MEMS gyroscope installation error calibration method based on large-range autocollimator

Abstract

This invention claims protection for a MEMS gyroscope installation error calibration method based on a large-range autocollimator, comprising the following steps: Step 1, designing an autocollimator with an angle measurement range of ±9° using the reflection angle sensitivity modulation mechanism of a non-standard corner cone mirror (HCCCR); Step 2, establishing an evaluation benchmark for three-axis MEMS gyroscope measurement data by orthogonally combining two large-range autocollimators; Step 3, establishing a common data acquisition serial port between the autocollimator camera and the IMU host computer, and achieving dimensional unification and time synchronization between the autocollimator and MEMS gyroscope output readings by integrating the MEMS gyroscope angular rate measurement readings with time; Step 4, calibrating the MEMS gyroscope installation error coefficient using a traditional precision angular rate output turntable and conducting a comparative experiment. The experiment shows that the relative deviation of the IMU output after compensation by the two calibration methods is less than 0.3%.

Description

Technical Field

[0001] This invention belongs to the field of optical measurement instruments. Specifically, it proposes a method for calibrating the installation error of the gyroscope by using two orthogonal large-range autocollimators and MEMS gyroscopes to synchronously output three-axis attitude sensing readings. Background Technology

[0002] MEMS (Micro Electro Mechanical Systems) gyroscopes, as miniaturized, low-cost, and high-frequency response angular rate sensors, possess the important characteristics of wearability and easy integration. They are rapidly evolving from a single application of beacon tracking into crucial sensors in multiple fields such as mechanical control, personnel navigation, and material positioning. When using MEMS gyroscopes for three-dimensional motion sensing and measurement of targets, it is essential to strictly control their orthogonality, ensuring that their measurement axes coincide with the XYZ coordinate system of the Inertial Measurement Unit (IMU) to avoid linear misinterpretations of the measured values. Currently, a precision three-axis angular rate output turntable is primarily used as a calibration benchmark to measure the misinterpreted readings of the MEMS gyroscope output, thereby calibrating the installation errors. However, precision three-axis turntables lack traceable benchmarks, have extremely high assembly and operation standards, and their high-frequency calibration cycles and stringent configuration conditions are difficult for many laboratories to withstand.

[0003] To date, numerous inertial sensing and precision measurement laboratories worldwide have conducted research on the application of optically traceable measurement technology in turntable angle reading calibration and gyroscope error calibration, as documented by Sachin Nadig (S. Nadig, S.). B. Clark and A. Lal, “DOME-DISC: Diffractive optics metrology enabled dithering inertial sensor calibration,” in Proc. IEEE 27th Int. Conf. on MicroElectro Mechanical Systems (MEMS), San Francisco, CA, USA, 2014, pp. 608-611., designed a method for measuring the scale factor and bias of a gyroscope on a jitter stage using a Nano Optical Scale Imaging System (NORIS). However, the calibration accuracy of this scheme is affected by the number of pixels in the imaging system and the system integration level, and further improvements are needed in the experimental details. H. Schwenke (in the literature H. Schwenke, R. Schmitt, P. Jatzkowski, C. Warmann, “On-the-fly calibration of linear and rotary axes of machine tools and CMMs using a tracking…”) The interferometer, “CIRP Annals, vol. 58, no. 1, pp. 477-480, 2009.”, uses laser tracking for the six-degree-of-freedom calibration of rotating machine tool axes. Although laser tracking has a certain reliability for motion measurements of machine tools at static and different speeds, the higher the speed, the greater the difference in continuous measurements. Jiapeng Mou (J. Mou, J. Su, L. Miao, and T. Huang, “Research on field application technology of dynamic angle measurement based on fiber optic gyroscope and autocollimator,” IEEE Sensors Journal, vol. 21, no. 13, pp. 15308-15317, July 1, 2021.) uses an optical autocollimator to calibrate the scaling factor of a fiber optic gyroscope (FOG) within a range of ±0.05°. Among these studies, the optical autocollimator has received more attention due to its theoretical advantages of easy structural integration and high angle measurement accuracy. However, its aperture size limits its angular measurement range, making it difficult to use for error correction in MEMS gyroscopes with a large dynamic range.

[0004] This paper discloses a method for calibrating the installation error of MEMS gyroscopes based on a large-range autocollimator. Utilizing the traceability of the autocollimator and overcoming its limited measurement range, an autocollimator with a ±9° measurement range was designed. A data evaluation benchmark for the MEMS gyroscope was then constructed using a spatially orthogonal structure of two large-range autocollimators. Under the control of a computer synchronously acquiring time serial port, the calibration measurement of the MEMS gyroscope installation error was completed. Summary of the Invention

[0005] This invention aims to solve the problems of the prior art. It proposes a method for calibrating the installation error of a MEMS gyroscope based on a large-range autocollimator. The technical solution of this invention is as follows:

[0006] A method for calibrating the installation error of a MEMS gyroscope based on a large-range autocollimator, comprising the following steps:

[0007] Step 1: Using the reflection angle sensitivity modulation mechanism of the non-standard corner cone HCCCR, an autocollimator with an angle measurement range of ±9° was designed.

