METHOD FOR CALIBRINGING A MICRO-FABRICATED INERTIAL ANGLE SENSOR AND CALIBRATION SYSTEM

Description

[0001] The present invention relates to a method for calibrating a micro-machined angular inertial sensor.

[0002] The present invention also relates to a calibration system configured to calibrate a micro-machined angular inertial sensor.

[0003] The invention relates to the field of embedded inertial angular sensors, particularly for navigation, piloting, guidance or heading search.

[0004] Such an angular inertial sensor is configured to measure an angular position, usually called a gyroscope, or to measure an angular velocity, usually called a gyrometer.

[0005] Such sensors are known in themselves, for example from Applicant's document EP 2 960 625 A1. Such a sensor includes means for excitation, detection, and often balancing. In the sensor, the masses are generally excited into vibration in a plane formed by the X and Y axes. This plane is perpendicular to a Z axis, which forms a so-called sensitive axis of the sensor. During rotation of the sensor about its sensitive axis, the vibration of the moving masses combined with an angular rotation vector generates, by the Coriolis effect, forces that set the moving masses into natural vibration perpendicular to the excitation vibration and the sensitive axis. The amplitude of this natural vibration is proportional to the rotational speed of the sensor and thus allows a value for the angular velocity about the sensitive axis to be deduced.

[0006] When a mass of such a sensor has an imbalance caused by manufacturing tolerances for example, the measurement is, in some cases, likely to be disturbed and is therefore not entirely correct.

[0007] Solutions have been proposed to address this problem, for example laser treatment to remove material.

[0008] Other solutions involve adding additional elements inside the sensor to allow for compensation of imbalance by mechanical means.

[0009] Other solutions propose direct control of the excitation means. However, such control is difficult to implement, as it depends on a large number of parameters, such as temperature, pressure, etc.

[0010] Li Cheng et al. in "A Digital Calibration Technique of MEMS Gyroscope for Closed-Loop Mode-Matching Control", Micromachines, vol.10, no.8, June 25, 2019, Chinwuba Ezekwe et al. in "A Mode-Matching Closed-Loop Vibratory Gyroscope Readout Interface With a 0.004 / s / VHz Noise Floor Over a 50 Hz Band", IEEE Journal of Solid-State Circuits, vol.43, no.12 December 1, 2019 and an IEEE article entitled "IEEE Recommended Practice for Inertial Sensor Test Equipment, Instrumentation, Data Acquisition and Analysis", January 1, 2005, propose solutions to address this problem.

[0011] These solutions can therefore still be improved, as they are difficult to implement and / or do not allow for a satisfactory increase in the reliability of the inertial angular sensor measurement.

[0012] Thus, an object of the present invention is to obtain a calibration method for a micro-machined angular inertial sensor which is simple to implement, while obtaining satisfactory measurement accuracy of the sensor.

[0013] To this end, the invention relates to a method for calibrating a micro-machined angular inertial sensor comprising a support, at least one vibrating mass movable relative to the support, at least one transducer for exciting a vibrational movement of the vibrating mass, at least one transducer for detecting a vibration of the vibrating mass, and at least one electrostatic transducer capable of applying an adjustable electrostatic stiffness to the vibrating mass, the calibration method comprising the steps of: reception by the angular sensor of a predetermined vibratory excitation emitted by an excitation device; measurement by the detection transducer of the vibration of the vibrating mass, to obtain a measurement signal from said measurement by the detection transducer; transformation of the measurement signal over a predetermined time window of said measurement signal into a power spectral density; determination of at least one value, called the noise value, as a function of a part of said power spectral density comprising a frequency less than or equal to a predetermined frequency; adjustment of the electrostatic stiffness by a feedback loop receiving as input the noise value and providing as output a control signal to the electrostatic transducer comprising the electrostatic stiffness to be applied to the vibrating mass minimizing the noise value, preferably minimizing the amplitude of the noise value;application of electrostatic stiffness by the electrostatic transducer according to the control signal.

[0014] The calibration process allows for very simple and reliable calibration of the inertial angular sensor, because the feedback loop adjustment reduces, or even eliminates, the sensor's sensitivity to vibrations, particularly linear and acoustic vibrations. This compensates for any imbalance in the sensor's mass(es) that causes such sensitivity.

[0015] Also, calibration according to the calibration process is particularly reliable since the noise value only presents low frequencies related to the presence of an imbalance in the mass(es) of the sensor when receiving the predetermined vibratory excitation.

