Method for calibrating a sensor with at least two axes

Abstract

Method for calibrating a sensor (1) with at least two axes, comprising the steps: - At least one defined excitation (S1) of the sensor (1) per excitation direction for multiple excitation directions, wherein the excitation directions are different, the number of excitation directions corresponding at least to the number of axes of the sensor (1); - Measuring (S2) all values ​​along all axes for each individual excitation; - Determining (S3) cross-sensitivities from the measured values; - Calculating (S4) correction parameters from the determined cross-sensitivities; - Providing (S5) the correction parameters for a signal path of the sensor (1) for output of corrected measured values, wherein alignment errors are determined based on the cross-sensitivities and axis deviation errors of axes of the sensor (1).

Description

Technical field The invention relates to a method for calibrating a sensor with at least two axes. The invention further relates to a method for calibrating a sensor system comprising at least two subsystems, each with at least one sensor. The invention further relates to a method for calibrating a device with a sensor system. The invention further relates to a sensor with at least two axes. The invention further relates to a calibration device for calibrating a sensor with at least two axes. Although the present invention is generally applicable to any sensors, the present invention is described with reference to MEMS sensors. State of the art Sensors such as accelerometers, gyroscopes, magnetic sensors, or sensor systems consisting of a combination of the aforementioned sensors or similar components with multiple spatial sensor axes exhibit cross-sensitivities. These sensitivities result from unavoidable tolerances in the manufacturing process or from design characteristics. The smaller the sensor or its structures, the greater the impact of manufacturing tolerances on cross-sensitivities and thus on the measurement accuracy of the sensor or sensor system. In sensor systems, in addition to the cross-sensitivity of each individual sensor, the cross-sensitivities between different sensors within the system must also be considered. Cross-sensitivities can be essentially divided into two categories: - Alignment errors cause a rectangular sensor coordinate system to be rotated relative to the sensor's axes and its external environment.- Intrinsic cross-sensitivity arises because individual axes of the sensor coordinate system are no longer perpendicular to the other axes of the sensor coordinate system. Alignment errors arise primarily from tolerances during the installation of individual MEMS sensors in the so-called second-level package. In sensor systems with multiple subsystems, these can also be rotated relative to each other. Furthermore, additional alignment errors occur during soldering on the customer's circuit board. Intrinsic cross-sensitivities, on the other hand, originate within the sensor system or the sensor itself and cannot be influenced by improving tolerances through further sensor positioning. For applications requiring high precision, accurate sensitivity calibration and very low cross-sensitivity are desirable. For applications demanding high accuracy, such as indoor navigation, dead reckoning, virtual reality, augmented reality, or similar, it is desirable for the cross-sensitivity to be of the same order of magnitude. US patent 2007 / 0073502 A1 discloses laser interferometric calibrations for determining the dynamic matrix sensitivity of an inertial sensor. US patent 2002 / 0100310 A1 discloses transformation matrix-based compensations for cross-coupling in sensor clusters. US patent 2011 / 0048103 A1 discloses a method for vector-based trajectory tracking, and US patent 2016 / 0109270 A1 discloses a method for topological sensor addressing using spatial action vectors. Disclosure of the invention In one embodiment, the present invention provides a method for calibrating an at least two-axis sensor according to claim 1. In a further embodiment, the invention provides a method for calibrating a sensor system according to claim 6. In a further embodiment, the invention provides a sensor with at least two axes, calibrated according to a method according to one of claims 1-6, comprising a memory for storing the provided correction parameters and a correction device for correcting measured values ​​of the sensor based on the provided correction parameters. In a further embodiment, the invention provides a method for calibrating a device with a sensor system, wherein the sensor system comprises at least two subsystems, each with at least one sensor, comprising the steps of: - providing correction parameters according to a method according to claim 6 for the sensor system, - arranging the sensor system in the device, - determining a position and / or orientation of at least one sensor of the sensor system in the device, - calculating adjustment parameters from the determined position and / or orientation for the at least one sensor, - calculating adjustment parameters for the sensor system based on the calculated adjustment parameters for the at least one sensor, - providing adjusted correction parameters based on the calculated adjustment parameters and the provided correction parameters for a signal path of the sensor system to output corrected measured values ​​of the sensor system.In a further embodiment, the present invention provides a calibration device for calibrating an at least two-axis sensor, comprising a receiving device for the defined receiving of the two-axis sensor, an excitation device for the defined individual excitation of the sensor per excitation direction for several excitation directions, wherein the number of excitation directions corresponds at least to the number of axes of the sensor, in particular wherein the excitation directions are perpendicular to each other, and a measuring device that can be connected to the sensor and that is configured to measure values ​​of all sensors along each of their axes for each individual excitation, and a computing device configured to determine cross-sensitivities from the measured values ​​and to calculate correction parameters from the determined cross-sensitivities.and for determining alignment errors based on the transverse sensitivities and axis deviation errors of the sensor's axes, and a provisioning device, in particular comprising a memory and an interface configured to provide the correction parameters for the sensor's signal path for outputting corrected measured values. The term "excitation direction" with respect to an axis refers to the direction of excitation along that axis. In other words, excitation