阅读说明：本技术 用于转速传感器的转速传感器信号的偏移校准的方法、系统、计算机程序 (Method, system, computer program for offset calibration of a tachometer signal of a tachometer ) 是由 J·古特 H·贝克尔 于 2019-09-26 设计创作，主要内容包括：一种用于转速传感器(1)的转速传感器信号(11)的偏移校准的方法,其中,在第一步骤中,求取转速传感器(1)处于静止状态(15)；在该第一步骤之后的第二步骤中,根据静止状态(15)中所测量的转速传感器的转速传感器测量值(x1)来确定滤波器参数(a)；在该第二步骤之后的第三步骤中,借助滤波器参数(a)确定经滤波的转速传感器测量值其中,借助经滤波的转速传感器测量值确定偏移(40)。(A method for offset calibration of a tachometer signal (11) of a tachometer sensor (1), wherein in a first step, it is determined that the tachometer sensor (1) is in a stationary state (15); in a second step following the first step, a filter parameter (a) is determined from a rotational speed sensor measurement (x1) of the rotational speed sensor measured in a stationary state (15); in a third step following the second step, the filtered measured values of the rotational speed sensor are determined using the filter parameters (a) Wherein, by means ofFiltered tachometer sensor measurements An offset is determined (40).)

1. A method for offset calibration of a tacho sensor signal (11) of a tacho sensor (1), wherein,

in a first step, it is determined that the rotational speed sensor (1) is in a stationary state (15);

in a second step, following the first step, a filter parameter (a) is determined from a rotational speed sensor measurement (x1) of the rotational speed sensor (1) measured in the stationary state (15);

in a third step following the second step, the filtered rotational speed sensor measured values are determined by means of the filter parameters (a)

Wherein the filtered rotational speed sensor measurement value is usedAn offset (40) is determined.

2. The method according to claim 1, wherein in the first step the rotational speed sensor (1) is determined to be in the stationary state (15) by means of an estimated mean value of the rotational speed sensor measured values (10) and/or by means of an estimated fluctuation value of the rotational speed sensor measured values (10), wherein the estimated fluctuation value preferably corresponds to an estimated variance.

3. Method according to claim 2, wherein the evaluation in the first step results in the rotation speed sensor (1) being in the stationary state if the estimated mean value is smaller than a first threshold value and/or the estimated variance is smaller than a second threshold value.

4. The method according to any one of the preceding claims, wherein the rotation speed sensor measurements measured in the stationary state (15) comprise only rotation speed sensor measurements (10) of: the rotational speed sensor measurement value is measured since the rotational speed sensor (1) is in the stationary state (15), wherein the rotational speed sensor measurement value (x1) measured in the stationary state (15) is preferably determined in a third step in the determination of the filtered rotational speed sensor measurement valueIn particular in the determination of the filtered rotational speed sensor measurement valueIs used in the case of (1).

5. Method according to one of the preceding claims, wherein in the second step the filter parameter (a) is determined as a function of a first intermediate parameter (δ 1) and a second intermediate parameter (a2), wherein the first intermediate parameter (a1) is proportional to the inverse of the number of speed sensor measurements (x1) measured in the stationary state (15), wherein the second intermediate parameter (a2) depends on fluctuations of the speed sensor measurements measured in the stationary state (15), wherein the fluctuations are preferably the estimated variance.

6. The method of claim 5, wherein the filter parameter (a) corresponds to a maximum of the first and second intermediate parameters (a1, a 2).

7. A method according to any of the preceding claims, wherein in the third step further based on previously filtered tacho sensor measurementsAnd the current tachometer sensor measurement (x n)]) Determining the filtered rotational speed sensor measurement value, in particular by means of exponential smoothing

8. Method according to one of the preceding claims, wherein in a fourth step following the third step the filtered tachometer sensor measurement value is reduced by means of a decimation filtering device (20)Thereby producing filtered tachometer sensor measurements (21) with reduced output rate.

9. Method according to one of the preceding claims, wherein, after the third step and in particular in a fifth step after the fourth step, the filtered tachometer measurement values are usedOr the offset (40) is determined by means of the filtered tachometer measurement (21) of the output rate reduction and further by means of a smoothing filter (30).