[0016] According to other advantageous aspects of the invention, the calibration process comprises one or more of the following features, taken individually or in any technically possible combination: the excitation device is separate from the excitation transducer; the angular sensor comprises at least two vibrating masses movable relative to the support, and movable relative to each other, suspended by suspension springs from fixed anchor points of the support and coupled to each other by coupling springs to vibrate in opposite phase, the angular sensor comprising, respectively for each vibrating mass, at least one excitation transducer, and at least one detection transducer, the measurement step comprising the measurement of the vibration of each vibrating mass by the respective detection transducer, the measurement signal being obtained from the measurement of the vibration of each vibrating mass; the angular sensor comprises at least one electrostatic transducer respectively for each vibrating mass;and the adjustment step includes providing each electrostatic transducer with the control signal to apply the respective electrostatic stiffness to each vibrating mass, minimizing the noise value; the control signal includes, for each vibrating mass, a different electrostatic stiffness relative to each other to account for an imbalance of the vibrating masses; one of the vibrating masses is called the external mass, and the other of the vibrating masses is called the internal mass; and, during the adjustment step, the feedback loop provides the control signal comprising electrostatic stiffnesses generating a force determined as follows: ; Kti − Kte Xi + Xe 2 + Kti + Kte 2 Xi − Xe , Or : Kti is a sum of electrostatic stiffnesses to be applied to the internal mass, Kte is a sum of electrostatic stiffnesses to be applied to the external mass, Xi is a displacement of the internal mass, and Xe is a displacement of the external mass; said force cancels out the following disturbing forces: Mi − Me Xι . . + Xe . . 2 And Ki − Ke Xi + Xe 2 , Or Mi is the mass of the internal mass, Me is the mass of the external mass, Ẍ is the acceleration of the internal mass, Ẍe is the acceleration of the external mass, Ki is a sum of the stiffnesses of said suspension springs connecting the internal mass to the support, and Ke is a sum of the stiffnesses of said suspension springs connecting the external mass to the support; the measurement step includes the measurement by the vibration detection transducer(s) of the vibrating mass(s) along a first axis in a plane of the support, and the measurement, for the vibrating mass(s), by an additional vibration detection transducer of the vibrating mass(s) along a second axis perpendicular to the first axis and included in said plane; the control signal includes the electrostatic stiffness to be applied along the first axis by the electrostatic transducer(s), for the vibrating mass(s), the control signal further including the electrostatic stiffness to be applied along the second axis by an additional electrostatic transducer, for the vibrating mass(s);The application step includes the application of the respective electrostatic stiffness by the electrostatic transducer(s) along the first axis, and by the additional electrostatic transducer(s) along the second axis, in accordance with the control signal; the calibration process includes at least one repetition of the reception, measurement, transformation, determination, adjustment, and application steps; during the repetition, the control signal includes an electrostatic stiffness to be applied along a direction, in the plane of the support, distinct from the first and second axes; the control signal further includes an electrostatic stiffness to be applied by an electrostatic transducer compensating for a quadrature bias, the quadrature bias corresponding to a stiffness coupling of the suspension springs along the first and second axes;the excitation device includes at least one element from among a loudspeaker, a piezoelectric element and a vibrating pot; the noise value is equal to an arithmetic mean of a plurality of values ​​of the power spectral density at frequencies less than or equal to 0.1 Hz.

[0017] The invention also relates to a calibration system configured to calibrate a micro-machined inertial angular sensor comprising a support, at least one vibrating mass movable relative to the support, at least one transducer for exciting a vibrational movement of the vibrating mass, at least one transducer for detecting a vibration of the vibrating mass, and at least one electrostatic transducer capable of applying an adjustable electrostatic stiffness to the vibrating mass. the calibration system comprising an excitation device configured to emit a predetermined vibrational excitation to the angular inertial sensor, and the calibration system further comprising a calculation device comprising: a receiving module configured to receive a measurement signal obtained from a measurement of the vibration of the vibrating mass by the detection transducer; a transformation module configured to transform the measurement signal over a predetermined time window of said measurement signal into a power spectral density; a determination module configured to determine at least one value, called the noise value, as a function of a part of said power spectral density comprising a frequency less than or equal to a predetermined frequency;a feedback loop configured to adjust the electrostatic stiffness, the feedback loop being configured to receive the noise value as input and to provide as output a control signal to the electrostatic transducer including the electrostatic stiffness to be applied to the vibrating mass minimizing the noise value, preferably minimizing the amplitude of the noise value, and an emission module configured to emit the control signal to the inertial angular sensor to apply the electrostatic stiffness by the electrostatic transducer in accordance with the control signal.

[0018] These features and advantages of the invention will become apparent upon reading the following description, given solely by way of non-limiting example, and made with reference to the accompanying drawings, in which: [ Fig 1 ] there figure 1 is a schematic view of an assembly comprising an inertial angular sensor and a calibration system according to the invention; [ Fig 2 ] there figure 2 is a schematic view of an example of the angular inertial sensor of the figure 1 ; Fig 3 ] there figure 3 is a schematic view of another example of the angular inertial sensor of the figure 1 ; And [ Fig 4 ] there figure 4 is a flowchart of a calibration process implemented by the calibration system of the figure 1 .

[0019] On the figure 1 , an assembly 1 includes a micro-machined angular inertial sensor 2 and a calibration system 4 configured to calibrate the angular sensor 2. Hereafter, the micro-machined angular inertial sensor 2 is referred to as angular sensor 2.

[0020] Angular sensor 2 is, for example, a gyroscope configured to measure angular velocity. Alternatively, or in addition, angular sensor 2 is a gyroscope for measuring angular position.

[0021] Angular sensor 2 is a micro-machined sensor, and thus forms a microelectromechanical system, also defined by its acronym MEMS (from the English "Microelectromechanical systems").