[0008] Step 2: By utilizing the orthogonal combination of two large-range autocollimators, an evaluation benchmark for the measurement data of a three-axis MEMS gyroscope is established.

[0009] Step 3: Establish a common data acquisition serial port between the autocollimator camera and the IMU host computer. By integrating the angular rate measurement readings of the MEMS gyroscope with time, the dimensional unification and acquisition time synchronization between the output readings of the autocollimator and the MEMS gyroscope are achieved.

[0010] Step 4: Calibrate the MEMS gyroscope installation error coefficient using a traditional precision angular rate output turntable, and conduct comparative tests to verify the reliability of this method.

[0011] Furthermore, in step 1, utilizing the reflection angle sensitivity modulation mechanism of a non-standard corner cone (HCCCR), an autocollimator with an angle measurement range of ±9° was designed, specifically including:

[0012] By rotating the standard corner cube mirror, two non-standard corner cube mirrors were designed with included angles of ∠2_3 = 90°13', ∠3_1 = 89°47', and ∠1_2 = 90°13', respectively. Replacing the plane mirror with these non-standard corner cube mirrors as the reflector structure, a large-range autocollimator measurement system was obtained. When the non-standard corner cube mirrors rotate with the object being measured, the large-range autocollimator reads the output voltage of the PSD to realize the displacement conversion of the light spot on the PSD plane. and Complete the attitude angle measurement of non-standard corner cones.

[0013] Furthermore, the normal vectors of the three reflecting surfaces of the non-standard corner cube mirror in step 1 are derived from the angle relationship between the mirror surfaces after rotation of the standard corner cube mirror as follows:

[0014]

[0015] Where -δ 12 δ is the angle of rotation of mirror 1 along the OZ axis. 23 It is the angle of rotation of mirror 2 along the OX axis, δ 13 It is the angle of rotation of mirror 1 along the OY axis; based on the above process of obtaining the non-standard corner cone, it can be deduced that if δ1=δ, δ2=δ, δ3=-δ, where δ is the mathematical relationship angle between the three reflecting surfaces of the non-standard corner cone, then the vector change of the measurement beam after reflection by the 3-2-1 reflecting surface is:

[0016]

[0017] Therefore, the reflection angle sensitivity of a non-standard corner cone is:

[0018]

[0019] According to formula (3), the sensitivity of the reflection angle of the non-standard corner cone can be adjusted by setting the value of the connection angle δ. Two non-standard corner cones with mathematical connection angle δ = 13' and mirror angles of ∠2_3 = 90°13', ∠3_1 = 89°47', and ∠1_2 = 90°13' were designed.

[0020] Traditional autocollimators use plane mirrors as reflectors. When the plane mirror rotates with the object being measured, producing yaw angles Θ1 and pitch angles Θ2, the collimated beam will tilt, resulting in a change in the beam spot displacement on the PSD detector. The mathematical description of this process is as follows:

[0021] ΔX=f·2·tan(Θ1) (4)

[0022] ΔY=f·2·tan(Θ2) (5)

[0023] Where f is the focal length of the collimating lens, ΔX is the displacement of the light spot along the X0 axis, and ΔY is the displacement of the light spot along the Y0 axis; if the non-standard corner cone mirror is replaced with the plane mirror, then formulas (4) and (5) should be rewritten as:

[0024]

[0025]

[0026] Measurements were performed using the NIKON-6D autocollimator equipped with a Sony MI-20 CMOS sensor, which can increase the measurement range from ±30° to ±9°, with a system resolution of less than 4″.

[0027] Furthermore, step 2, establishing an evaluation benchmark for three-axis MEMS gyroscope measurement data by utilizing an orthogonal combination of two large-range autocollimators, specifically includes:

[0028] Since the autocollimator is only sensitive to yaw angles Θ1 and pitch angles Θ2, but not to roll angle Θ3, in order to obtain three-axis motion angle sensing, two autocollimators are calibrated using two adjacent coated surfaces of a high-precision cubic reflector module, so that they are respectively aimed at the coordinate axes of the reflector module and adjusted to be mutually orthogonal. The cubic reflector module is replaced with a non-standard corner cube mirror. By restoring the light spot state on the detection surface of the autocollimator, the alignment of the two autocollimators with the non-standard corner cube mirror is completed. When the turntable rotates Θ1 along the yaw X-axis and Θ2 along the pitch Y-axis, autocollimator 1 serves as the calibration reference for the gyroscope; when the turntable rotates Θ3 along the roll Z-axis, autocollimator 2 serves as the calibration reference.

[0029] Furthermore, in step 2, the rotation angle of each axis of the three-axis turntable is read simultaneously by an autocollimator and a three-axis MEMS gyroscope. In order to obtain high-precision three-dimensional motion angle perception, two large-range autocollimators are orthogonally combined to complete the reference angle measurement of the IMU three-axis attitude.