10. System (100) for offset calibration of a tachometer signal (11) of a tachometer sensor (1), wherein the system (100) is designed in such a way that the tachometer sensor is determined to be in a stationary state (15),

wherein the system (100) is designed in such a way that a filter parameter (a) is determined as a function of a rotational speed sensor measured value (x1) of the rotational speed sensor (1) measured in the stationary state (15),

wherein the system (100) is designed in such a way that the filtered rotational speed sensor measured values are determined by means of the filter parameters (a)

Wherein the system (100) is designed in such a way that the filtered rotational speed sensor measurement values are usedAn offset (40) is determined.

11. A computer program comprising instructions which, when executed by a computer and/or the system (100) according to claim 10, cause the computer and/or the system (100) to carry out the method according to any one of claims 1 to 9.

Technical Field

The invention relates to a method for offset calibration of a tachometer signal of a tachometer. The invention also relates to a system for calibrating an offset of a tacho sensor signal of a tacho sensor and a computer program comprising instructions.

Background

Rotational speed sensors, in particular rotational speed sensors with microelectromechanical systems, are used in a wide variety of fields. It is used, for example, in the field of electronic consumer products as a component of a smartphone or a so-called "wearable device". Here, it can be used to determine the orientation of the device in space. For this purpose, the tachometer signal is usually integrated. In the motionless state of the tacho sensor, the tacho sensor signal should be zero. Such tachometer sensors typically have a manufacturing-defined offset in their signal. This leads to systematic errors in the integration of the tacho sensor signal, in particular to drifts which have a negative effect on the orientation estimate. In this case, an orientation error of 180 ° can occur rapidly within a few minutes. In order to avoid such drift, it is necessary to correct the rotational speed offset. Since the offset can change during the operation of the sensor and during its service life, for example due to the influence of temperature, a factory-side pre-calibration during the manufacture of the sensor generally does not lead to the desired avoidance of the offset.

Methods for calibrating the rotational speed offset are known from the prior art.

US 8548766B 2, for example, describes a sensor platform with at least one gyroscope, an acceleration sensor and a magnetometer. The compensation for the gyroscope is determined by means of the measured values of the acceleration sensor and the magnetometer. US 20110301900 a1 describes a look-up table with temperature values and associated zero rotational speed offset.

However, the known methods have the disadvantage that they are only partially difficult and expensive to implement and are premised, for example, on the presence of a plurality of different sensor types on a chip system. On the other hand, with many known systems, effective offset correction for the rotation rate sensor in a wide variety of application areas is only difficult to achieve or cannot be achieved at all, for example rotation rate sensors in cell phones, which are typically placed without movement at a table or the like, whereas rotation rate sensors in wearable devices, even in their rest position, are still in significant noise due to slight movements of the wearer.

Disclosure of Invention

The object of the present invention is to provide a method for offset calibration of a tachometer signal of a tachometer, which method makes it possible to achieve rapid and reliable offset correction for a wide range of fields of application of the tachometer.

The method according to the invention for offset calibration of a tachometer signal of a tachometer sensor according to the main claim has the following advantages over the prior art: the filter parameters are adapted to the stationary state conditions of the rotation speed sensor. Thus, different filter parameters can be determined and used for a rotational speed sensor which (almost) does not move at all in the stationary state, for example, than for a sensor which is subject to slight movements and fluctuations even in the stationary state (Ruhezustand). For a single sensor, the filter can also be advantageously adapted to the currently specified conditions of the standstill state. Accordingly, it is possible to achieve a time-matching of the filter to the quiescent state conditions and thus an optimization is possible. Thus, for example, a reduced time can be used for the offset for a (completely) motionless sensor with low noise, whereby a more precise correction and adaptation to changing temperature conditions can be achieved. For sensors that are still subject to significant movement even in the stationary state, the time for offset determination can be increased by determining the filter parameters in order to better compensate for the added noise and also to obtain a suitable offset correction.

The offset determined according to the invention can be used in particular subsequently for offset correction or offset calibration of the tachometer signal.

According to one embodiment of the invention, an offset of the speed sensor signal can also be understood as a misalignment (Versatz) of the speed sensor signal.

The tacho signal comprises, in particular, a plurality of tacho sensor measurement values which are determined or generated and output by the tacho sensor in a sequence at a rate (sampling rate).

The rotation speed sensor may have one or more measuring shafts, in particular one, two or three shafts. The tachometer signal and the tachometer measurement value can accordingly each comprise the values of all measuring axes of the sensor.

Advantageous embodiments and embodiments result from the dependent claims.