along an axis can have two directions. For example, an accelerometer can be excited along both the positive and negative x-directions. One advantage is that measuring values ​​on all axes for each stimulus along excitation directions enables extremely precise calibration of sensors and multi-sensor systems. Another advantage is that it significantly increases the accuracy of measured values ​​from sensors and sensor systems. Furthermore, false signals can be reduced, allowing for higher precision in sensor fusion applications. Further features, advantages and further embodiments of the invention are described below or become apparent therein. Alignment errors are determined based on the transverse sensitivities and axis deviation errors of the sensor's axes. This allows for the simple determination of all transverse sensitivity categories. Axis deviation errors are those errors that result in a non-perpendicular sensor axis coordinate system. According to a further advantageous development, the correction parameters are stored in a memory of the sensor. This increases the user-friendliness of the sensor, for example, if a user can read the relevant data from the memory and use it for other purposes. According to a further advantageous embodiment, after providing the correction parameters and installing the sensor in a device, the sensor is calibrated in its installation position within the device, and the provided correction parameters are adjusted. This allows the sensor to be calibrated in a simple and reliable manner within a finished product, significantly improving the reliability and accuracy of the sensor's measured values ​​in the final product. For example, a less-than-ideal installation position in the final product, which leads to inaccurate measurements, can be compensated for accordingly. According to a further advantageous embodiment, the excitation directions are provided perpendicular to each other. The advantage of this is that with stimuli that are perpendicular to each other, simultaneous excitation of several axes is possible. This reduces the time required to calibrate the sensor. The excitations can be parallel to the sensor axes, but can also be applied at angles of 30 degrees, 45 degrees, or 60 degrees to them. According to a further advantageous refinement, adjusted values ​​are calculated from the measured values, from which the cross-sensitivities are then determined. The advantage of this is that, for example, offsets, drifts, or similar factors in the measured values ​​can be compensated for. This increases the overall accuracy in determining the cross-sensitivities. The method for calibrating a sensor system uses the determined transverse sensitivities to calculate relative alignment errors between at least two subsystems. This allows the individual subsystems or sensors in multisensor systems, such as those with angular rate and acceleration sensors, to be calibrated relative to each other with respect to their alignment. According to a further advantageous embodiment of the sensor, the sensor has at least one interface for reading and / or adjusting the correction parameters. This increases the flexibility of the sensor, for example for the end customer, by giving them access to the correction parameters via the interface and, in particular, allowing them to modify these parameters if necessary, for example depending on the sensor's installation position. Further important features and advantages of the invention will become apparent from the dependent claims, the drawings, and the accompanying description of the figures based on the drawings. It is understood that the features mentioned above and those to be explained below can be used not only in the combination specified, but also in other combinations or on their own, without leaving the scope of the present invention. Preferred embodiments and configurations of the invention are shown in the drawings and are explained in more detail in the following description, wherein identical reference numerals refer to identical or similar or functionally identical components or elements. Brief description of the drawings Fig. 1 shows steps of a method according to an embodiment of the present invention; Fig. 2 shows steps of a method according to an embodiment of the present invention; Fig. 3 shows a sensor according to an embodiment of the present invention in schematic form; Fig. 4 shows a calibration device according to an embodiment of the present invention; Fig. 5 shows steps of a method for calibrating a sensor system according to an embodiment of the present invention; and Fig. 6 shows steps of a method for calibrating a device with a sensor according to an embodiment of the present invention. Embodiments of the invention Fig. 1 shows steps of a method according to an embodiment of the present invention. Figure 1 shows in detail the steps of a method for calibrating a sensor with at least two axes. The method comprises the following steps. In a first step S1, at least one defined excitation of the sensor takes place per excitation direction for several excitation directions, wherein the excitation directions are different, and the number of excitation directions corresponds at least to the number of axes of the sensor. In a second step S2, all values ​​along all axes are measured for each individual excitation. For example, steps S1 and S2 can be performed as follows: - Excitation of the sensor in a first direction, and measurement of all values ​​along each axis, - Excitation of the sensor in a second direction, and measurement of all values ​​along each axis, In other words, parts of steps S1 and S2 can be repeated. With an n-axis sensor system, this yields n*n measured values ​​or values ​​calculated from the measured values. In general, further quantities can first be calculated from the measured values. The following step S3 can then be carried out using these quantities derived from the measured values. In the next step S3, cross-sensitivities are determined from the measured values. In a further step S4, correction parameters are calculated from the determined cross-sensitivities. In a further step S5, the correction parameters for a signal path of the sensor are provided for the output of corrected measured values. Fig. 2 shows steps of a method according to an embodiment of the present invention. Figure 2 shows in detail the steps of a procedure for the live correction of sensor measurements. In the first step (W1), sensor measurement data is provided. In the second step (W2), corrected measurement data is calculated based on provided correction parameters (reference symbol W3), and this data is output in a further step (W4). Fig. 3 shows in schematic form a sensor according to an