[0022] Angular sensor 2 is in particular a sensor intended to be mounted in a vehicle, not shown, for example in an aircraft, a drone or a ship.

[0023] Angular sensor 2, for example, is intended to be integrated into a vehicle navigation, control or guidance system.

[0024] The angular sensor 2 is in particular a tuning fork type gyroscope, specifically a tuning fork type gyroscope with two vibrating masses.

[0025] By "vibrating mass", it is understood that the mass is capable of carrying out oscillations, for example driven by means described later, and by Coriolis effect during a rotation of the angular sensor 2.

[0026] With reference to the examples of figures 2 And 3 , the angular sensor 2 includes a support 6 extending in a plane of the support 6, along a first axis X and a second axis Y perpendicular to the first axis X.

[0027] The angular sensor 2 comprises at least one vibrating mass 8, 10, preferably two vibrating masses 8, 10 arranged one around the other, to form a mass called internal mass 8 and a mass called external mass 10.

[0028] The vibrating masses 8 and 10 are mobile relative to the support 6, and mobile relative to each other. The centers of gravity 0 of the vibrating masses 8 and 10 coincide at rest.

[0029] The angular sensor 2 includes suspension springs 12, for example four for each vibrating mass 8, 10, suspending each vibrating mass 8, 10 from a respective anchor point 14 which is fixed relative to the support 6.

[0030] The angular sensor 2 further includes coupling springs 16, for example four when the angular sensor 2 includes two masses 8, 10, coupling the vibrating masses 8, 10 together to allow vibration of the masses 8, 10 in opposite phase.

[0031] The example of angular sensor 2 of the figure 3 differs from the example of the figure 2 in that, on the figure 3 , the suspension springs 12 of the external mass 10 are oriented towards the internal mass 8 and fixed to the same anchorage points 14 as the corresponding suspension springs 12 of the internal mass 8.

[0032] The angular sensor 2 further includes, particularly visible on the figure 3 , at least one excitation transducer Ex, Ey of a vibrational movement of the vibrating mass 8, 10, at least one detection transducer Dx, Dy of a vibration of the vibrating mass 8, 10, and at least one electrostatic transducer Tx, Ty capable of applying an adjustable electrostatic stiffness to the vibrating mass 8, 10.

[0033] The fact that the static stiffness is "adjustable" indicates that the electrostatic transducer Tx, Ty is capable of modifying the electrostatic stiffness.

[0034] The angular sensor 2 preferably further comprises at least one electrostatic transducer for compensating a quadrature bias Q+, Q- configured to compensate for a quadrature bias. The quadrature bias corresponds to a coupling of the stiffness of the suspension springs 12 along the first X axis and the second Y axis.

[0035] For greater visibility, the Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- transducers are not shown on the figure 2 Preferably, the angular sensor 2 according to the example of the figure 2 includes the same transducers as those in the example of the figure 3 .

[0036] An example of the arrangement of the Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- transducers is described below, with reference to the figure 3 .

[0037] The angular sensor 2 includes for example an excitation transducer Ex configured to excite the internal mass 8 along the first X axis, and an excitation transducer Ey configured to excite the internal mass 8 along the second Y axis, in particular during an operation of the angular sensor 2, for example to obtain an angular velocity.

[0038] The angular sensor 2 includes, for example, two Dx sensing transducers configured to detect the vibration of the internal mass 8 along the first X axis, and two Dy sensing transducers for such detection along the second Y axis.

[0039] The angular sensor 2 further includes an electrostatic transducer Tx capable of applying the adjustable electrostatic stiffness to the internal mass 8 along the first X axis, and a corresponding electrostatic transducer Ty for the second Y axis, also called the additional electrostatic transducer.

[0040] Angular sensor 2 includes, for example, two Q+ compensation electrostatic transducers configured to compensate for a positive quadrature bias, and two Q- compensation electrostatic transducers configured to compensate for a negative quadrature bias.

[0041] On the figure 3 , the transducers Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- are represented only for the internal ground 8. Preferably, the angular sensor 2 further includes at least one, preferably each transducer Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- also for the external ground 10.

[0042] For example, each transducer Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- is as described in EP 2 960 625 A1.

[0043] Each transducer Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- is in particular a transducer comprising interdigitated combs. These interdigitated combs include a movable comb attached to the internal mass 8, in particular fixed relative to the internal mass 8, and a comb attached to the support 6, and in particular fixed relative to the support 6.

[0044] Each transducer Ex, Ey, Dx, Dy, Tx, Ty, Q+ and Q- is configured either to apply a force to the internal mass 8, according to a received voltage, or to detect a movement of the internal mass 8 by measuring variations in charge between the fixed and moving combs.

[0045] The calibration system 4 includes an excitation device 18 configured to emit a predetermined vibrational excitation to the angular sensor 2, and a calculation device 20.

[0046] The excitation device 18 includes, for example, a loudspeaker, a piezoelectric element and / or a vibrating pot.

[0047] The excitation device 18 is for example configured to emit to the angular sensor 2 a vibrational excitation of a frequency range centered around a resonance frequency of the excitation device 18, for example of a frequency range equal to + / -2kHz with respect to the resonance frequency.

