METHOD FOR MEASURING AN ANGULAR ROTATIONAL VELOCITY AND / OR AN ANGULAR POSITION

The method enhances gyroscopic sensor performance by using four control loops and mode switching to minimize computational demands and errors, achieving accurate angular speed and position measurements across various rotational ranges.

FR3140161B1Active Publication Date: 2024-11-08JXSENS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
FR2022009684
Authority / Receiving Office
FR · FR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-09-23
Publication Date
2024-11-08
Estimated Expiration
2042-09-23

AI Technical Summary

Technical Problem

Existing gyroscopic sensors face challenges in achieving high performance measurement of angular speed and position due to complex algorithms requiring high computing power and being susceptible to errors, particularly when measuring large inertial rotations.

Method used

A method involving a vibrating gyroscopic sensor with four control loops and data processing to minimize differences between detection signals and setpoint amplitudes using regulators, allowing for efficient measurement of angular speed and position by constraining vibrations in orthogonal modal bases, and switching between FTR and WA modes based on threshold comparisons.

Benefits of technology

The method reduces computational load, improves measurement accuracy, and extends the rotational measurement range without complex calculations, enabling precise angular speed and position estimation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000029_0000
    Figure 00000029_0000
  • Figure 00000030_0000
    Figure 00000030_0000
  • Figure 00000031_0000
    Figure 00000031_0000
Patent Text Reader

Abstract

Method for measuring an angular velocity of rotation and / or an angular position comprising the steps of: (a)Receiving first and second vibration detection signals according to primary and secondary resonance modes of a resonator (3); (b)Implementation of at least four control loops using first, second, third and fourth regulators (R1, R2, R3, R4) respectively; (c)Estimation of said angular velocity and / or said angular position, as a function of the outputs of the regulators;characterized in that the first regulator (R1) seeks to minimize the difference between the phase component of the first detection signal and the product, by a first coefficient C1 function of the azimuthal angle θ in the orthogonal modal basis of the primary and secondary modes, of a setpoint amplitude of the vibration of the resonator (3), and the third regulator (R3) seeks to minimize the difference between the phase component of the second detection signal and the product, by a second coefficient C2 function of said angle θ, of said setpoint amplitude of the vibration of the resonator (3).Gyroscopic sensorsFigure for the abstract: Fig. 1.;
Need to check novelty before this filing date? Find Prior Art

Description

Title of the invention: METHOD FOR MEASURING AN ANGULAR SPEED OF ROTATION AND / OR AN ANGULAR POSITION

[0001] The present invention relates to the general technical field of rotation sensors, and more precisely to that of gyroscopic sensors based on Coriolis forces for measuring rotation speeds and / or angular positions, such as Coriolis effect vibrating gyrometers or gyroscopes known as CVG (Coriolis Vibratory Gyroscope).

[0002] The present invention relates more particularly to a method for measuring the angular speed and / or the angular position, along an axis of rotation by a vibrating axisymmetric gyroscopic sensor.

[0003] Vibrating structure gyrometers, intended to measure angular speeds, are well known. These vibrating gyrometers are based on the Coriolis effect which causes a vibrating object (the resonator) to undergo a force, when the latter rotates, to continue to vibrate in a single plane called inertial. The application of an opposing force makes it possible to rotate the vibration plane with the object. The vibration is then stationary relative to the rotating reference frame linked to the object and the measurement of this force makes it possible to determine the angular speed.

[0004] Axisymmetric resonators, i.e. resonators having a symmetry of revolution around an axis, also have the property that in the absence of the application of a force opposed to the Coriolis forces, the measurement of the position of the reference frame linked to the rotating object with respect to the fixed vibration directly gives the information on the angular position of the object. This is referred to as operation in gyroscope mode, as opposed to the previous mode of operation, called gyrometer mode.

[0005] Such axisymmetric CVGs generally use a hemispherical, cylindrical, annular or polygonal resonator with 8xk sides of identical lengths equi-angularly distributed with k a non-zero integer. In this case, the resonator has two resonance modes of identical frequencies called primary and secondary modes having the property of being modally orthogonal, which results in an angle of 45 degrees in the physical space of the resonator.

[0006] These axisymmetric CVGs incorporate advanced electronics allowing vibration control, i.e. the primary vibration mode, as well as the secondary vibration mode, or a combination of these modes. These electronics also implement algorithms aimed at identifying and eliminating any drift in order to obtain the greatest possible measurement accuracy.