embodiment of the present invention. In detail, a sensor 1, in particular a MEMS sensor in the form of a gyroscope, an accelerometer, a magnetic sensor or the like, is shown in schematic form. The sensor 1 comprises a memory 2 for storing correction parameters, an interface 3 for reading and / or adjusting the correction parameters, and a correction device 4 for correcting measured values ​​of the sensor based on the provided correction parameters. Fig. 4 shows a calibration device according to an embodiment of the present invention. Figure 4 shows in detail a calibration device 10 for calibrating a sensor with at least two axes. The calibration device 10 comprises a receiving device 11 for the defined receiving of a sensor with at least two axes 1. The receiving device 11 is connected to an excitation device 12. The excitation device 12 is configured for the defined individual excitation of the sensor for multiple excitation directions, wherein the number of excitation directions corresponds at least to the number of axes of the sensor, and in particular, wherein the excitation directions are perpendicular to each other. A measuring device 13 is connected to the excitation device 12 and / or the receiving device 11, and is configured to measure values ​​of all sensors along each of their axes for each individual excitation.The measuring device 13 forwards the measured values ​​to a computing unit 14, which uses the measured values ​​to determine the cross-sensitivities and calculates correction parameters for the measurement signals of the sensor 1. The computing unit 14 transmits the values ​​to a provisioning unit 15. The provisioning unit 15 comprises a memory 15a and an interface 15b. The provisioning unit 15 is configured to provide the correction parameters for the sensor's signal path for the output of corrected measured values. In detail, the calibration device 10 works as follows. In n-axis sensor systems, n defined stimuli are applied to the recording device 11 by the excitation device 12, and these stimuli are positioned exactly perpendicular to each other. For each stimulus, all n sensor axes are measured by the measuring device 13. From this, an n x n matrix is ​​generated, which allows for the determination of alignment errors, including positioning errors within the recording device 11, as well as intrinsic cross-sensitivities. In multi-sensor systems, a constant positioning error within the recording device allows the positioning errors of the subsystems relative to each other to be determined from the measured alignment information of the subsystems. Both pieces of information, i.e., intrinsic cross-sensitivities and positioning errors, are stored in the sensor's memory.To compensate for the transverse sensitivity, the sensor system can be computationally realigned in subsequent signal processing by feeding signal components from one or more axes to another or more axes. This can be achieved using a simple algorithm with multiplication and addition functions, either directly within the sensor's ASIC or subsequently via an internal microcontroller / FPGA or an external processing unit. The stored orientation information for multi-sensor systems can be made available to the customer for further alignment operations, thus providing added value. Using the intrinsic transverse sensitivities and positioning errors stored in memory, or the corresponding correction parameters for the individual subsystems, it is possible to computationally rotate all subsystems to a customer platform, such as a mobile phone or similar device.With a multi-sensor system featuring accelerometers and gyroscopes, a customer can align the sensor system and their platform based on the accelerometer readings and calculate the gyroscope readings using the stored intrinsic lateral sensitivities and positioning errors, or the corresponding correction parameters. This eliminates the need for the customer to define a gyroscope rate; instead, they can align the entire multi-sensor system using, for example, the acceleration due to gravity, which is significantly simpler and more reliable. Fig. 5 shows steps of a method for calibrating a sensor system according to an embodiment of the present invention. Figure 5 shows in detail the steps of a procedure for calibrating a sensor system. The procedure comprises the following steps. In a first step T1, at least one defined, individual excitation of each sensor in each subsystem is performed per excitation direction for several excitation directions, where the number of excitation directions corresponds at least to the number of axes of the respective sensor. In a second step T2, values ​​of all sensors are measured along each of their axes for each individual excitation. In a further step T3, cross-sensitivities are determined from the measured values. In a further step T4, correction parameters are calculated from the determined cross-sensitivities. In a further step T5, the correction parameters for a signal path of the respective subsystem are provided for the output of corrected measured values. Fig. 6 shows steps of a method for calibrating a device with a sensor system according to an embodiment of the present invention. Figure 6 shows in detail the steps of a method for calibrating a device with a sensor system according to an embodiment of the present invention. The method comprises the following steps: In a first step V1, correction parameters are provided for the sensor system according to a method according to one of claims 7-8. In a second step V2, the sensor system is arranged in the device. In a further step V3, the position and / or orientation of at least one sensor of the sensor system in the device is determined. In a further step V4, adjustment parameters are calculated from the determined position and / or orientation for the at least one sensor. In a further step V5, adjustment parameters for the sensor system are calculated based on the calculated adjustment parameters for the at least one sensor.In a further step V6, adapted correction parameters are provided based on the calculated adaptation parameters and the provided correction parameters for a signal path of the sensor system to output corrected measured values ​​from the sensor system. In summary, at least one embodiment of the invention offers at least one of the following advantages: • Reduction of cross-sensitivities between the sensor's measuring axes. • Easy determination of the alignment error in the end application of sensor systems. • Higher precision in sensor fusion applications. • Reduction of false signals. • Higher accuracy. Although the present invention has been described using preferred embodiments, it is not limited to these, but can be modified in many ways.