[0030] Using two adjacent coated surfaces of a high-precision cube reflector module, two autocollimators are aligned. By observing that the crosshairs of the two autocollimators are located at the center of the sensor, the two autocollimators are adjusted to be orthogonal to each other. Next, the spatial attitude angles of the cube and the two autocollimators are finely adjusted until the light spot of autocollimator 1 remains stationary while the turntable rotates along the Z-axis, and the light spot of autocollimator 2 moves only horizontally along the detection plane. This step ensures that the detection surfaces of autocollimators 1 and 2 coincide with the XOY and XOZ coordinate planes of the cube, respectively. Finally, a non-standard corner cube reflector is used to replace the cube reflector module. Based on the principle of planar reflection of the aperture surface, the light spot state on the detection surface of the autocollimator is observed, thus completing the alignment of the autocollimator with the non-standard corner cube emitter.

[0031] Furthermore, step 3, establishing a common data acquisition serial port between the autocollimator camera and the IMU host computer, and achieving dimensional unification and acquisition time synchronization between the autocollimator and MEMS gyroscope output readings by integrating the angular rate measurement readings with time using the MEMS gyroscope, specifically includes:

[0032] Two autocollimators were used to measure the attitude angle of a non-standard corner cone mirror, and the readings from the three-axis gyroscope in the collimator IMU were compared. During the comparison, a computer sent acquisition commands to the common serial port of the autocollimator PSD and the IMU corresponding to the measurement task, thus synchronizing the acquisition time between the two systems. The exposure time of the PSD was used as the data comparison time for one IMU acquisition. θ i =[(ω i +ω i+1 )]*[(t i+1 -t i The angle measured by the MEMS gyroscope can be obtained by ]] / 2, realizing the dimensional unification between the output readings of the autocollimator and the MEMS gyroscope.

[0033] Furthermore, the MEMS gyroscope installation error calibration system in step 3 consists of two mutually orthogonal large-range autocollimators and a manual three-axis turntable. When the three-axis turntable rotates sequentially along its standard axis, it provides the IMU and non-standard corner cones with corresponding yaw Θ1, pitch Θ2, or roll Θ3 rotation angles. When the turntable starts rotating, the computer sends acquisition commands to the autocollimator PSD and IMU via a common serial port, enabling them to simultaneously measure readings. When the turntable rotates along the yaw X-axis Θ1 and Y-axis pitch Θ2 angles, autocollimator 1 serves as the calibration reference for the gyroscope readings, and when the turntable rotates along the Z-axis roll Θ3 angle, autocollimator 2 serves as the reading reference.

[0034] The turntable performs measurements within a range of ±9°. The exposure time of the PSD is used as the comparison time for one IMU data acquisition. Since the MEMS gyroscope acquires data n times at the same frequency within the PSD exposure time, the angle measured by the gyroscope is:

[0035]

[0036] Where ω i Θ is the angular velocity of the three-axis MEMS gyroscope during a single PSD exposure, where i = 1, 2, 3…n, and t is the sampling interval of the gyroscope; i This corresponds to the angle measured by the MEMS gyroscope within one exposure time of the autocollimator; by comparing the attitude angle measurement values ​​of the non-standard corner cone mirror by two autocollimators and the attitude angle sensing values ​​of the three-axis gyroscope in the IMU, the installation error of the MEMS gyroscope is calibrated.

[0037] Furthermore, step 4, calibrating the MEMS gyroscope installation error coefficient using a traditional precision angular rate output turntable, specifically includes:

[0038] An IMU host computer equipped with a three-axis MEMS gyroscope and a non-standard corner cube mirror were rigidly fixed on a PT5 three-axis manual turntable. The three-dimensional turntable was rotated within ±9° along its coordinate axes. The measurements from the three-axis MEMS gyroscope and the large-range autocollimator within each PSD exposure interval were linearly fitted and averaged. The installation error coefficient of the three-axis MEMS gyroscope was recorded as K. ij (i = x, y, z, j = x, y, z); then, the reliability of the present invention is verified by calibrating the installation error coefficient of the MEMS gyroscope using a traditional precision angular rate output turntable; finally, the installation error coefficients calibrated by the two methods are used for data compensation of the experimental results.

[0039] Furthermore, in step 4, the PT5 type three-dimensional manual turntable is rotated within a range of ±9° along yaw Θ1, pitch Θ2, and roll Θ3 angles, respectively, so that the three-axis MEMS gyroscope obtains the rotation angle measured by the large-range autocollimator at each measurement reading. The linear slope of the ratio of the three-axis MEMS gyroscope and the large-range autocollimator within each PSD exposure interval, after averaging, can be used to calculate the installation error coefficient K of the three-axis MEMS gyroscope. ij (i = x, y, z, j = x, y, z).

[0040] Using a precision triaxial angular rate output turntable I6082 (measurement range 0.001° / s~300° / s, measurement accuracy better than 0.05%), the triaxial mounting error of the MEMS gyroscope was measured in steps of 20° / s within a range of ±100° / s, and the mounting error coefficient was recorded as K'. ij (i = x, y, z, j = x, y, z); The installation error coefficients determined by the two methods are used for data compensation of the experimental results, and the compensation method is expressed as follows:

[0041]

[0042] It indicates that the carrier has an angular velocity. During input, the output matrix of the MEMS gyroscope, ωx0, ωy0, and ωz0, represents the zero-position output of the gyroscope when it is stationary, and their values ​​are very small and can be ignored. Experiments show that the relative deviation of the IMU output after compensation by the two calibration methods is less than 0.3%.

