Inertial measurement device and method for operating measurement device

Abstract

The invention relates to an inertial measurement device (1) for providing output sensor data as a function of a force or motion exerted on the inertial measurement device (1), the inertial measurement device (1) comprising a sensor unit (2) and a processing unit (4, 5). The sensor unit (2) comprises one or more sensor elements (21) for detecting motion and is configured to continuously provide motion sensor data samples for each sensor element (21a, 21b). A processing unit (4, 5) for filtering the data in accordance with or corresponding to a motion sensor data sample obtained by the sensor element (21) to provide output sensor data, where the processing unit is configured to select one or more sets of filtering parameters determined by at least one predetermined rule applied to the sensor data sample, and filtering the data based on the selected set of filtering parameters.

Description

Technical Field

[0001] The present invention relates to inertial measurement devices (IMU) for measuring inertia (linear (translation) acceleration), angular velocity and / or angular acceleration in one or more dimensions. The present invention particularly relates to inertial measurement devices, which include pre-processing of raw sensor signals, including data cleaning, filtering, etc. Background Art

[0002] IMUs are used in a variety of applications, such as for tilt sensors and for navigation purposes in mobile devices, etc. In particular, IMUs are often used in guidance systems used in vehicles, providing an independent and alternative source for navigation.

[0003] A standard IMU may be implemented with one or more sensor units, which may be implemented as a unit based on a MEMS (Micro-Electro-Mechanical System). The sensor unit typically includes a capacitively coupled spring-biased mass to provide a variable capacitance output based on a motion-induced force (such as an inertial force or a centripetal force) applied thereto. The variable capacitance is typically measured to provide an electrical sensor signal in the form of, for example, a voltage signal.

[0004] Typically, the electrical sensor signal provided by the sensor unit is a raw sensor signal. Since the sensing of the sensor unit is sensitive to a wide range of physical conditions (such as temperature, ambient pressure, mechanical stress, magnetic / electromagnetic fields, crosstalk, etc.), the raw sensor signal must be preprocessed to obtain clean output sensor data that can be used for downstream applications. Therefore, within a close distance, the preprocessing unit is coupled to the sensor unit to allow calibration, recalibration and / or filtering of the raw sensor signal. Preprocessing can typically include compensation of the raw sensor signal, which allows correction of misalignment of the sensing unit and allows application of scaling factors and offsets to correct sensing mismatches. Such compensation can be defined by a compensation table to store relevant compensation parameters.

[0005] Furthermore, preprocessing may include filtering of the raw sensor signal, in particular in order to reduce or eliminate dynamic effects such as those from high frequency vibrations. The filter may be parameterized by filter parameters. Typically, compensation tables and filter parameters are determined by calibration, however, due to the high sensitivity of the sensor signal measurement to the physical conditions of the environment, it is not possible to obtain an accurate sensor signal in all operating situations.

[0006] Furthermore, sensor devices may be exposed to different dynamic conditions, which may require online adjustment of filtering parameters to provide output sensor data suitable for direct use in downstream applications.

[0007] The object of the present invention is to provide an improved inertial measurement device which allows obtaining output sensor data which are well adapted to the dynamic conditions to which the device is exposed and which take into account variable operating conditions. Summary of the invention

[0008] This object has been achieved by an inertial measurement device with adaptive filtering and a method for operating an inertial measurement device according to the invention.

[0009] The present invention also presents other embodiments.

[0010] According to a first aspect, there is provided an inertial measurement device for providing output sensor data based on a force or motion exerted on the inertial measurement device, the device comprising:

[0011] a sensor unit comprising one or more sensor elements for detecting motion and configured to continuously provide motion sensor data samples for each sensor element;

[0012] - a processing unit for filtering data according to or corresponding to motion sensor data samples obtained by the sensor element to provide output sensor data, wherein the processing unit is configured to select one or more filter parameter sets determined by at least one predetermined rule applied to the sensor data samples, and to filter the data based on the selected filter parameter sets.