Claims

Method for calibrating a sensor (1) with at least two axes, comprising the steps: - At least one defined excitation (S1) of the sensor (1) per excitation direction for multiple excitation directions, wherein the excitation directions are different, the number of excitation directions corresponding at least to the number of axes of the sensor (1); - Measuring (S2) all values ​​along all axes for each individual excitation; - Determining (S3) cross-sensitivities from the measured values; - Calculating (S4) correction parameters from the determined cross-sensitivities; - Providing (S5) the correction parameters for a signal path of the sensor (1) for output of corrected measured values, wherein alignment errors are determined based on the cross-sensitivities and axis deviation errors of axes of the sensor (1). Method according to claim 1, wherein the correction parameters are stored in a memory (2) of the sensor (1) and made available for readout via an interface (3). Method according to one of claims 1-2, wherein after providing the correction parameters and after installing the sensor (1) in a device, the sensor (1) is calibrated in its installation position in the device and the provided correction parameters are adjusted. Method according to one of claims 1-3, wherein the excitation directions are provided perpendicular to each other. Method according to one of claims 1-4, wherein adjusted values ​​are calculated from the measured values, from which the cross-sensitivities are then determined. Method for calibrating a sensor system, comprising at least two subsystems, each with at least one sensor (1), comprising the steps: - At least one defined, single excitation (T1) of each sensor (1) in each subsystem per excitation direction for multiple excitation directions, wherein the number of excitation directions corresponds at least to the number of axes of the respective sensor (1), - Measuring (T2) values ​​of all sensors (1) along each of their axes for each single excitation, - Determining (T3) cross-sensitivities from the measured values, - Calculating (T4) correction parameters from the determined cross-sensitivities, and - Providing (T5) the correction parameters for a signal path of the respective subsystem for outputting corrected measured values, wherein relative alignment errors of the at least two subsystems to each other are determined on the basis of the determined cross-sensitivities. Sensor (1) with at least two axes, calibrated according to a method according to one of claims 1-6, comprising a memory (2) for storing the provided correction parameters and a correction device (4) for correcting measured values ​​of the sensor (1) based on the provided correction parameters. Sensor (1) according to claim 7, wherein the sensor (1) has at least one interface (3) for reading and / or adjusting the correction parameters. A method for calibrating a device with a sensor system, wherein the sensor system comprises at least two subsystems, each with at least one sensor (1), comprising the steps: - providing (V1) correction parameters according to a method according to claim 6 for the sensor system, - arranging (V2) the sensor system in the device, - determining (V3) a position and / or orientation of at least one sensor (1) of the sensor system in the device, - calculating (V4) adjustment parameters from the determined position and / or orientation for the at least one sensor (1), - calculating (V5) adjustment parameters for the sensor system based on the calculated adjustment parameters for the at least one sensor (1), - providing (V6) adjusted correction parameters based on the calculated adjustment parameters and the provided correction parameters for a signal path of the sensor system for outputting corrected measured values ​​of the sensor system. Calibration device (10) for calibrating a sensor (1) with at least two axes, comprising a receiving device (11) for receiving the sensor (1) with at least two axes in a defined manner, an excitation device (12) for the defined individual excitation of the sensor (1) per excitation direction for several excitation directions, wherein the number of excitation directions corresponds at least to the number of axes of the sensor (1), in particular wherein the excitation directions are perpendicular to each other, and a measuring device (13) which is connectable to the sensor (1) and which is configured to measure values ​​of all sensors (1) along each of their axes for each individual excitation, and a computing device (14) configured to determine cross-sensitivities from the measured values.for calculating correction parameters from the determined transverse sensitivities and for determining alignment errors based on the transverse sensitivities and axis deviation errors of the axes of the sensor (1), and a provisioning device (15), in particular comprising a memory (15a) and an interface (15b), configured for providing the correction parameters for the signal path of the sensor (1) for outputting corrected measured values.

Images