[0043] The advantages and beneficial effects of this invention are as follows:

[0044] This invention proposes a MEMS gyroscope installation error calibration method based on a large-range autocollimator. By designing a non-standard corner cone with a mathematical connection angle δ of 13', the angular measurement range of the NIKON-6D autocollimator is increased from ±30' to ±9°. Using a measurement system with two orthogonally aligned large-range autocollimators, the attitude characterization of the yaw, pitch, and roll rotation processes of the IMU to be calibrated is achieved. Finally, a computer sends serial port signals to the autocollimator and the IMU's host computer to achieve synchronous data acquisition. By integrating the MEMS gyroscope angular rate measurement readings with time, dimensional unification and time synchronization between the output readings of the large-range autocollimator and the MEMS gyroscope are achieved. Experiments show that this method achieves a calibration accuracy deviation of less than 0.3% for MEMS gyroscope installation errors compared to traditional precision three-axis turntable calibration methods.

[0045] Previously, the calibration of MEMS gyroscope installation errors mainly relied on precision triaxial angular rate output turntables as the calibration benchmark. However, due to the lack of traceable benchmarks, the assembly and operation specifications of precision triaxial turntables place extremely high demands on experimental personnel, and the high-frequency calibration cycle and stringent configuration conditions are difficult for many laboratories to bear. Based on previous research, although a series of error sources of gyroscopes can be calibrated using traceable optical technologies such as laser interferometers and nano-optical scale imaging, these technologies are generally limited by the size of the optical aperture, resulting in a very small angular measurement range. Therefore, they are difficult to use for installation error calibration of MEMS gyroscopes, which have error characteristics manifested under a large dynamic range. This method overcomes the limitation of insufficient measurement range of autocollimators, which helps to promote the development of inertial sensing calibration technology towards traceability, thereby significantly reducing and simplifying laboratory equipment costs and maintenance requirements. Attached Figure Description

[0046] Figure 1 This is a theoretical structural diagram of a non-standard corner cone reflector provided by the present invention in a preferred embodiment;

[0047] Figure 2 This is a schematic diagram of a large-range autocollimator measurement system;

[0048] Figure 3 This is an orthogonal state spot imaging diagram based on a large-range autocollimator using a non-standard corner cone.

[0049] Figure 4 This is a schematic diagram of a MEMS gyroscope installation error calibration platform;

[0050] Figure 5 This is a schematic diagram of a data acquisition system. Detailed Implementation

[0051] The technical solutions of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0052] The technical solution of the present invention to solve the above-mentioned technical problems is:

[0053] A method for calibrating the installation error of a MEMS gyroscope based on a large-range autocollimator, comprising the following steps:

[0054] 1) Based on the reflection angle sensitivity factor in the mathematical model of a standard corner cone, by rotating the standard corner cone mirror to change the normal vector of each reflecting surface, and then adjusting the mathematical connection angle according to the changed reflection vector matrix, two non-standard corner cone mirrors were designed with mirror angles ∠2_3=90°13', ∠3_1=89°47', and ∠1_2=90°13'. Replacing the plane mirror with non-standard corner cone mirrors as the reflector structure, a large-range autocollimator measurement system was obtained. When the non-standard corner cone mirrors rotate with the object being measured, the large-range autocollimator reads the output voltage of the PSD to realize the displacement conversion of the light spot on the PSD plane. and The attitude angle measurement of the non-standard corner cone was completed. To ensure that the imaging spot is within the maximum allowable displacement range of the photoelectric sensor plane, the mathematical connection angle δ of the non-standard corner cone was designed to be 13'. Through experiments, it was verified that the angle measurement range of the NIKON-6D autocollimator was extended from ±30' to ±9°, while ensuring a resolution higher than 4″ and an overall accuracy better than 18″.

[0055] 2) Since the autocollimator is only sensitive to yaw Θ1 and pitch Θ2 angles, but not to roll Θ3, to obtain three-axis motion angle sensing, two adjacent coated surfaces of a high-precision cubic reflector module are aligned with two large-range autocollimators. By observing that the crosshair spots are located at the center of the PSD, they are aligned with the coordinate axes of the reflector module. Then, the spatial attitude angle of the cubic mirror is finely adjusted until the spot of the large-range autocollimator 1 remains stationary during the rotation of the turntable along the Z-axis, while the spot of the large-range autocollimator 2 moves only horizontally along the detection plane, ensuring that the two large-range autocollimators are orthogonal to each other. Then, a non-standard corner cube mirror is used to replace the cubic reflector. By restoring the spot state on the PSD, the alignment of the two large-range autocollimators and the non-standard corner cube mirror is completed. This ensures that when the turntable rotates Θ1 along the yaw X-axis and Θ2 along the pitch Y-axis, autocollimator 1 serves as the calibration reference for the gyroscope; when the turntable rotates Θ3 along the roll Z-axis, autocollimator 2 serves as the calibration reference.