[0013] According to another embodiment, the processing unit may include:

[0014] a filtering unit having at least one filter element parameterized with filter parameters of a selected filter parameter set and configured to provide output sensor data as a function of the motion sensor data samples and the applied filter parameters;

[0015] - a filtering parameter unit comprising a rule-based filtering determination unit configured to provide one of a plurality of preset filtering parameter sets to the filtering element in response to motion sensor data samples obtained by the sensor element according to a given rule set.

[0016] According to another embodiment, an inertial measurement device for providing output sensor data based on a force or motion applied to the inertial measurement device is provided, the device comprising:

[0017] a filtering unit having a plurality of filter elements, each filter element being parameterized with filter parameters of a respectively predetermined filter parameter set and each filter element being configured to provide filtered data depending on the motion sensor data samples and the applied filter parameters, wherein the filtering unit comprises a selection unit for selecting at least one filter element to output the filtered data as output sensor data;

[0018] - a filter parameter unit comprising a rule-based filter determination unit configured to select one of a plurality of filter elements to output the output sensor data via the selection unit.

[0019] In addition, the sensor unit may also include one or more operating sensor elements, which in particular include at least one of a temperature sensor, an ambient pressure sensor, a mechanical stress sensor, a magnetic sensor and an electromagnetic field sensor, wherein the sensor unit is configured to provide an operating sensor data sample for each operating sensor element, wherein the processing unit and the filtering parameter unit are respectively configured to receive the operating sensor data samples, and provide filtering parameters to the filtering unit or select one of multiple filtering units based on the operating sensor data samples.

[0020] The filtering unit may include at least one of a complementary filter and a Kalman filter as a filtering element.

[0021] For example, the filtering parameters may include at least one of: a time constant, one or more elements of a state transition matrix of the Kalman filter, one or more elements of a control input matrix of the Kalman filter, one or more elements of a measurement matrix of the Kalman filter, a Kalman gain of the Kalman filter, a zero position of the compensation filter, a pole position of the compensation filter, a proportional gain of the compensation filter, an integral gain of the compensation filter, and a differential gain of the compensation filter.

[0022] The inertial measurement device may continue to operate in subsequent sampling cycles, wherein sensor signals of the motion sensor elements are sampled and digitized.

[0023] In essence, the inertial measurement device comprises: a sensing unit having one or more sensing elements, such as a motion sensor, such as an accelerometer, a gyroscope, etc.; and a pre-processing unit comprising a filtering unit having one or more filtering elements suitable for applying to all sensor data originating from the one or more sensing elements.

[0024] By filtering the filter parameter unit, the filter unit is permanently adapted to the operating conditions of the inertial measurement device. The filter parameter unit is continuously operated and is configured to select a filter parameter set from various preset filter parameter sets in response to the motion sensor data samples and optionally in response to one or more operation sensor data samples.

[0025] The filter parameter unit applies the rules to the sensor data samples, which may optionally include one or more operational sensor data samples, to select a set of filter parameters that best suits the filter unit.

[0026] It can be provided that the rule database is configured to provide a plurality of rule sets, each rule set being associated with one of a plurality of filter parameter sets, wherein at least one rule is defined using a threshold value, wherein in the rule-based filter determination unit, it is configured to retrieve the plurality of rule sets and to evaluate the plurality of rule sets, wherein, in particular, a result of the evaluation is associated with a filter parameter set to be applied to a filter unit, or with the selection of one of a plurality of filter units, in particular by using a lookup table.