[0048] The resonance frequency of the excitation device 18 is a predetermined frequency.

[0049] For example, the excitation device 18 is integrated into a calibration bench, not shown, allowing the angular sensor 2 to be positioned at a predetermined position for calibration by receiving the vibrational excitation.

[0050] In another example, the excitation device 18 is attached to the angular sensor 2, for example by being housed together with the angular sensor 2 in the same casing. This allows, for example, self-calibration of the angular sensor 2.

[0051] In some examples, the excitation device 18 is arranged outside or inside the angular sensor 2. In one example not shown, the excitation device 18 is integrated into the angular sensor 2.

[0052] The computing device 20 includes a receiving module 22, a transformation module 24, a determination module 25, a feedback loop 26 and a transmitting module 28.

[0053] The receiving module 22, the transformation module 24, the determination module 25, the feedback loop 26 and the transmitting module 28 are each, for example, integrated into at least one computer.

[0054] In this case, each of the elements among the receiving module 22, the transformation module 24, the determination module 25, the feedback loop 26 and the transmitting module 28 is at least partially in the form of software executable by a processor and stored in a memory of the computer.

[0055] Alternatively or in addition, each of the elements among the receiving module 22, the transformation module 24, the determination module 25, the feedback loop 26 and the transmitting module 28 is integrated, at least partially, into a physical device, such as for example a programmable logic circuit, such as an FPGA (Field Programmable Gate Array), or in the form of a dedicated integrated circuit, such as an ASIC (Application Specific Integrated Circuit).

[0056] The receiving module 22 is configured to receive a measurement signal S m obtained from a measurement of the vibration of the vibrating mass 8, 10 by each detection transducer Dx, Dy.

[0057] The transformation module 24 is configured to transform the measurement signal S m over a predetermined time window of this measurement signal S m into a power spectral density.

[0058] The determination module 25 is configured to determine at least one value, called the noise value V, as a function of a portion of the power spectral density comprising a frequency less than or equal to a predetermined frequency. This portion then has, in particular, a frequency less than or equal to the predetermined frequency.

[0059] The predetermined frequency is, for example, equal to 0.1 Hz.

[0060] According to other examples, the predetermined frequency has a value greater than or less than 0.1 Hz.

[0061] The feedback loop 26 is configured to adjust the electrostatic stiffness to be applied by each electrostatic transducer Tx, Ty to the vibrating mass 8, 10.

[0062] The feedback loop 26 is configured to receive the noise value V as input and to provide a control signal S c as output to the electrostatic transducer Tx, Ty including the electrostatic stiffness to be applied to the vibrating mass 8, 10 minimizing the noise value V. Preferably, the feedback loop 26 is configured to provide the control signal S c minimizing the amplitude of the noise value V as output.

[0063] The transmitting module 28 is configured to transmit the control signal S c to the angular sensor 2 to apply the electrostatic stiffness by each electrostatic transducer Tx, Ty in accordance with the control signal S c.

[0064] A calibration procedure for angular sensor 2 is now described, with reference to the figure 4 showing a flowchart of this process.

[0065] The calibration process 100 includes a receiving step 110, a measuring step 120, a transformation step 130, a determining step 135, an adjustment step 140 and an application step 150.

[0066] During the reception step 110, the angular sensor 2 receives the predetermined vibratory excitation emitted by the excitation device 18 which is separate from the excitation transducer Ex, Ey.

[0067] Each vibrating mass 8, 10 is subjected to different forces during the receiving stage 110. An example is described below.

[0068] Each vibrating mass 8, 10 is subjected to restoring forces by the suspension springs 12 connecting each vibrating mass 8, 10 to the support 6. The suspension springs 12 generate a sum of stiffnesses Ki applied to the internal mass 8 and a sum of stiffnesses Ke applied to the external mass 10.

[0069] The sum of the stiffnesses Ki applied to the internal mass 8 by the suspension springs 12 of the internal mass 8 is, for example, defined as follows: Ki = Kxi Kxyi Kxyi Kyi , Or Kxi is the stiffness along the first X axis generated by the suspension springs 12 connecting the internal mass 8 to the support 6; Kyi is the stiffness along the second Y axis generated by the suspension springs 12 connecting the internal mass 8 to the support 6; Kxyi represents the quadrature stiffness forming a coupling between the first X axis and the second Y axis.

[0070] The sum of the stiffnesses Ke applied to the external mass 10 is defined equivalently, with the index e instead of i: Ke = Kxe Kxye Kxye Kye .

[0071] We further define: Xi = xi yi as being the displacement of the internal mass 8, and Xe = xe ye like that of the external mass 10.

[0072] Each vibrating mass 8, 10 is also subjected to forces generated by the electrostatic transducers Tx, Ty. The electrostatic transducers Tx, Ty generate a sum of stiffnesses Kti applied to the internal mass 8 and a sum of stiffnesses Kte applied to the external mass 10.

