Methods for drift correction in a sensor system

Abstract

The invention relates to a method (100) for drift correction in a sensor system (200), comprising: Determining (101) values ​​of an operating parameter (201) of the sensor system (200); Executing (103) a correction model (203) taking into account the values ​​of the operating parameter (201) and determining a correction value (205) of a sensor value (227) of the sensor system (200), wherein the correction model (203) represents a model description of an operating state of the sensor system (200) as a function of the operating parameter (201) and describes a relation between values ​​of the operating parameter (201) and corrected sensor values ​​(207) of the sensor system (200); Performing (105) a correction of sensor values ​​(227) of the sensor system (200) taking into account the correction values ​​(205) of the correction model (203); and Providing (107) sensor values ​​(207) corrected with respect to the correction values ​​(205).

Description

[0001] The present invention relates to a method for drift correction in a sensor system. State of the art

[0002] Methods for drift correction in sensor systems are known from the prior art.

[0003] The task is to provide an improved method for drift correction in a sensor system.

[0004] The problem is solved by the drift correction method in a sensor system of the independent claim. Advantageous embodiments are the subject of the dependent claims.

[0005] One aspect is described as a drift correction method in a sensor system, encompassing: Determining the values ​​of an operating parameter of the sensor system; Executing a correction model taking into account the values ​​of the operating parameter and determining a correction value of a sensor value of the sensor system, wherein the correction model represents a model description of an operating state of the sensor system as a function of the operating parameter and describes a relation between values ​​of the operating parameter and corrected sensor values ​​of the sensor system; Performing a correction of sensor values ​​of the sensor system, taking into account the correction values ​​of the correction model; and Providing sensor values ​​corrected with respect to the correction values.

[0006] This approach offers the technical advantage of providing an improved drift correction method for a sensor system. For this purpose, the values ​​of at least one operating parameter of the sensor system are evaluated by a suitably designed correction model, which then determines correction values ​​to correct the sensor system's readings. The operating parameter is a parameter that is available as a quantitative value during operation. This eliminates the need for a separate measurement of a quantity required to check for drift, such as a temperature measurement to determine the temperature within the sensor system. By considering the operating parameter, a temperature-related drift in the sensor system's measurement accuracy can still be determined by selecting an operating parameter that exhibits a temperature dependency.In addition to temperature, other environmental influences that cause a drift in the sensor parameters can also be corrected by appropriately selecting the respective operating parameter that is affected by the corresponding environmental influence, for example humidity or pressure.

[0007] Micromechanical sensors, especially inertial sensors with combined angular rate and acceleration sensors, experience changes in sensitivity, offset, or nonlinearities of output signals due to loads applied during the component's lifetime. To compensate for these drifts or changes, temperature is typically considered one of the most significant influencing factors for correcting the output values. The temperature is measured within the sensor system and computationally accounted for in the data path based on physically motivated or purely data-driven models. For this purpose, measurement series can be recorded to precisely determine the temperature dependence of the measurement accuracy, i.e., the temperature-driven drifts, of the sensor systems.

[0008] A disadvantage of this methodology, however, is that while temperature has many effects within the sensor system, the actual effects within the sensor system cannot be directly attributed to the temperature of the sensor system itself, but rather are directly based on stress states within the sensor system caused by the temperature. Instead of directly considering the stress state of the sensor system, which is the actual cause of the sensor system's drift, the prior art takes a detour via temperature by measuring the temperature and thereby determining the temperature dependence of the sensor system's measurement accuracy.

[0009] Due to the indirect nature of temperature measurements, the stress state of the sensor system cannot be accurately determined. However, by considering the operating states, the (stress) states of the sensor system can be described much more accurately and precisely, which in turn enables significantly improved drift correction.

[0010] For example, this method can reduce hysteresis effects that occur after temperature ramps. By making the stress within the sensor system directly observable through consideration of the operating parameters, hysteresis effects can be taken into account.

[0011] The method is particularly powerful for applications where multiple measured variables are implemented on the same MEMS chip, for example, a measured rotation rate from a rotation rate sensor and an acceleration from an accelerometer.

[0012] In such cases, it can be assumed that both sensor elements, i.e., the gyroscope and the accelerometer, are subjected to comparable loads. Non-redundant operating parameters of the different sensor elements can be taken into account, thus enabling the most precise drift correction possible.