[0056] 3) By rotating the three-axis turntable sequentially along its coordinate axes, the IMU host computer and the non-standard corner cube mirror are given corresponding yaw, pitch, and roll angles. The attitude angle measurements of the non-standard corner cube mirror are compared with the attitude angle sensing values ​​from the three-axis gyroscope in the collimator IMU using two autocollimators. During the comparison of the two system readings, the computer sends acquisition commands to the common serial port of the autocollimator PSD and the IMU corresponding to the measurement task, thus achieving time synchronization between the two systems. The exposure time of the PSD is used as the data comparison time for one IMU acquisition. θ i =[(ω i +ω i+1 )]*[(t i+1 -t i The angle measured by the MEMS gyroscope can be obtained by ]] / 2, realizing the dimensional unification between the output readings of the autocollimator and the MEMS gyroscope.

[0057] 4) Using this method, the angular velocity measured by the three-axis MEMS gyroscope during each exposure time of the PSD is obtained. By integrating with the sampling time, the measured rotation angle is obtained. The measured values ​​of the three-dimensional MEMS gyroscope and the large-range autocollimator are linearly fitted and averaged. The installation error coefficient of the three-axis MEMS gyroscope is recorded as K. ij (i = x, y, z, j = x, y, z); then, the method of calibrating the installation error coefficient of MEMS gyroscope using the traditional precision angular rate output turntable is used to verify the actual effect of the calibration of installation error in this method; finally, the installation error coefficients calibrated by the two methods are used for data compensation of the experimental results. The compensation results show that the relative deviation of the IMU output after compensation by the two calibration methods is less than 0.3%.

[0058] 2. Preferably, the reflection vector of the light beam after passing through each reflecting surface of the standard corner cube in step 1) is:

[0059]

[0060] M d Let N be the reflection matrix of each reflecting surface of the cornerstone. x N y N z These are the normal vectors of each reflecting surface in the XYZ coordinate system. If the incident beam is reflected in the reflection sequence of the standard corner cube mirror 3-2-1, the outgoing light vector can be expressed as:

[0061] B 321 =M3·M2·M1·A (2)

[0062] A is the unit vector of the incident light hitting the corner cone. According to formulas (1) and (2), the beam vector after reflection by the standard corner cube mirror is... According to the law of reflection, incident light reflected by a standard corner cube will always be parallel to outgoing light, and its reflection angle sensitivity is 1.

[0063] The non-standard corner cone mirror designed in this invention is obtained by continuously rotating a standard corner cone mirror. There is an angle ∠2-3 = 90°-δ2 between mirror surfaces 2 and 3, an angle ∠1-3 = 90°-δ3 between mirror surfaces 1 and 3, and an angle ∠1-2 = 90°-δ1 between mirror surfaces 1 and 2. The normal vector of the rotating mirror surfaces can be obtained as follows:

[0064]

[0065] Where -δ 12 δ is the angle of rotation of mirror 1 along the OZ axis. 23 It is the angle of rotation of mirror 2 along the OX axis, δ 13 It is the angle of rotation of mirror 1 along the OY axis. Based on the above process of obtaining a non-standard corner cone, we know that δ2 = δ 23 ,δ3=δ 13 Therefore, the angle between normal vectors N1 and N2 can be expressed as:

[0066]

[0067] Substituting formula (3) into formula (4) yields δ 12 The expression is

[0068]

[0069] When δ1, δ2, and δ3 are small angles, formula (5) can be approximately equal to

[0070] δ 12 =δ1+δ2·δ3 (6)

[0071] Based on the reflection matrices of each reflecting surface of the corner cone, if δ1=δ, δ2=δ, δ3=-δ, where δ is the mathematical relationship angle between the three reflecting surfaces of the non-standard corner cone, then the vector change of the measurement beam after reflection by the 3-2-1 reflecting surface is:

[0072]

[0073] Therefore, the reflection angle sensitivity of a non-standard corner cone is:

[0074]

[0075] According to formula (8), the sensitivity of the reflection angle of the non-standard corner cone can be adjusted by setting the value of the connection angle δ. This invention designs two non-standard corner cones with a mathematical connection angle δ = 13' and mirror angles of ∠2_3 = 90°13', ∠3_1 = 89°47', and ∠1_2 = 90°13', respectively. This allows the non-standard corner cone to be used as a reflector in an autocollimator, so that the reflected beam is not limited by the aperture size of the collimating objective lens. Within a large range of angular changes of the reflector, it can always form an image on the photodetector.

[0076] Taking the NIKON-6D autocollimator equipped with a Sony MI-20 CMOS sensor as an example, its measurement range can be increased from ±30' to ±9°, while ensuring a resolution higher than 4″ and an overall accuracy better than 18″. Based on the test performance of the above-mentioned large-range autocollimator, it can be seen that this method is suitable for dynamic mounting error calibration tests of MEMS gyroscopes with angular rate resolution lower than 4″ / 0.55s = 0.002° / s and noise higher than 18″ / 0.55s = 0.009° / s, where 0.55s is the minimum exposure time of the CMOS.