[0027] Additionally, at least one rule can depend on one of the following:

[0028] - actual values ​​of sensor data samples of one or more motion sensor elements;

[0029] - actual values ​​of sensor data samples of one or more operating sensor elements,

[0030] - a maximum or minimum value of sensor data samples of one or more motion sensor elements within a given number between 2 and 10 of the most recent sensor data samples;

[0031] - a maximum or minimum value of the sensor data samples of one or more operating sensor elements within a given number between 2 and 10 of the most recent sensor data samples,

[0032] - the differential of the most actual sensor data sample of one or more motion sensor elements;

[0033] - the differential of the most actual sensor data sample of one or more operating sensor elements,

[0034] - integration of the most actual sensor data samples of one or more motion sensor elements;

[0035] - the integration of the most actual sensor data samples of one or more operating sensor elements,

[0036] - a maximum gradient of the characteristic of the sensor data samples within a given number of 2 to 10 most recent sensor data samples,

[0037] - a sliding average of the 2 to 10 most recent sensor data samples for each motion sensor element; and

[0038] - A sliding average of the 2 to 10 most recent sensor data samples for each operating sensor element.

[0039] For example, for a sensor data sample, a rule may be applied to an actual value of the sensor data or to a sliding average of a plurality of latest sensor data samples. In particular, the comparative sensor data value thus obtained may be evaluated by comparison with a corresponding given threshold value determining the filter parameter set to be selected. Basically, the filter parameter set is selected from the results of the respective comparisons with the given threshold values, e.g. by a lookup table or the like. Each filter parameter set determines a plurality of filter parameters to process the dynamic characteristics of the sensor data sample. For example, it has been determined that a translational acceleration exceeding a given threshold value results in a filter parameter of a time constant to reduce the dynamic influence of the sensor data.

[0040] Furthermore, the filter parameter unit may be configured to prohibit application of a newly selected filter parameter set after a change in selection of the filter parameter set for a period of time and before a given dead time expires. Thus, the filter parameter unit may also be configured to apply a dead time after a change in the filter parameter set based on dynamic changes according to the motion sensor data samples and optionally according to one or more operational sensor data samples. The dead time may be in the range of 0.1 to 20 seconds, preferably in the range of 0.5 to 4 seconds, more preferably between 0.1 and 5 seconds. During the dead time, the rule-based determination of the filter parameter set determination or selection is suspended to avoid large changes between filter parameter sets within a short period of time. This helps to avoid artifacts in the resulting output sensor data characteristics.

[0041] However, it may be provided that if one or more rules determine that the comparison sensor data value exceeds a given threshold by more than a predetermined amount, for example more than 100% of the comparison sensor data value, or causes the given threshold to be greater than 50% of the comparison sensor data value, then the dead time may be ignored and changes to the filter parameter set may be permitted.

[0042] Furthermore, provision can be made for one or more of the filter parameters of the determined or selected filter parameter set to be ramped up with a given slope to the new value of the respective filter parameter in order to avoid abrupt changes in the filter behavior.

[0043] According to another embodiment, the filter unit can be formed by a plurality of filters, each filter operating with a different set of filter parameters calculated simultaneously and continuously. After each cycle, the actual sensor data is evaluated in the filter parameter unit and the corresponding output sensor data of one of the filter units is selected and provided as the output sensor data of the IMU.

[0044] Furthermore, if a change in the set of filter parameters has been determined, the previously selected filter unit (corresponding to the previous set of filter parameters) and the newly selected filter unit (corresponding to the newly determined filter parameters) can be used to generate a value of the output sensor data value, such as an average between the values ​​of the two filter unit output data, so as to avoid unexpected behavior of the output sensor data. Furthermore, the selection of the most reasonable result within two different results can be determined based on the previous data.

[0045] A compensation unit may be provided to pre-process the motion sensor data samples to compensate for systematic sensor errors.

[0046] According to one embodiment, the smoothing element may be configured to low pass filter or ramp the output sensor data samples depending on a change in the set of applied filter parameters or the selected filter element, respectively.