[0073] The sum of the stiffnesses Kti applied to the internal mass 8 is, for example, defined as follows: Kti = Ktxi Ktxyi Ktxyi Ktyi , Or Ktxi is the stiffness along the first X axis generated by the electrostatic transducer Tx along the X axis; Ktyi is the stiffness along the first X axis generated by the electrostatic transducer Ty along the Y axis.

[0074] The sum of the stiffnesses Kte applied to the external mass 10 is defined equivalently with the index e instead of the index i.

[0075] The internal mass 8, for example, is subjected to an oscillating force F oscillant which is defined as follows: Mi . Xι . . + Ki . Xi = F oscillant , Or Mi is the mass of the internal mass 8; X is the acceleration of the internal mass 8.

[0076] Similarly, the external mass 10, for example, exhibits an oscillating force F oseillant which is defined as follows: Me . Xe . . + Ke . Xe = F oscillant , Or Me is the mass of the external mass 10; Ẍe is the acceleration of the external mass 10.

[0077] The person skilled in the art will understand that these definitions of oscillating forces are simplified definitions, not taking into account certain forces, for example coupling forces between the internal and external mass 8, 10, Coriolis forces, excitation force differences applied to each mass.

[0078] Starting from the first terms Mi. X̂ And Me. Ẍe equations of oscillating force F oseillant For each mass 8, 10, we have: Mi . X ¨ i − Me . X ¨ e = Mi − Me X ¨ i + X ¨ e 2 + Mi + Me 2 X ¨ i − X ¨ e .

[0079] This equation notably features phase acceleration X ¨ i + X ¨ e 2 , average acceleration of the two masses 8, 10, associated with the mass difference (Mi - Me) and an acceleration in opposite phase ( Ẍi - Ẍe ), associated with the average mass Mi + Me 2 ,

[0080] We now consider the second terms Ki. Xi And Ke. Xe equations of oscillating force F oscillant for each mass 8, 10. These terms relate to the forces generated by the suspension springs 12 connecting the internal mass 8 and external mass 10 to the support 6.

[0081] The movement Xi + Xe 2 described below is driven by the excitation device 18.

[0082] We have: Kxi Kxyi Kxyi Kyi xi yi − Kxe Kxye Kxye Kye xe ye = Ki . Xi − Ke . Xe = Ki − Ke Xi + Xe 2 + Ki + Ke 2 Xi − Xe

[0083] Either : Mi + Me 2 X ¨ i − X ¨ e + Ki + Ke 2 Xi − Xe = − Mi − Me X ¨ i + X ¨ e 2 − Ki − Ke Xi + Xe 2 .

[0084] During the reception step 110, the movement of the internal and external mass 8, 10 presents, in particular, two disturbing forces Mi − Me X ¨ ι + X ¨ e 2 And Ki − Ke Xi + Xe 2 , related to a difference in masses 8, 10 and a difference in the sums of the stiffnesses of the suspension springs 12. These disturbing forces are in particular reduced, preferably cancelled, by the feedback loop 26 during the adjustment step 140, described later.

[0085] During measurement step 120, the detection transducer or each Dx, Dy measures the vibration of the vibrating mass 8, 10, to obtain the measurement signal S m from the measurement by the detection transducer or each Dx, Dy.

[0086] The measurement signal S m is in particular a signal supplied at the output of the angular sensor 2, for example an angular velocity or an angular position.

[0087] For example, the angular sensor 2 determines, as a function of the amplitude of the vibration of the vibrating mass 8, 10, an angular velocity corresponding to the measurement signal S m.

[0088] For example, when the angular sensor 2 is a gyroscope, the measurement signal Sm is a measurement of the direction of vibration of the vibrating mass 8, 10, in the plane of the support 6, measured by the sensing transducers Dx, Dy. As an example, each sensing transducer Dx, Dy measures the amplitude of the vibration along the corresponding X and Y axes, and the angular sensor 2 obtains the measurement signal Sm from these measurements.

[0089] When the angular sensor 2 is a gyroscope, the measurement signal Sm is, for example, proportional to the excitation forces applied by the excitation transducer(s) Ex, Ey. The angular sensor 2 determines these excitation forces, in particular, based on the measurement of the vibration of the vibrating mass 8, 10 by the detection transducer(s) Dx, Dy.

[0090] For example, measurement step 120 includes the measurement by the detection transducer Dx of the vibration of the internal mass 8 along the first axis X, and the measurement by detection transducer Dy, also called additional detection transducer, of the vibration of the internal mass 8 along the second axis Y.

[0091] According to one example, measurement step 120 includes measurements in any direction in the plane of support 6, including different from the X or Y axis.

[0092] Specifically, this vibration corresponds to the displacement known as Xi above and possibly its first and / or second order derivatives.

[0093] Preferably, measurement step 120 further includes the measurement of the vibration of the external mass 10 along the first X axis and the second Y axis by the corresponding detection transducers, not shown.

[0094] Specifically, this vibration corresponds to the displacement known as Xe above, and possibly its first and / or second order derivatives.

[0095] In the case of two masses, namely the internal mass 8 and the external mass 10, the measurement step 120 includes in particular the measurement of a movement of each mass 8, 10 along the first axis X, and along the second axis Y. For example, the angular sensor 2 determines the measurement signal S m from the difference in movement, in particular in opposite phase, between the internal mass 8 and the external mass 10 along the first axis X, and the difference in movement between these masses 8, 10 along the second axis Y.