According to one embodiment of the invention, it can be advantageously achieved in a reliable manner that the rotational speed sensor is in a stationary state, i.e. does not move or hardly moves, by: in a first step, the stationary state of the speed sensor is determined by means of an estimated mean value of the measured values of the speed sensor and/or by means of an estimated fluctuation value of the measured values of the speed sensor, wherein the estimated fluctuation value preferably corresponds to the estimated variance. The estimated fluctuation value is preferably an estimated maximum variance or fluctuation of the measured value of the rotation speed sensor. In particular, it is conceivable to derive or estimate the estimated mean value and/or the estimated fluctuation value (or the estimated variance) by means of a smoothing filter. In this case, it is conceivable that the estimated mean value and/or the estimated variance do not exactly correspond to the mean value of the measured values and/or the variance of the measured values.

According to one embodiment of the invention, a particularly advantageous standstill state determination can be achieved by: in a first step, the rotational speed sensor is determined to be in a standstill if the estimated mean value is smaller than a first threshold value and/or the estimated variance is smaller than a second threshold value. In this case, the estimated mean value and the estimated variance (or maximum fluctuation) are preferably formed for a number of (most recent) rotational speed measured values by means of a smoothing filter. Particularly preferably, it is conceivable that a standstill state is determined if the estimated mean value is smaller than a first threshold thr1 and the estimated variance is smaller than a second threshold thr 2. Preferably, the following steps of the method are carried out only in the stationary state of the rotation speed sensor.

According to one embodiment of the invention, only the measured values measured when the rotational speed sensor is in the standstill state can be taken into account during the offset correction and in particular during the determination of the filter parameters in the following manner: the measured rotational speed sensor measured values in the stationary state comprise only the measured rotational speed sensor measured values since the rotational speed sensor was in the stationary state, wherein the measured rotational speed sensor measured values in the stationary state are preferably used when determining the filtered rotational speed sensor measured values in the third step, in particular when determining the filtered rotational speed sensor measured values.

According to one embodiment of the invention, a particularly advantageous adaptation of the filter parameters (and therefore of the correction method as a whole) to the given conditions in the standstill state and to external influences on the rotational speed sensor can be achieved by: in a second step, a filter parameter is determined as a function of a first intermediate parameter and a second intermediate parameter, wherein the first intermediate parameter is proportional to the inverse of the number of measured rotational speed sensor values in the stationary state, wherein the second intermediate parameter depends on fluctuations of the measured rotational speed sensor values in the stationary state, wherein the fluctuations are preferably the estimated variance.

In particular, according to one specific embodiment, first intermediate parameter a1 may be determined as follows:

where # Z is the number of rotation speed sensor measurement values measured in the stationary state or the number of samples used for calibration. The second intermediate parameter a2 can be preferably determined in particular in such a way that the range from 0 to the second threshold thr2, i.e. in particular from 0 to the maximum variance or fluctuation, is divided into a plurality of regions or subranges. For example, an interval may be provided for each typical noise level that occurs in different wearable device calibration scenarios. Each interval is assigned a fixed value of the second intermediate parameter. Thus, the second intermediate parameter obtains the values assigned to the following intervals: the (maximum) fluctuation or variance found in the stationary state falls within this interval.

According to one embodiment of the invention, the filter parameter can be determined particularly advantageously by the filter parameter corresponding to the maximum value of the first and second intermediate parameters, wherein both fluctuations in the measured values of the speed sensor measured in the standstill state and the number of measured values of the speed sensor measured in the standstill state can be taken into account. Thus, for example, it is possible to carry out a particularly rapid and precise offset calibration not only for speed sensors which have only low noise in the stationary state, but also for speed sensors which have increased noise, a relatively slower but still precise offset calibration.

According to one embodiment of the invention, a particularly advantageous offset calibration can be achieved by: in a third step, the filtered rotational speed sensor measured value is determined further on the basis of the previous filtered rotational speed sensor measured value and the current rotational speed sensor measured value, in particular by means of exponential smoothing.

It is in particular conceivable to use the following relationship in the third step:

wherein the content of the first and second substances,is a filtered tachometer sensor measurement, where a is a filter parameter, where,is a previously filtered tachometer sensor measurement, where x n]Is the current tachometer sensor measurement, where n is the tachometer sensor measurement or the filtered tachometer sensor measurement number. In this case, the class (for determining the filtered rotational speed sensor measurement value) is preferably used in the iteration of the preceding third step or of the second and third steps or of the first, second and third stepsA previous filtered tachometer sensor measurement is determined in a similar manner. Accordingly, the parameter, i.e. the filter parameter a, can be used in the third step for the exponential smoothing. For the third step, in principle, other filtering methods can also be considered.