[0047] It may be provided that, upon change of the respective one or more filter parameters, the filter parameter unit is configured to provide a ramp-up from previous values ​​of the one or more filter parameters to newly determined values ​​of the one or more filter parameters which may be completed within 2 to 10 sampling periods. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Embodiments are described in more detail with reference to the accompanying drawings, in which:

[0049] Figure 1 An inertial measurement device according to an embodiment of the present invention is shown;

[0050] Figure 2 Shown according to Figure 1 Functional diagram of a parametric model of a filter unit of an inertial measurement device;

[0051] Figure 3 An inertial measurement device according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0052] Figure 1Schematically, an inertial measurement device 1 is shown that includes a sensor unit 2, which includes one or more sensor elements 21. The sensor element 21 can be basically formed as a motion sensor, such as an accelerometer 21a, a gyroscope 21b, etc., which can provide sensor signals up to three dimensions. For example, the accelerometer 21a includes a three-axis accelerometer that measures the acceleration of the rate of change of velocity along the x-axis, y-axis and z-axis of the IMU 1.

[0053] The motion sensors 21a, 21b can be formed as inertial masses with spring bias, which are movable under the impact of applied force. The movement of the inertial mass can be detected by physical effects. For example, the capacitance change can be measured according to the magnitude of the applied force. The measurement can provide a voltage, current or frequency signal as the electrical sensor signal x, respectively. Other physical effects for measuring the force caused by the inertia of the mass can also be applied, and any type of electrical signal can be provided. The output sensor data can also be an angle and / or a compensated angle and / or other raw data.

[0054] In addition, an operating condition sensor element 22 as a non-motion sensor may be included in the sensor unit 2 to detect an operating condition of the inertial measurement device 1. The operating condition sensor element 22 may include one or more of a temperature sensor, an ambient pressure sensor, a mechanical stress sensor (for measuring strain), a magnetic / electromagnetic field sensor (for example, for measuring a magnetic field of the earth or EMC influences), and the like.

[0055] The sensor signals x from the sensors 21a, 21b and 22 (which are essentially provided as analog electrical signals X) are sampled and digitized by the analog-to-digital converter 23 to provide information of the sensor signals at the output of the sensor unit 2 as raw sensor data samples X raw . Raw sensor data sample X raw The raw sensor data samples X may include motion sensor data samples associated with the motion sensors 21a, 21b and operation sensor data samples associated with the operation sensor 22. In the operation of the IMU 1, the raw sensor data samples X raw The data is provided as a stream of continuous sampling cycles, the sampling rate of which can be in the low frequency range of a few Hz to hundreds of kHz.

[0056] Raw sensor data sample X raw may be provided to an optionally provided compensation unit 3 which may be implemented by hardware and / or software. In the compensation unit 3, the raw sensor data samples X are performed rawThe compensation may be performed according to a calibrated compensation table comprising a scale factor f, an offset value b for each sensor element 21a, 21b, and a misalignment matrix C for correcting the misalignment of the sensor elements 21a and 21b, which allows compensation of systematic sensor errors. In particular, the compensated sensor data sample X comp For example, X comp =C*f*X raw +b is obtained, where b is the given offset compensation.

[0057] The compensated sensor data samples X can then be comp Provided to the filtering unit 4 which can be implemented by hardware and / or software. The filtering unit 4 may include one or more filters 41 to compensate the sensor data samples X comp In other embodiments, the raw sensor data samples X raw directly to the filtering unit 4, wherein the original sensor data sample X is not raw Make compensation.

[0058] For example, the filtering unit 4 may comprise a complementary filter and / or a Kalman filter and / or any other signal processing as a specific filtering element 41. The type of filtering element 41 may be selected according to given filtering parameters, and the filtering element 41 itself may be parameterized by the filtering parameters. The type of filtering is determined by the filtering parameters of a given set of filtering parameters. The output of the filtering element 41 may be at the output of the inertial measurement device 1 as output sensor data samples X out Direct output.

[0059] In addition, the output sensor data samples X out may be provided to the smoothing element 6, which is configured to sample the output sensor data X of the filter element 41 according to the change of the filter parameter set P out Low pass filtering is performed, or the output of the filter element 42 is ramped. This allows abrupt changes in the output of the filter element 41 to be smoothed when the set of filter parameters and therefore the filtering behavior are changed.