[0077] 3. Preferably, in step 2), considering that the autocollimator is only sensitive to the yaw Θ1 and pitch Θ2 angles, in order to obtain high-precision three-dimensional motion angle perception, we used two large-range autocollimators orthogonally combined to complete the reference angle measurement of the IMU's three-axis attitude.

[0078] Two large-range autocollimators were aligned with two adjacent coated surfaces of the high-precision cube reflector module. The crosshairs of the two autocollimators were observed to ensure they were centered on the sensor. Then, the spatial attitude angles of the cube and the two autocollimators were finely adjusted until the spot of autocollimator 1 remained stationary while the turntable rotated along the Z-axis, and the spot of large-range autocollimator 2 moved only horizontally along the detection plane. This ensured that the detection surfaces of autocollimators 1 and 2 coincided with the XOY and XOZ coordinate planes of the cube, respectively. Next, a non-standard corner cube reflector was replaced with a non-standard corner cube. Based on the planar reflection principle of the non-standard corner cube aperture, the spot state on the autocollimator's detection surface was observed, completing the alignment of the autocollimator with the non-standard corner cube emitter. Finally, the non-standard corner cube and IMU were rigidly fixed on a PT5 three-axis manual turntable. The three-axis turntable rotated within ±9° along each axis, and readings were simultaneously obtained through the autocollimators and the three-axis MEMS gyroscope.

[0079] 4. Preferably, in step 3), when the turntable starts to rotate, the computer sends a data acquisition command to the large-range autocollimator PSD camera and IMU via a common serial port, so that they can simultaneously measure the readings. When the turntable rotates along the yaw X-axis Θ1 and the pitch Y-axis Θ2 angle, the large-range autocollimator 1 serves as the calibration reference for the gyroscope readings. When the turntable rotates along the roll Z-axis Θ3 angle, the large-range autocollimator 2 serves as the reading reference.

[0080] The turntable performs measurements within a range of ±9°. The rotation angle measured by the MEMS gyroscope is obtained by integrating the angular velocity measured by the three-axis MEMS gyroscope within each exposure time of the PSD with the sampling time. The calculation method is expressed as follows:

[0081]

[0082] Where ω i (i = 1, 2, 3…n) represents the angular velocity of the three-axis MEMS gyroscope during a single PSD exposure, and t is the gyroscope's sampling interval. Θ i (i = 1, 2, 3...n) represents the angle measured by the MEMS gyroscope within one exposure time of the autocollimator.

[0083] 5. Preferably, in step 4), the installation error coefficient K of the three-axis MEMS gyroscope can be obtained by comparing the measured values ​​of the three-axis MEMS gyroscope and the large-range autocollimator within each PSD exposure interval and averaging the results (the linear slope of the ratio of the measured readings). ij (i = x, y, z, j = x, y, z), K ij The angle sensed by the MEMS gyroscope along the j-axis when the turntable rotates along the i-axis is the ratio of the input angle to the angle sensed by the gyroscope. The X-axis corresponds to the yaw angle Θ1, the Y-axis to the pitch angle Θ2, and the Z-axis to the roll angle Θ3. To verify the practical effect of calibrating the gyroscope installation error coefficient based on this invention, a comparative experiment was conducted using a traditional precision angular rate output turntable to calibrate the MEMS installation error coefficient. The experiment involved modulating the precision turntable within a range of ±100° / s, varying it in 20° / s increments, and measuring the three-axis installation error of the MEMS gyroscope in the IMU. The experimental data were then linearly fitted and recorded as the installation error coefficient K'. ij (i = x, y, z, j = x, y, z).

[0084] The installation error coefficients calibrated by the two methods are used for data compensation of the experimental results. The compensation method is expressed as follows:

[0085]

[0086] After processing, the error values ​​of the three-axis MEMS gyroscope after calibration and compensation using both methods were suppressed to within 0.12° / s. To compare the deviations between the two installation error calibration methods, the measurement data were averaged. Analysis showed that the relative deviation of the IMU output after compensation by the two calibration methods was less than 0.3%.

[0087] Finally, according to this invention, a large-range autocollimator calibration platform for MEMS gyroscope installation errors with traceability was constructed, achieving a calibration accuracy deviation of better than 0.3% for MEMS gyroscope installation errors. Furthermore, the calibration platform parameters can calibrate and complete the calibration of installation error coefficients for MEMS gyroscopes with angular rate resolutions below 0.002° / s and noise levels above 0.009° / s. This method helps promote the development of inertial sensing calibration technology towards traceability, thereby significantly reducing and simplifying laboratory equipment costs and maintenance requirements.