According to one embodiment of the invention, a reduced offset output rate can be achieved and is independent of the measurement or sampling rate of the rotational speed sensor measurement values in particular by: after the third step, the output rate, in particular the sampling rate, of the filtered tachometer measurement values is reduced by means of a decimation filter device (dezimatoreichtung), so that filtered tachometer measurement values with a reduced output rate are produced.

According to one embodiment of the invention, additional smoothing in the offset calibration can be achieved by: in a fifth step after the third step and in particular after the fourth step, the offset is determined by means of the filtered rotational speed sensor measured values or by means of the filtered rotational speed sensor measured values with a reduced output rate and further by means of a smoothing filter. For example, slight fluctuations in the estimated value (i.e., the filtered rotational speed sensor measured value or the filtered rotational speed sensor measured value with a reduced output rate) can be reduced. In this case, the offset preferably corresponds to a value obtained from the smoothing filter. The smoothing filter may consider various methods such as exponential smoothing. In the smoothing in the fifth step, a plurality of filtered tachometer measurement values (from a plurality of iterations of the third step and/or the fourth step) or filtered tachometer measurement values with a reduced output rate are typically used.

According to one embodiment of the invention, it is conceivable for the third step, preferably the second and third steps, particularly preferably the first, second and third steps, further preferably the first, second, third and fourth steps, and even further preferably the first, second, third, fourth and fifth steps to be repeated a plurality of times, in particular in such a way that a plurality of filtered rotational speed sensor measured values are generated in the third step. Thus, continuous matching offset calibration can be achieved.

A further subject matter of the invention is a system for offset calibration of a tachometer signal of a tachometer, wherein the system is designed in such a way that the tachometer is determined to be in a stationary state, wherein the system is designed in such a way that a filter parameter is determined from a tachometer measurement value of the tachometer measured in the stationary state, wherein the system is designed in such a way that a filtered tachometer measurement value is determined by means of the filter parameter, wherein the system is designed in such a way that an offset is determined by means of the filtered tachometer measurement value.

Another subject matter of the invention is a computer program comprising instructions which, when executed by a computer and/or a system according to an embodiment of the invention, cause the computer and/or the system to carry out a method according to an embodiment of the invention.

The features, configurations and advantages described in connection with the method according to the invention for offset calibration of a tacho sensor signal of a tacho sensor or in connection with an embodiment of the method according to the invention can be applied to the system according to the invention and the computer program according to the invention. The features, configurations and advantages described in connection with the system according to the invention, the computer program according to the invention or in connection with an embodiment of the system according to the invention or in connection with an embodiment of the computer system according to the invention can be applied to the method according to the invention for offset calibration of a tacho sensor signal of a tacho sensor.

Drawings

Embodiments of the invention are illustrated in the drawings and are further described in the following description.

Fig. 1 shows a schematic block diagram of a system for offset calibration of a tachometer signal of a tachometer according to an embodiment of the present invention.

FIG. 2 shows a schematic block diagram of a portion of a system for offset calibration according to an embodiment of the present invention.

Detailed Description

In the different figures, identical parts are provided with the same reference numerals throughout and are therefore usually also referred to or referred to once, respectively.