[0060] The filter element 41 can be operated based on the filter parameters. For example, the filter parameters can be time constants, etc. For a Kalman filter as a filter element, the filter parameters can include at least one of the following: one or more elements of a state transition matrix, one or more elements of a control input matrix, one or more elements of a measurement matrix, and / or a Kalman gain. As a compensation filter, a lead compensator, a lag compensator, a lead-lag compensator, or a PID controller can be applied. For a compensation filter as a filter element, the filter parameters can include at least one of the following: zero position, pole position, proportional gain, integral gain, and differential gain.

[0061] Furthermore, a filter parameter unit 5 may be provided, which comprises a rule-based filter determination unit 51. The filter determination unit 51 may be implemented in hardware and / or software. The filter determination unit 51 may be provided with access to a defined filter parameter set for configuring the filter element 41. The filter parameter set may be associated with a selection criterion, such that the filter element 41 is selected according to the rule applied to one or more comparison sensor data values ​​X. c Based on the results above, select the filter parameter set to be applied.

[0062] The association between the application result of the rule and the set of filter parameters to be applied can be achieved by means of a stored lookup table 53. The lookup table 52 can be pre-configured and stored in the filter parameter unit 5.

[0063] The filter parameter unit 5 receives the sensor data X raw , the sensor data X raw The filter parameter unit 5 may receive the compensated sensor data samples X, including inertial values ​​of linear and / or angular acceleration and / or angular velocity, and operating parameters such as the temperature of the inertial measurement device 1 and other operating conditions measured such as ambient pressure, mechanical stress, magnetic / electromagnetic field exposure, etc. comp .

[0064] Each comparison sensor data value X c Each is associated with a sensor element 21a, 21b, 22 and may each correspond to or may each be directly obtained from one or more timely continuous sensor data samples X (of one motion sensor 21a, 21b or of the operation sensor 22) in the comparison value unit 52. raw Export in.

[0065] For example, actual values ​​of sensor data samples of one or more motion sensor elements 21a, 21b or operation sensor elements 22 may be used as comparison sensor data values. Furthermore, at least one of the comparison sensor data values ​​may correspond to a maximum or minimum value of a sensor data sample (for one sensor 21a, 21b, 22) within a given number of the most recent sensor data samples, or to a differential of the most actual sensor data sample (for one sensor 21a, 21b, 22). Furthermore, the comparison sensor data value may be determined by a maximum gradient of a sensor data characteristic within a most recent time interval of 2 to 20 sensor data samples (for one sensor 21a, 21b, 22). Furthermore, a sliding average of, for example, 2 to 10 most recent sensor data samples associated with the respective sensor elements 21a, 21b, 22 may be used as comparison sensor data values. The comparison sensor data value of each sensor data element, for example associated with acceleration along the x-axis, may be applied to a rule, which may be in the form of a threshold comparison with a given threshold.

[0066] The following table illustrates the selection of a set of filtering parameters based on a rule-based approach to evaluating sensor data:

[0067]

[0068] Thresholds a to m are given thresholds associated with a particular rule.

[0069] Generally speaking, in each set of rules, the thresholds for each axis can be different.

[0070] The filter parameter unit 5 may also provide functionality in which, after a change in the selection of a filter parameter set, the next change of the filter parameter set is inhibited before a given dead time expires. The given dead time may be between 0.5s and 4s and is considered to avoid unstable conditions caused by multiple changes in filtering behavior in a very short period of time.

[0071] However, if one of the thresholds of the rule-based scheme is exceeded by more than a given amount, e.g. 50%, or exceeds a given amount, e.g. 50%, the dead time function may be suspended and the filtering parameter set immediately changed according to the filtering parameter set determined by the rule-based filtering parameter unit 5 or the default parameter set.