[0013] By considering the operating parameters already present in the sensor system instead of the temperature in the inventive method, the stress state of the sensor system can be determined directly. The detour via temperature can thus be avoided. This further means that additional sensor elements, such as temperature sensors, are not necessary.

[0014] The method according to the invention can in particular be designed as a computer-implemented method.

[0015] According to one embodiment, the sensor system is a microelectromechanical sensor system and is designed as an inertial sensor system with a gyroscope element and / or an accelerometer element.

[0016] This allows the technical advantage that the method can be applied to microelectromechanical sensor systems.

[0017] According to one embodiment, the operating parameter includes: drive voltage and / or drive frequency of the gyroscope, quadrature signal of the gyroscope, test signal amplitude of the sensor elements, noise behavior of the sensor elements, transient test signal behavior, common-mode signal suppression.

[0018] This allows the technical advantage to be achieved by taking into account operating parameters that are available during the operation of the sensor system without the need for additional measurements, and on which a drift of the sensor system can be clearly determined.

[0019] According to one embodiment, the correction model is designed as an observer-based model or as a data-based model.

[0020] This allows for the technical advantage of providing a powerful and meaningful correction model, enabling precise determination of correction values ​​based on operating parameters.

[0021] According to one embodiment, the correction model is set up as an observer-based model to compare the values ​​of the operating parameter with a model behavior.

[0022] This allows for the technical advantage of providing a reliable correction model.

[0023] In one embodiment, the correction model is designed as a Kalman filter.

[0024] This allows the technical advantage that, by being designed as a Kalman filter, the correction model can determine precise correction values ​​for the sensor values ​​of the sensor system solely based on the provided values ​​of the drive parameter.

[0025] According to one embodiment, the correction model is derived as a data-based model from characterization measurements of the sensor system, in which the relationship between the values ​​of the operating parameter and corrected sensor values ​​was determined.

[0026] This allows the technical advantage to be achieved that the data-based correction model can be precisely and individually adapted to the respective sensor system and the drift behavior occurring within it through the characterization measurements.

[0027] According to one embodiment, the operating state describes a stress state of the sensor system and describes a stress load on the sensor system.

[0028] This offers the technical advantage that the correction model can take stress loads within the sensor system into account. This enables the most precise drift correction possible for the sensor system's values.

[0029] According to one aspect, a computing unit is provided which is set up to execute the drift correction procedure in a sensor system according to one of the preceding embodiments.

[0030] According to one aspect, a computer program product is comprehensively provided with commands which, when the program is executed by a data processing unit, cause it to execute the drift correction method in a sensor system according to one of the preceding embodiments.

[0031] Embodiments of the invention are described with reference to the following figures. The figures show: Fig. 1 a schematic representation of a sensor system according to one embodiment; Fig. 2 a further schematic representation of a sensor system according to a further embodiment; Fig. 3 a flowchart of a drift correction method for a sensor system according to one embodiment; and Fig. 4. A schematic representation of a computer program product.

[0032] Fig. Figure 1 shows a schematic representation of a sensor system 200 according to one embodiment.

[0033] In the embodiment shown, the sensor system 200 is designed as a microelectromechanical sensor system 200 and comprises a MEMS element or sensor element 209, an ASIC 211 and a substrate 213.

[0034] The MEMS element 209 is arranged on the ASIC 211 via a bonding layer 215, which in turn is arranged on the substrate 213.

[0035] According to the invention, the ASIC 211 is configured to carry out the drift correction method according to the invention in a sensor system 200.

[0036] In the embodiment shown, the sensor system 200 further comprises a stress sensor 217, which is arranged on the MEMS element 209, and via which a stress state of the sensor system 200 can be determined.

[0037] Fig. Figure 2 shows a further schematic representation of a sensor system 200 according to a further embodiment.

[0038] Based on the graphic in Fig. 2 Various steps of the inventive method for drift correction in a sensor system are described.

[0039] The illustration shown only depicts the sensor element 209 and the ASIC 211 of the sensor system 200.

[0040] For drift correction, values ​​of an operating parameter 201 are first determined and transferred from the sensor element 209 to a correction model 203.