[0096] According to one example, the angular sensor 2 determines a resultant direction of the vibration of each vibrating mass 8, 10, in the plane of the support 6, from the measurements of the detection transducers Dx and Dy, to obtain the measurement signal S m.

[0097] During transformation step 130, the transformation module 24 transforms the measurement signal S m over a predetermined time window of this measurement signal S m into a power spectral density.

[0098] Power spectral density is in particular the power spectral density of the measurement signal S m.

[0099] The transformation module 24 obtains the power spectral density for example by applying the periodogram method to the measurement signal S m.

[0100] Power spectral density represents the frequency distribution of the power of the measured signal Sm according to the frequencies of that signal. For example, power spectral density has the unit: ° / s Hz , Or ° / s is the rotation speed obtained by measurement by the angular sensor 2 in degrees per second; and Hz is the frequency in Hertz.

[0101] During the determination step 135, the determination module 25 determines the noise value V, as a function of a part of the power spectral density comprising a frequency less than or equal to the predetermined frequency, for example 0.1 Hz.

[0102] This part then has, in particular, a frequency less than or equal to the predetermined frequency.

[0103] For example, the determination module 25 determines the arithmetic mean of each power spectral density value less than or equal to the predetermined frequency to obtain the noise value V.

[0104] According to another example, the determination module 25 determines an Allan variance to obtain the noise value V.

[0105] According to one example, the determination module 25 determines a plurality of noise values ​​V, corresponding for example to different directions of measurement in the plane of the support 6.

[0106] During the adjustment step 140, the feedback loop 26 adjusts the electrostatic stiffness to be applied to each mass 8, 10.

[0107] By "an electrostatic stiffness" is meant a stiffness which is constant for a predetermined time and which is applied to the vibrating mass 8, 10, in particular constant for a time several orders of magnitude greater than a period of oscillation of the predetermined vibratory excitation generated by the excitation device 18.

[0108] According to one example, the electrostatic stiffness depends on a direction in the plane of the support 6 formed by the X and Y axes and / or on a temperature of the angular sensor 2.

[0109] The feedback loop 26 receives the noise value V as input and provides the control signal S c as output to each electrostatic transducer Tx, Ty including the electrostatic stiffness to be applied to the respective vibrating mass 8, 10, minimizing the noise value V.

[0110] Preferably, the feedback loop 26 minimizes the amplitude of the noise value V by providing the control signal S c.

[0111] The feedback loop 26 notably implements negative feedback which tends to reduce the noise value V.

[0112] For example, the control signal S c includes a different electrostatic stiffness relative to each vibrating mass 8, 10 to take account of an imbalance of the vibrating masses 8, 10, preferably a different stiffness along the X axis relative to the Y axis.

[0113] In the example of an angular sensor 2 comprising the two vibrating masses 8, 10, the unbalance corresponds to the difference in mass between the vibrating masses 8 and 10.

[0114] The control signal S c includes for example the electrostatic stiffness to be applied along the X axis by the electrostatic transducer Tx, and the electrostatic stiffness to be applied along the Y axis by the additional electrostatic transducer Ty, for each vibrating mass 8, 10.

[0115] For example, the control signal S c further includes an electrostatic stiffness to be applied by each electrostatic quadrature bias compensation transducer Q+, Q-.

[0116] For example, the feedback loop 26 provides the control signal S c comprising electrostatic stiffnesses generating a force determined as follows: Kti − Kte Xi + Xe 2 + Kti + Kte 2 Xi − Xe .

[0117] In particular, this determined force minimizes, and preferably cancels, the following disturbing forces: Mi − Me X ¨ ι + X ¨ e 2 et Ki − Ke Xi + Xe 2 .

[0118] During application step 150, the electrostatic transducer Tx, Ty applies the electrostatic stiffness according to the control signal S c.

[0119] In particular, the electrostatic transducer Tx applies the respective electrostatic stiffness along the X axis, and the electrostatic transducer Ty along the Y axis, according to the control signal S c.

[0120] The electrostatic transducers Tx, Ty cancel out, in particular, the diagonal terms of the disturbing forces, by applying the following stiffnesses: Kti − Kte Xi + Xe 2 = Ktxi − Ktxe 0 0 Ktyi − Ktye Xi + Xe 2 .

[0121] For example, each Q+, Q- compensation electrostatic transducer applies the respective electrostatic stiffness according to the control signal S c.

[0122] Electrostatic Q+, Q- compensation transducers cancel out, in particular, the non-diagonal terms of the disturbing forces.

[0123] For example, the calibration process 100 is repeated several times, as illustrated by arrow R on the figure 4 .

[0124] For example, during the repetition of the calibration process 100, the excitation device 18 continuously emits a predetermined vibratory excitation to the angular sensor 2.

[0125] After a first execution of steps 110, 120, 130, 135, 140 and 150 and the application of the electrostatic stiffness according to the control signal S c, the noise value V, and in particular its amplitude, is, during a second execution, reduced compared to the first execution.