[0072] When a new selection of a set of filter parameters occurs, the filter parameter unit 5 may provide a ramp between one or more of the filter parameters such as a time constant, so that when the corresponding one or more filter parameters change, a (linear) ramp from a previous value of one or more filter parameters to a newly determined value of one or more filter coefficients may be completed within more than 2 to 10 sampling periods.

[0073] In addition, according to Figure 3 In the different embodiments shown, multiple filter elements 41 can be used. As shown by the same reference numerals, Figure 3 The embodiment is basically similar to Figure 1 The difference is that each filter element 41 calculates a separate filtered data sample according to an associated one of the predetermined filter parameter sets simultaneously within a single sampling period. The filtered data sample of each filter element 41 is applied to the selection unit 42. Therefore, the filter parameter unit 5 is configured to control the selection unit 42 so that according to the result of applying the rule set, one of the filter elements 41 is selected to output the filtered data sample obtained with the respective filter parameter set as the output sensor data sample.

Claims

1. An inertial measurement device (1) for providing output sensor data based on a force or motion applied to the inertial measurement device (1), the inertial measurement device (1) comprising: a sensor unit (2) comprising one or more motion sensor elements (21) for detecting motion and configured to continuously provide motion sensor data samples for each motion sensor element; and a processing unit for filtering data according to or corresponding to the motion sensor data samples obtained by the motion sensor element (21) to provide the output sensor data (X) out ), wherein the processing unit is configured to: select one or more filter parameter sets determined by at least one predetermined rule applied to the motion sensor data sample, and filter the data based on the selected filter parameter sets.

2. The inertial measurement device (1) according to claim 1, wherein: The processing unit comprises: a filtering unit (4), the filtering unit (4) having at least one filtering element (41), the filtering element (41) being parameterized with filtering parameters of the selected filtering parameter set and being configured to provide output sensor data (X) as a function of the motion sensor data samples and the applied filtering parameters out );and - a filtering parameter unit (5), the filtering parameter unit (5) comprising a rule-based filtering determination unit (51), the rule-based filtering determination unit (51) being configured to provide one of a plurality of preset filtering parameter sets to the filtering element (41) in response to the motion sensor data sample obtained by the motion sensor element (21) according to a given rule set.

3. The inertial measurement device (1) according to claim 1, wherein: The processing unit comprises: - a filtering unit (4), the filtering unit (4) having a plurality of filtering elements (41), each of the filtering elements (41) being parameterized with filtering parameters of a respectively preset filtering parameter set, and each of the filtering elements (41) being configured to provide filtering data according to the motion sensor data samples and the applied filtering parameters, wherein the filtering unit (4) comprises a selection unit (42) for selecting at least one of the filtering elements (41) to output the output sensor data (X) according to or in correspondence to the filtering data provided by at least one of the filtering elements (41) out ); a filtering parameter unit (5), the filtering parameter unit (5) comprising a rule-based filtering determination unit, the rule-based filtering determination unit being configured to select one of the plurality of filtering elements (41) through the selection unit (42) to output the output sensor data (X) out ).

4. An inertial measurement device (1) according to any one of claims 2 to 3, wherein: The filtering unit (4) includes at least one of a complementary filter and a Kalman filter as a filtering element (41).

5. The inertial measurement device (1) according to claim 1, wherein: The sensor unit (2) further comprises one or more operational sensor elements (22), wherein the sensor unit (2) is configured to provide operational sensor data samples to each of the one or more operational sensor elements (22), wherein the processing unit is configured to: receive the operational sensor data samples and select a set of filtering parameters based on at least one rule of the operational sensor data samples depending on one or more of the operational sensor data samples.

6. The inertial measurement device (1) according to claim 1, wherein: The filtering parameters include at least one of the following: a time constant, an offset and a gain, one or more elements of a state transition matrix of the Kalman filter, one or more elements of a control input matrix of the Kalman filter, one or more elements of a measurement matrix of the Kalman filter, a Kalman gain of the Kalman filter, a zero position of the compensation filter, a pole position of the compensation filter, a proportional gain of the compensation filter, an integral gain of the compensation filter, and a differential gain of the compensation filter.