[0041] Correction model 203 is configured to determine correction values ​​205 of the sensor values ​​227 of sensor system 200, taking into account the values ​​of operating parameter 201. Correction model 203 represents a model description of an operating state of sensor system 200 as a function of operating parameter 201 and describes a relationship between the values ​​of operating parameter 201 and corrected sensor values ​​207 of sensor system 200.

[0042] The operating state of sensor system 200 can represent a stress state of sensor system 200, in which a stress load on sensor system 200 and, in particular, on individual components such as the sensor elements 209 of sensor system 200 is described. This stress load on sensor system 200 can be caused, for example, by temperature influences or other environmental influences acting on sensor system 200. The stress load refers in particular to the various microelectromechanical components, such as the MEMS sensor elements 209, of the microelectromechanical sensor system 200.

[0043] The sensor system 200 can be designed as a micromechanical sensor system, as is the case in Fig. Figure 1 already shows that the sensor system 200 can be configured as an inertial sensor system with a gyroscope element and an accelerometer element. In the embodiment shown, the gyroscope element and the accelerometer element can be combined in the sensor element 209 shown.

[0044] The operating parameter 201, which is available to the controller during the operation of the sensor element 209, can be, for example, a drive voltage, a drive frequency of the gyroscope element, a quadrature signal of the gyroscope element, a test signal amplitude of the gyroscope or acceleration sensor element, a noise characteristic of the gyroscope or acceleration sensor element, a transient test signal characteristic, or a common-mode signal suppression. The operating parameter 201 can also be a different type of operating parameter that allows for a clear conclusion to be drawn about the operating state, in particular the stress state, of the sensor system 200.

[0045] The operating parameters 201 are provided by a control system 219 of the sensor element 209. The control system 219 is configured to regulate or control the sensor system 200 based on the operating parameters 201, which primarily serve as parameters for the control system 219.

[0046] The values ​​of the operating parameter 201 can be recorded during the operation of the sensor system 200 or during short self-test phases, or provided by the control system 219.

[0047] In the case of the sensor element 200 designed as a gyroscope, the quality, i.e., the internal pressure in the sensor cavity, of the gyroscope can be determined by monitoring the drive voltage as operating parameter 201. If the sensor element 209 also includes the accelerometer, as described above, and this same sensor cavity is used, the stress state of both the gyroscope and the accelerometer can be determined by monitoring the drive voltage.

[0048] According to one embodiment, the correction model 203 is configured as an observer-based model or as a data-based model. When configured as an observer-based model, the correction model 203 can, in particular, be set up to compare the values ​​of the operating parameter 201 with a model behavior. The correction model 203 can, for example, be configured as a Kalman filter. Using the Kalman filter, corresponding correction values ​​205 for drift correction of the sensor values ​​227 of the sensor element 209 can thus be determined for different values ​​of the operating parameter 201.

[0049] In the training as a data-based model, the correction model 203 is derived from characterization measurements of the sensor system, in which relationships between the values ​​of the operating parameter 201 and the corrected sensor values ​​207 were determined. The characterization measurements can be performed on the respective sensor system 200 or on representative sensor systems 200.

[0050] In the illustrated embodiment, the correction model 203 interacts with a correction function 221. Through the interaction between the correction model 203 and the correction function 221, the corresponding correction values ​​205 for drift correction of the sensor values ​​227 of the sensor elements 209 can be determined. The correction values ​​205 represent a drift correction to be performed, by means of which, when the measured values ​​227 of the sensor system 200 are applied with the corresponding correction values ​​205, correspondingly drift-corrected sensor values ​​207 can be determined.

[0051] For this purpose, the correction values ​​205 from the correction function 221 are forwarded to a sensitivity / offset correction 223. The correction model 203, or the correction function 221, which can also be interpreted as part of the correction model 203, considers only the drift corrections that are based on the correspondingly determined stress state of the sensor system 200. The sensitivity / offset correction 223 also takes into account corrections regarding sensitivity or measurement offset that are, for example, sensor-related and exist independently of the respective stress state of the sensor system 200. In addition to the correction values ​​205 of the correction model 203 or the correction function 221, the sensitivity / offset correction 223 also considers the uncorrected sensor values ​​227 of the sensor element 209.

[0052] After performing sensitivity / offset correction 223, taking into account the correction values ​​205 of the correction model 203 and the uncorrected sensor values ​​227, a corresponding drift-corrected sensor value 207 is output by a data output 225 for each sensor value 227.