[0088] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0089] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0090] The above embodiments should be understood as illustrative only and not as limiting the scope of protection of the present invention. After reading the description of the present invention, those skilled in the art can make various alterations or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A MEMS gyroscope installation error calibration method based on a large range autocollimator, characterized in that, The method comprises the following steps: Step 1, a collimator with a measuring angle range of ±9° is designed by using the reflection angle sensitivity modulation mechanism of a non-standard corner cube HCCCR; Step 2, an evaluation benchmark for three-axis MEMS gyroscope measurement data is built by using orthogonal combination between two large-range collimators; Step 3, a common data acquisition serial port between the collimator camera and the IMU host computer is built, and dimension unification and acquisition time synchronization between the output readings of the collimator and the MEMS gyroscope are realized by integrating the MEMS gyroscope angular rate measurement readings and time; Step 4, finally, the MEMS gyroscope installation error coefficient is calibrated by a traditional precise angular rate output turntable.

2. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 1, characterized in that, In step 1, the collimator with a measuring angle range of ±9° is designed by using the reflection angle sensitivity modulation mechanism of a non-standard corner cube HCCCR, and specifically comprises: By rotating the standard corner cube mirror, two non-standard corner cube mirrors are designed, whose mirror angles are ∠2_3=90°13′, ∠3_1=89°47′, ∠1_2=90°13′, respectively. Using the non-standard corner cube mirror to replace the plane mirror as the reflector structure, a large range of collimator measuring system is obtained. When the non-standard corner cube mirror rotates with the measured object, the large range of collimator realizes the conversion of the spot displacement in the PSD plane by reading the output voltage of the PSD, and according to and The attitude angle measurement of the non-standard corner cube mirror is completed.

3. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 2, characterized in that, The normal vector of the three reflecting surfaces of the non-standard corner cube is derived according to the mirror surface included angle relationship after the rotation of the standard corner cube, and is expressed as: where δ 12 is the angle of rotation of mirror 1 along the OZ axis, δ 23 is the angle of rotation of mirror 2 along the OX axis, δ 13 is the angle of rotation of mirror 1 along the OY axis; according to the obtaining process of the above non-standard corner cube mirror, it can be deduced that if δ1= δ, δ2= δ, δ3= - δ, where δ is the mathematical connection angle of the three reflection angles of the non-standard corner cube mirror, then the vector change of the measuring beam after being reflected by the 3-2-1 reflection surfaces is: Therefore, the reflection angle sensitivity of the non-standard corner cube is: According to formula (3), the reflection angle sensitivity of the non-standard corner cube can be adjusted by setting the contact angle δ value; two non-standard corner cubes with a mathematical contact angle δ of 13' and mirror surface included angles of ∠2_3=90°13', ∠3_1=89°47', and ∠1_2=90°13' are designed.

4. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 2, characterized in that, According to the traditional collimator, a plane mirror is used as a reflecting body, when the plane mirror is rotated by a yaw Θ1 and a pitch Θ2 angle with the measured object, the collimated light beam will be tilted, and the light spot displacement change is presented on the PSD detector, and the mathematical description of the process is: ΔX=f·2·tan(Θ1) (4) ΔY=f·2·tan(Θ2) (5) Wherein f is the focal length of the collimator, ΔX is the displacement of the light spot along the X0 axis, and ΔY is the displacement of the light spot along the Y0 axis; if the non-standard corner cube is replaced by the plane mirror, formulas (4) and (5) should be rewritten as: When the NIKON-6D collimator with a Sony MI-20 CMOS sensor is used, the measurement range can be increased from ±30' to ±9°, and the system resolution is less than 4''.

5. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 1, characterized in that, In step 2, an evaluation benchmark for three-axis MEMS gyroscope measurement data is built by using orthogonal combination between two large-range collimators, and specifically comprises: Since the collimator is only sensitive to the yaw Θ1 and the pitch Θ2 angles, and is not sensitive to the roll angle Θ3, in order to obtain three-axis motion angle sensing, two adjacent coated surfaces of a high-precision cube reflection module are used to calibrate two collimators, so that they are aimed at the coordinate axes of the reflection module and adjusted to be orthogonal to each other; the non-standard corner cube is used to replace the cube reflection module, the light spot state on the detection surface of the collimator is restored, the alignment of the two collimators and the non-standard corner cube is completed, so that when the turntable is rotated by a yaw X axis Θ1 and a pitch Y axis Θ2 angle, the collimator 1 is the calibration benchmark for the gyroscope; when the turntable is rotated by a roll Z axis Θ3 angle, the collimator 2 is the calibration benchmark.

6. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 5, characterized in that, The rotation angle of each axis of the step 2 three-axis turntable is obtained by the autocollimator and the three-axis MEMS gyroscope simultaneously; in order to obtain high-precision three-dimensional motion angle sensing, two large-range autocollimators are used in orthogonal combination design to complete the reference angle measurement of the IMU three-axis attitude; Two adjacent coated surfaces of the high-precision cube reflector module are used to align the two autocollimators, and the cross-shaped light spots of the two autocollimators are observed to be located at the center position of the sensor, and the two autocollimators are adjusted to be orthogonal to each other. Secondly, the space attitude angle of the square block and the two autocollimators is adjusted, until the light spot of the autocollimator 1 remains stationary during the rotation of the turntable along the Z axis, and at the same time, the light spot of the autocollimator 2 only moves horizontally along the detection plane. The detection surfaces of the autocollimators 1 and 2 are overlapped with the XOY and XOZ coordinate planes of the cube, respectively. Finally, the non-standard corner mirror is used to replace the cube reflector module, and based on the plane reflection principle of the aperture surface, the light spot state on the detection surface of the autocollimator is observed to complete the alignment of the autocollimator and the non-standard corner mirror emitter.

7. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 5, characterized in that, In step 3, the common data acquisition serial port between the autocollimator camera and the IMU host computer is built, and the dimension unification and acquisition time synchronization between the output readings of the autocollimator and the MEMS gyroscope are realized by integrating the angular rate measurement readings of the MEMS gyroscope with time, which specifically includes: The attitude angle sensing values of three-axis gyroscope in the non-standard angle pyramid attitude angle measurement are compared by using two autocollimators; in the comparison of two types of system values, the computer gives the acquisition instruction to the common serial port of the autocollimator PSD and IMU corresponding to the measurement task, so as to form the acquisition time synchronization between the two systems; the exposure time of the PSD is used as the comparison time of the acquisition data of the IMU, and the angle measured by the MEMS gyroscope is obtained through θ i = [(ω i + ω i+1 )] * [(t i+1 -t i )] / 2, which realizes the dimension unification between the output values of the autocollimator and the MEMS gyroscope.

8. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 7, characterized in that, The step 3 MEMS gyroscope installation error calibration system is composed of two mutually orthogonal large-range autocollimators and a manual three-axis turntable. When the three-axis turntable is rotated along its reference axes in turn, the corresponding yaw Θ1, pitch Θ2 or roll Θ3 rotation angles of the IMU and the non-standard corner mirror are given. When the turntable starts to rotate, the computer gives the autocollimator PSD and the IMU common serial port to execute the acquisition instruction, so that they can measure the readings at the same time. When the turntable rotates along the yaw X axis Θ1 and the pitch Y axis Θ2, the autocollimator 1 is the calibration reference for the gyroscope readings, and when the turntable rotates along the Z axis roll Θ3, the autocollimator 2 is the reading reference; The turntable is measured within ±9°, and the exposure time of the PSD is used as the comparison time of the acquisition data of the IMU. Since the MEMS gyroscope collects n times of data within the exposure time of the PSD, the angle measured by the gyroscope is: where ω i is the angular rate of the three-axis MEMS gyroscope collected in the first exposure time of the PSD, i = 1, 2, 3…n, and t is the sampling interval time of the gyroscope; Θ i is the angle measured by the MEMS gyroscope in the first exposure time of the autocollimator; by comparing the measured values of the non-standard angle cone mirror by two autocollimators and the perceived values of the attitude angle by the three-axis gyroscope in the IMU, the installation error of the MEMS gyroscope is calibrated.

9. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 1, characterized in that, In step 4, the installation error coefficient of the MEMS gyroscope is calibrated by the traditional precise angular rate output turntable, and the method is compared. Specifically, it includes: The IMU upper computer with a three-axis MEMS gyroscope and a non-standard corner cube mirror are rigidly fixed on a PT5 three-axis manual turntable, so that the three-dimensional turntable rotates within a range of ±9° along its coordinate axes, the measurement values of the three-axis MEMS gyroscope and the large-range autocollimator within the exposure interval of each PSD are linearly fitted and processed by averaging, and the installation error coefficient K of the three-axis MEMS gyroscope is recorded ij , i=x, y, z, j=x, y, z; then, the reliability of the application is verified by using a conventional precise angular rate output turntable to calibrate the installation error coefficient of the MEMS gyroscope; finally, the installation error coefficients calibrated by the two methods are respectively used for data compensation of experimental results.

10. The MEMS gyroscope installation error calibration method based on a large-range autocollimator according to claim 9, characterized in that, In step 4, the PT5 type three-dimensional manual turntable rotates within a range of ±9° along yaw Θ1, pitch Θ2, and roll Θ3 angles, respectively, so that the three-axis MEMS gyroscope obtains the rotation angle measured by the large-range autocollimator at each measurement reading. The linear slope of the ratio of the three-axis MEMS gyroscope and the large-range autocollimator within each PSD exposure interval, after averaging, can be used to calculate the installation error coefficient K of the three-axis MEMS gyroscope. ij i = x, y, z, j = x, y, z; The three-axis installation error of the MEMS gyroscope is measured by using a precision three-axis angular rate output turntable I6082, with a measurement range of 0.001° / s to 300° / s, a measurement accuracy of better than 0.05%, and a step change of every 20° / s within a range of ±100° / s, and installation error coefficients K' are recorded ij , i = x, y, z, j = x, y, z; the installation error coefficients calibrated by the two methods are respectively used for data compensation of experimental results, and the compensation method is represented as: which represents the angular velocity of the carrier The output matrix of MEMS gyroscope, ωx0, ωy0, ωz0, is the zero output of the gyroscope when it is at rest, and its value is very small and can be ignored. According to the experiment, the relative deviation of the IMU output after compensation by the two calibration methods is less than 0.3%.

Images