7. The inertial measurement device (1) according to claim 1, wherein: The inertial measurement device (1) operates continuously in subsequent sampling periods, in which the sensor signals of the motion sensor elements (21) are sampled and digitized.

8. The inertial measurement device (1) according to claim 5, wherein: The rule database (52) is configured to provide a plurality of rule sets, each of the plurality of rule sets being associated with one of a plurality of filter parameter sets, wherein at least one rule is defined using a threshold value, wherein, in a rule-based filter determination unit (51), the at least one rule is configured to retrieve and evaluate the plurality of rule sets, wherein a result of the evaluation is associated with a filter parameter set to be applied to one of a plurality of filter elements (41), or is associated with selecting one of a plurality of the filter elements (41).

9. The inertial measurement device (1) according to claim 8, wherein: At least one rule depends on one of the following: - actual values ​​of sensor data samples in one or more of said motion sensor elements; - actual values ​​of sensor data samples in one or more of said operating sensor elements, - a maximum or minimum value of the sensor data samples of one or more of said motion sensor elements within a given number between 2 and 10 of the most recent sensor data samples; - a maximum or minimum value of the sensor data samples of one or more of said operating sensor elements within a given number between 2 and 10 of the most recent sensor data samples, - a differential of the most actual sensor data sample of one or more of said motion sensor elements; - the differential of the most actual sensor data sample of one or more of said operating sensor elements, - integration of the most actual sensor data samples of one or more of said motion sensor elements; - an integration of the most actual sensor data samples of one or more of said operating sensor elements, - a maximum gradient of the characteristic within a given number of sensor data samples between 2 and 10 of the most recent sensor data samples, - a sliding average of the 2 to 10 most recent sensor data samples of each of said motion sensor elements; and - a sliding average of the 2 to 10 most recent sensor data samples of each of said operating sensor elements.

10. An inertial measurement device (1) according to any one of claims 8 to 9, wherein: The at least one rule determines a threshold comparison, or a combination of a threshold comparison and a given threshold comparison as selection criterion.

11. The inertial measurement device (1) according to claim 1, wherein: The compensation unit (3) is arranged to pre-process the motion sensor data samples and the operation sensor data to compensate for the system sensor errors.

12. The inertial measurement device (1) according to claim 2, wherein: The smoothing element (6) is configured to smooth the output sensor data (X) according to the detected changes of the applied set of filter parameters or the selected filter element (41), respectively. out ) to perform low-pass filtering or to make the output sensor data (X out ) rises obliquely.

13. The inertial measurement device (1) according to claim 2, wherein: The filtering parameter unit (5) is configured to prohibit application of the newly selected filtering parameter set after the time of changing the selection of the filtering parameter set and before the expiration of a given dead time.

14. The inertial measurement device (1) according to claim 2, wherein: The filter parameter unit (5) is configured to provide, when one or more filter parameters are changed, a ramp-up from previous values ​​of the one or more filter parameters to newly determined values ​​of the one or more filter parameters that can be completed within 2 to 10 sampling periods.

15. The inertial measurement device (1) according to claim 3, the sensor unit (2) further comprising one or more operating sensor elements (22), wherein: The sensor unit (2) is configured to provide an operation sensor data sample to each of the one or more operation sensor elements (22), wherein the filter parameter unit (5) is configured to: receive the operation sensor data samples and select a set of filter parameters based on at least one rule of the operation sensor data samples depending on one or more of the operation sensor data samples.

16. The inertial measurement device (1) according to claim 5, the sensor unit (2) comprising at least one of a temperature sensor, an ambient pressure sensor, a mechanical stress sensor, a magnetic sensor, an electromagnetic field sensor and a humidity sensor.

Images