Fig. 1 shows a schematic block diagram of a system 100 for offset calibration of a tachometer signal 11 of a tachometer sensor 1 according to an embodiment of the present invention. The tachometer sensor 1 provides a tachometer sensor signal 11 comprising a sequence of tachometer sensor measurements 10. This is provided for the monitoring device 14 or the verification device 14. The monitoring device 14 determines whether the speed sensor 1 is in a stationary state 15 or in a non-stationary state 16 or in a moving state 16 on the basis of the acquired speed sensor measurement values 10. For this purpose, the monitoring device 14 preferably estimates the estimated mean value of the measured value 10 of the rotational speed sensor and the estimated fluctuation value, in particular the variance, of the measured value 10 of the rotational speed sensor by means of a smoothing filter. If the estimated average value is smaller than the first threshold thr1 and the estimated variance is smaller than the second threshold thr2, the monitoring device 14 determines that the rotation speed sensor 1 is in the stationary state. The first and/or second threshold values thr1, thr2 are determinable and may be selected corresponding to requirements. If the rotational speed sensor 1 is found to be in the non-stationary state 16, no further calibration steps are carried out. Conversely, if the rotational speed sensor 1 is found to be in the stationary state 15, the calibration is continued. The filter parameter a is determined by means of the filter parameter device 17 from the measured value x1 of the rotational speed sensor 1 measured in the stationary state 15. The speed sensor measurement x1 measured in the stationary state 15 here includes only such speed sensor measurements: the tachometer sensor measurement value is measured since the tachometer sensor 1 was in a stationary state 15 or transitioned to the stationary state. Fig. 2 illustrates one possible exemplary embodiment of the determination of the filter parameters a or the configuration of the filter parameter device 17. Using the filter parameters a obtained, the filtered measured values of the rotational speed sensor are subsequently determined in an adaptive filter device 18The filtered tachometer sensor measurement values are illustrated in fig. 2One possible example implementation of the configuration of the determining or filtering means 18. After the filtered measured value of the rotating speed sensor is determinedThen, the result is supplied to the decimation filter 20. Reduction of filtered rotational speed sensor measured values by means of a decimation filter 20To provide filtered tachometer sensor measurements 21 with reduced output rate, particularly sample rate. The filtered tachometer measurement 21 with the reduced output rate is subsequently smoothed by means of a smoothing filter 30. The following offset 40 is thus determined: the offset can be used to correct or calibrate the tacho sensor signal 11. The smoothing filter 30 may comprise, for example, exponential smoothing. But other filtering techniques may alternatively be considered. Alternatively, it is also possible to consider the measured value directly from the filtered rotational speed sensorObtained or used as offset 40. Alternatively, it is also conceivable to use the filtered tachometer measurement 21 with a reduced output rate directly as the offset 40.

FIG. 2 shows a schematic block diagram of a portion of a system 100 for offset calibration according to an embodiment of the present invention. In particular, an embodiment of the filter parameter device 17 is shown. In a first part 17' of the filter parameter means 17, the relation is passedA first intermediate parameter a1 is found. Here, # Z is the number of rotation speed sensor measurement values x1 measured in the stationary state 15 since the first sample in the stationary state 15 or the number of samples used for calibration. In a second part 17' of the filter parameter device 17, a second intermediate parameter is thus determinedThe number a2 is such that the range from 0 to the second threshold thr2 (i.e. in particular the range from 0 to the maximum variance of the rotational speed sensor measurement 10) is divided into a plurality of sections or sub-ranges. For example, an interval may be provided for each typical noise level that occurs in different wearable device calibration scenarios. Each interval is assigned a fixed value of the second intermediate parameter a 2. Therefore, the second intermediate parameter a2 obtains values assigned to the following intervals: the (maximum) fluctuation or variance of the measured rotational speed sensor measurement x1 in the stationary state 15 falls within this interval. Therefore, the second intermediate parameter a2 is based on the noise power of the rotational speed sensor 1. In a third part 17' ″ of the filter parameter device 17, the filter parameter a is then determined from the first and second intermediate parameters a1, a2 and is selected in particular to be a max (a1, a 2). It is generally contemplated that the first, second and third portions 17', 17 ", 17'" may be implemented as one common functionality. The derived filter parameters a may then be provided to the adaptive filter means 18. Determination of a filtered measured value of a rotational speed sensor by means of an adaptive filter device 18For this purpose, the adaptive filtering means 18 may use exponential smoothing, in particular the following relation:

wherein the content of the first and second substances,is a filtered tachometer sensor measurement, a is a filter parameter,is the previously filtered tachometer sensor measurement (as a previous sample), x n]Is the current tachometer measurement value, n describes the number of tachometer measurement values or filtered tachometer measurement values. In this case, it is preferably filtered in a preceding iteration in a manner similar to the above-described determinationMeasured value of a rotational speed sensorTo determine previously filtered tachometer sensor measurementsOther filters than the above-mentioned exponential smoothing may also be considered for the adaptive filter means 18.

With the aid of the embodiment of the system 100 depicted in fig. 1 and 2, a method for offset calibration of the tachometer signal 11 of the tachometer sensor 1 can be carried out, in which:

in a first step, it is determined that the rotational speed sensor 1 is in a standstill state 15;

in a second step, which follows the first step, a filter parameter a is determined from the measured value x1 of the rotational speed sensor 1 measured in the stationary state 15;

in a third step, which follows the second step, the filtered measured value of the rotational speed sensor is determined using the filter parameter a

Wherein the measured value is measured by means of a filtered rotational speed sensorAn offset 40 is determined.

The offset 40 can then advantageously be used for offset correction or offset calibration of the tacho sensor signal 11.