third coefficient (X[2p]) of the Discrete Fourier Transform (DFT) is based on a harmonic oscillation with order (2p), the value "p" corresponding to the period (p) of the first and second processed measurement signals (a, b).

7. 7. The method (100) according to claim 6, wherein a first correction value (O1) is calculated as the average value of the angular errors (dW) of the plurality (N) of at least two pre-provided measurement signals (vMS1, vMS2, vMS3) corresponding to the real part of the first coefficient (X[0]) of the Discrete Fourier Transform (DFT).

8. 8. The method (100) according to claim 6 or 7, wherein a second correction value (O2) and a third correction value (O3) are determined based on the second coefficient (X[p]) of the discrete Fourier transform (DFT), and the second correction value (O2) and the third correction value (O3) are suitable for compensating for a component of the angle error (dW) due to harmonic vibrations of order (p).

9. 9. The method (100) of claim 8, wherein the second correction value (O2) is additionally scaled with a first scaling factor (S1) based on the amplitude determined for the sine channel (12.1), and the third correction value (O3) is additionally scaled with a second scaling factor (S2) based on the amplitude determined for the cosine channel (12.2).

10. 10. The method (100) according to any one of claims 6 to 9, wherein a fourth correction value (K4) and a fifth correction value (K5) are determined based on the third coefficient (X[2p]) of the discrete Fourier transform (DFT), and the fourth correction value (K4) and the fifth correction value (K5) are suitable for compensating for a first component of an angle error (dW) due to harmonic vibrations of order (2p).

11. 11. The method (100) according to claim 6, wherein a sixth correction value (K6) is determined based on the third coefficient (X[2p]) of the discrete Fourier transform (DFT), the sixth correction value (K6) being suitable for compensating for a second component of the angle error (dW) due to harmonic vibrations of order (2p).

12. 12. The method (100) according to claim 10 or 11, wherein the second correction value (O2) and the third correction value (O3) are taken into account when calculating the fourth correction value (K4) and / or the fifth correction value (K5) and / or the sixth correction value (K6).

13. 13. The method (100) according to any one of claims 10 to 12, wherein the cosine channel (12.2) is selected as a reference and the fifth correction value (K5) is assigned the value "1".

14. 14. The method (100) according to any one of claims 10 to 13, wherein the first corrected measurement signal (ac) is generated based on the processed first measurement signal (a1), the processed second measurement signal (b1), the second correction value (O2), the third correction value (O3), the fourth correction value (K4), the fifth correction value (K5) and the sixth correction value (K6).

15. 15. The method (100) according to any one of claims 10 to 14, wherein a second corrected measurement signal (bc) is generated based on the processed second measurement signal (b1), the third correction value (O3) and the fifth correction value (K5).

16. A sensor device (1) comprising at least one sensor unit (3) and at least one evaluation and control unit (10), the sensor device (1) being configured to perform a method (100) according to any one of claims 1 to 15.

Images