[0126] Preferably, the calibration procedure 100 is repeated until a predetermined value of the noise value V is obtained.

[0127] According to an example, the calibration process 100 includes at least one repetition of the steps reception 110, measurement 120, transformation 130, determination 135, adjustment 140, application 150.

[0128] For example, the control signal S c includes an electrostatic stiffness to be applied in a direction, in the plane of the support 6, which is distinct from the first X axis and the second Y axis.

[0129] For example, the calibration process 100 includes the implementation of several repetitions of the steps: reception 110, measurement 120, transformation 130, determination 135, adjustment 140, and application 150. In this case, during a given repetition, the control signal Sc includes an electrostatic stiffness to be applied in a first direction that is distinct from a second direction. The second direction corresponds, in particular, to the direction of the electrostatic stiffness to be applied during a previous repetition. The first and second directions lie within the plane of the support 6 and have, for example, a predetermined angle between them, such as 10°.

[0130] Of course, other embodiments than those described above can be considered.

[0131] For example, angular sensor 2 is a different type of sensor from the one described above. For example, angular sensor 2 comprises only one vibrating mass or more than two vibrating masses.

[0132] The calibration system 4 and the calibration process 100 according to the invention have a large number of advantages.

[0133] In particular, the calibration system 4 and the calibration process 100 allow calibration of the angular sensor 2 which is simple to implement thanks to the excitation by the excitation device 18 which is separate from the excitation transducer Ex, Ey, and thanks to the adjustment from the measurement at the output of the angular sensor 2.

[0134] Also, calibration by the method according to the invention makes it possible to obtain good measurement accuracy of the angular sensor 2, because the low frequency noise of the measurement of the angular sensor 2 is reduced by the feedback implemented by the feedback loop 26.

[0135] In the example where the sensor receives a predetermined vibrational excitation emitted by an excitation device that is separate from the excitation transducer, the implementation of the calibration process is particularly simple, as it is sufficient to position the angular sensor, for example, in a predefined area to receive this excitation for calibration, especially in the absence of adding compensation elements inside the sensor.

Claims

1. A method of calibrating (100) a micromachined inertial angle sensor (2) comprising a support (6), at least one vibrating mass (8, 10) movable relative to the support (6), at least one excitation transducer (Ex, Ey) for exciting a vibrating movement of the vibrating mass (8, 10), at least one detection transducer (Dx, Dy) for detecting a vibration of the vibrating mass (8, 10), and at least one electrostatic transducer (Tx, Ty) capable of applying an adjustable electrostatic stiffness to the vibrating mass (8, 10), the calibration method (100) comprising the steps of: - receiving (110), by the angle sensor (2), a predetermined vibrational excitation emitted from an excitation device (18); - measuring (120), by the detection transducer (Dx, Dy), the vibration of the vibrating mass (8, 10), to obtain a measurement signal (Sm) based on the measurement by the detection transducer (Dx, Dy); the method being characterized in that it comprises the steps of: - converting (130) the measurement signal (Sm) over a predetermined time window of said measurement signal (Sm) into a power spectral density; - determining (135) at least one value, called noise value (V), as a function of a portion of said power spectral density comprising a frequency less than or equal to a predetermined frequency; - adjusting (140) the electrostatic stiffness, by a feedback loop (26) receiving as input the noise value (V) and providing as output a control signal (Sc) to the electrostatic transducer (Tx, Ty) comprising the electrostatic stiffness to be applied to the vibrating mass (8, 10) that minimises the noise value (V), preferably that minimises the amplitude of the noise value (V); - applying (150), by the electrostatic transducer (Tx, Ty), the electrostatic stiffness according to the control signal (Sc).

2. The calibration method according to claim 1, wherein the excitation device (18) is separate from the excitation transducer (Ex, Ey).

3. The calibration method according to claim 1 or claim 2, wherein the angle sensor (2) comprises at least two vibrating masses (8, 10) movable relative to the support (6), and movable relative to each other, suspended by suspension springs (12) from fixed anchoring points (14) of the support (6) and coupled together by coupling springs (16) to vibrate in phase opposition, the angle sensor (2) comprising, respectively for each vibrating mass (8, 10), at least one excitation transducer (Ex, Ey) and at least one detection transducer (Dx, Dy); wherein the measuring step (120) comprises measuring the vibration of each vibrating mass (8, 10) by the respective detection transducer (Dx, Dy), the measurement signal (Sm) being obtained from the measurement of the vibration of each vibrating mass (8, 10).

4. The calibration method according to claim 3, wherein the angle sensor (2) comprises at least one electrostatic transducer (Tx, Ty) for each vibrating mass (8, 10) respectively; and wherein the adjusting step (140) comprises supplying, to each electrostatic transducer (Tx, Ty), the control signal (Sc) for applying the respective electrostatic stiffness to each vibrating mass (8, 10) that minimises the noise value (V).

5. The calibration method according to claim 4, wherein the control signal (Sc) comprises, for each vibrating mass (8, 10), a different electrostatic stiffness with respect to each other to take into account an imbalance of the vibrating masses (8, 10).