[0053] According to the invention, the drift correction of the sensor values ​​227 is thus performed exclusively based on the operating parameters 201. Consideration of temperature values ​​from a temperature sensor is not required according to the invention.

[0054] According to the embodiment in Fig. 1 The sensor system 200 can further comprise stress sensors 217, by means of which stress loads of the individual components of the sensor system 200, for example the sensor elements 209 and / or the ASIC 211, can be measured directly. The stress load determined via the stress sensor 217 can be taken into account as additional input data in the drift correction of the correction model 203.

[0055] Fig. Figure 3 shows a flowchart of a method 100 for drift correction for a sensor system 200 according to an embodiment.

[0056] To correct drift in a sensor system 200, values ​​of an operating parameter 201 of the sensor system 200 are first determined in a first process step 101.

[0057] In a further process step 103, a correction model 203 is executed taking into account the values ​​of the operating parameter 201, and correction values ​​205 of the sensor values ​​227 of the sensor system 200 are determined. The correction model 203 represents a model description of an operating state of the sensor system 200 as a function of the operating parameter 201 and describes a relationship between the values ​​of the operating parameter 201 and the corrected sensor values ​​207 of the sensor system 200.

[0058] In a further process step 105, a correction of the sensor values ​​227 of the sensor system 200 is carried out taking into account the correction values ​​205 of the correction model 203.

[0059] In a further process step 107, corrected sensor values ​​207 are provided according to the correction values ​​205.

[0060] Fig.Figure 4 shows a schematic representation of a computer program product 300, comprising instructions which, when the program is executed by a data processing unit, cause it to execute the drift correction method 100 in a sensor system 200.

[0061] In the embodiment shown, the computer program product 300 is stored on a storage medium 301. The storage medium 301 can be any storage medium known from the prior art or a computing unit 301. In one embodiment, the computing unit 301 can, in particular, comprise logic gates for executing the method 100.

Claims

[1] Method (100) for drift correction in a sensor system (200), comprising: Determining (101) values of an operating parameter (201) of the sensor system (200); Executing (103) a correction model (203) taking into account the values of the operating parameter (201) and determining a correction value (205) of a sensor value (227) of the sensor system (200), wherein the correction model (203) represents a model description of an operating state of the sensor system (200) as a function of the operating parameter (201) and describes a relationship between values of the operating parameter (201) and corrected sensor values (207) of the sensor system (200); Carrying out (105) a correction of sensor values (227) of the sensor system (200) taking into account the correction values (205) of the correction model (203); and Providing (107) sensor values (207) corrected with respect to the correction values (205). [2] Method (100) according to claim 1, wherein the sensor system (200) is a microelectromechanical sensor system (200) and is designed as an inertial sensor system with a rotation rate sensor element and / or an acceleration sensor element. [3] Method (100) according to claim 2, wherein the operating parameter (201) comprises: drive voltage and / or drive frequency of the yaw rate sensor, quadrature signal of the yaw rate sensor, test signal amplitude of the sensor elements, noise behavior of the sensor elements, transient test signal behavior, common-mode signal suppression, DC capacitances of the sensor element. [4] Method (100) according to one of the preceding claims, wherein the correction model (203) is designed as an observer-based model or as a data-based model. [5] Method (100) according to claim 4, wherein the correction model (203) is configured as an observer-based model to compare the values of the operating parameter (201) with a model behavior. [6] Method (100) according to claim 5, wherein the correction model (203) is designed as a Kalman filter. [7] Method (100) according to one of the preceding claims 4 to 6, wherein the correction model (203) is derived as a data-based model from characterization measurements of the sensor system (200), in which the relationship between the values of the operating parameter (201) and corrected sensor values (207) was determined. [8] Method (100) according to one of the preceding claims, wherein the operating state is a stress state of the sensor system (200) and describes a stress load of the sensor system (200). [9] Computing unit (301) configured to execute the method (100) for drift correction in a sensor system (200) according to one of the preceding claims 1 to 8. [10] Computer program product (300) comprising instructions which, when the program is executed by a data processing unit, cause the data processing unit to carry out the method (100) for drift correction in a sensor system (200) according to one of the preceding claims 1 to 8.

Images