6. The calibration method according to any one of claims 3 to 5, wherein one of the vibrating masses (8, 10) is referred to as the outer mass (10), and the other of the vibrating masses is referred to as the inner mass (8); and wherein, in the adjusting step (140), the feedback loop (26) provides the control signal (Sc) comprising electrostatic stiffnesses generating a force determined as follows: Kti − Kte Xi + Xe 2 + Kti + Kte 2 Xi − Xe , where: Kti is a sum of electrostatic stiffnesses to be applied to the inner mass (8), Kte is a sum of electrostatic stiffnesses to be applied to the outer mass (10), Xi is a displacement of the inner mass (8), and Xe is a displacement of the outer mass (10).

7. The calibration method according to claim 6, wherein said force cancels out the following so-called disturbing forces: Mi − Me X ¨ ι + X ¨ e 2 and Ki − Ke Xi + Xe 2 , where Mi is the mass of the inner mass (8), Me is the mass of the outer mass (10), Ẍ is the acceleration of the inner mass (8), Ẍe is the acceleration of the outer mass (10), Ki is a sum of stiffnesses of said suspension springs (12) connecting the inner mass (8) to the support (6), and Ke is a sum of stiffnesses of said suspension springs (12) connecting the outer mass (10) to the support (6).

8. The calibration method according to any of the preceding claims, wherein the measuring step (120) comprises the or each detection transducer (Dx) measuring the vibration of the or each vibrating mass (8, 10) along a first axis (X) in a plane of the support (6) and an additional detection transducer (Dy) measuring, for the or each vibrating mass (8, 10), the vibration of the or each vibrating mass (8, 10) along a second axis (Y) perpendicular to the first axis (X) and comprised within said plane.

9. The calibration method according to claim 8, wherein the control signal (Sc) comprises the electrostatic stiffness to be applied along the first axis (X) by the or each electrostatic transducer (Tx), for the or each vibrating mass (8, 10), the control signal (Sc) further comprising the electrostatic stiffness to be applied along the second axis (Y) by an additional electrostatic transducer (Ty), for the or each vibrating mass (8, 10); and wherein the applying step (150) comprises applying the respective electrostatic stiffness by the or each electrostatic transducer (Tx) along the first axis (X), and by the or each additional electrostatic transducer (Ty) along the second axis (Y), in accordance with the control signal (Sc).

10. The calibration method according to claim 9, comprising at least one repetition of the steps of receiving (110), measuring (120), transforming (130), determining (135), adjusting (140), applying (150), wherein, during the repetition, the control signal (Sc) comprises an electrostatic stiffness to be applied in a direction, in the plane of the support (6), distinct from the first axis (X) and the second axis (Y).

11. The calibration method according to any one of claims 8 to 10 taken in combination with claim 3, wherein the control signal (Sc) further comprises an electrostatic stiffness to be applied by an electrostatic transducer for compensation of a quadrature bias (Q+, Q-), the quadrature bias corresponding to a stiffness coupling of the suspension springs (12) along the first axis (X) and the second axis (Y).

12. The calibration method according to any of the preceding claims, wherein the excitation device (18) comprises at least one of a loudspeaker, a piezoelectric element, and a vibration generator.

13. The calibration method according to any of the preceding claims, wherein the noise value (V) is equal to an arithmetic mean of a plurality of power spectral density values at frequencies less than or equal to 0.1 Hz.

14. A calibration system (4) configured to calibrate a micromachined inertial angle sensor (2) comprising a support (6), at least one vibrating mass (8, 10) movable relative to the support (6), at least one excitation transducer (Ex, Ey) for exciting a vibrating movement of the vibrating mass (8, 10), at least one detection transducer (Dx, Dy) for detecting a vibration of the vibrating mass (8, 10), and at least one electrostatic transducer (Tx, Ty) capable of applying an adjustable electrostatic stiffness to the vibrating mass (8, 10), the calibration system (4) comprising an excitation device (18) configured to emit a predetermined vibrational excitation to the inertial angle sensor (2), and the calibration system (4) further comprising a computing device (20) comprising: - a reception module (22) configured to receive a measurement signal (Sm) obtained from a measurement of the vibration of the vibrating mass (8, 10) by the detection transducer (Dx, Dy); the system being characterized in that the computing device comprises: - a transformation module (24) configured to transform the measurement signal (Sm) over a predetermined time window of said measurement signal (Sm) into a power spectral density; - a determination module (25) configured to determine at least one value, called noise value (V), as a function of a portion of said power spectral density comprising a frequency less than or equal to a predetermined frequency; - a feedback loop (26) configured to adjust the electrostatic stiffness, the feedback loop (26) being configured to receive as input the noise value (V) and provide as output a control signal (Sc) to the electrostatic transducer (Tx, Ty) comprising the electrostatic stiffness to be applied to the vibrating mass (8, 10) that minimises the noise value (V), preferably minimising the amplitude of the noise value (V), and - a transmission module (28) is configured to transmit the control signal (Sc) to the angle sensor (2) so as to apply the electrostatic stiffness by the electrostatic transducer (Tx, Ty) in accordance with the control signal (Sc).