[0007] The control algorithms, also called implementation algorithms, are generally based on principles presented in the IEEE 1431-2004 guide, entitled “IEEE Standard Specification Format Guide and Test Procedure for Coriolis Vibratory Gyros”. These algorithms typically implement four vibration control loops and are called FTR (Force To Rebalance) for a gyrometer implementation, and WA (Whole Angle) for a gyroscope implementation.

[0008] Concerning the algorithms for correcting sensor errors, generally associated with bias and scale factor, the most rudimentary ones proceed outside control loops (or downstream of the loops) by the simple application of compensation polynomials depending on the temperature, and the most complex ones, using the properties of CVGs with symmetrical vibration modes described above, intervene in the control loops. These complex algorithms generally require rotating the vibration plane while no inertial rotation is applied to the resonator. This is called electric rotation. To achieve this, the gyroscope mode is implemented, with the addition of an electric precession force electrically pushing the vibration in a controlled manner while maintaining the vibration amplitude constant.Error compensation then involves implementing correction matrices on the detection and excitation signals, followed by other correction matrices targeting the physical errors of the sensor itself. However, and particularly when excellent performance is sought, algorithms based on the implementation of these rotation matrices require high computing power, and remain perfectible.

[0009] The present invention improves the situation.

[0010] The objects assigned to the invention are achieved using a method for measuring an angular speed of a rotation and / or an angular position, along a sensitive axis of a vibrating gyroscopic sensor, comprising the implementation, by data processing means of the vibrating gyroscopic sensor, of steps of: (a) Receiving, from an electromechanical coupling instrumentation of a resonator of the vibrating gyroscopic sensor, a first signal for detecting a vibration of said resonator according to a primary resonance mode, and a second signal for detecting a vibration of said resonator according to a secondary resonance mode; each of the first and second detection signals comprising an in-phase component and a quadrature-phase component; said resonator being vibrated via said electromechanical coupling instrumentation according to a first excitation signal of the primary resonance mode and a second excitation signal of the secondary resonance mode; (b) Implementation of at least four control loops including: O a first control loop of the amplitude of the vibration of the resonator according to the axis of the primary resonance mode, using a first regulator; O a second phase-locked loop generating in-phase and quadrature-phase reference harmonic signals of the primary mode and / or secondary mode, making it possible to separate said in-phase and quadrature-phase components from the detection signals, using a second regulator; O a third loop for compensating the Coriolis forces due to said rotation of the resonator exciting a resonance mode perpendicular to the vibration, using a third regulator; O a fourth quadrature control loop of said resonance mode perpendicular to the vibration of the resonator using a fourth regulator; (c) Estimation of said angular velocity and / or said angular position, based on the outputs of the regulators; characterized in that the vibration is constrained by the first and second excitation signals so as to present an azimuthal angle in the orthogonal modal base of the primary and secondary modes, the first regulator seeking to minimize the difference between the in-phase component of the first detection signal and the product, by a first coefficient depending on the azimuthal angle, of a setpoint amplitude of the vibration of the resonator, and the third regulator seeking to minimize the difference between the in-phase component of the second detection signal and the product, by a second coefficient depending on the azimuthal angle, of said setpoint amplitude of the vibration of the resonator, the first loop also being a Coriolis force compensation loop and the third loop also being an amplitude control loop of the vibration of the resonator along the axis of the secondary resonance mode.

[0011] According to advantageous and non-limiting characteristics:

[0012] The method comprises a prior step (i) of selecting an operating mode from at least one mode called Force-To-Rebalance any attitude, FTR TT, in which said azimuthal angle 0 has a manually selected value, while a fifth control loop is not implemented, and a mode called Whole Angle, WA, in which the fifth control loop is implemented.

[0013] Said selection of an operating mode is carried out according to the result of the comparison between the estimated angular speed and at least one threshold.

[0014] The objects assigned to the invention are also achieved using a vibrating gyroscopic sensor for measuring an angular speed or a rotation along a sensitive axis, comprising: - a resonator having electromechanical coupling instrumentation for setting said resonator into vibration according to a first excitation signal from a primary resonance mode and a second excitation signal from a secondary resonance mode resonance; - data processing means configured to: O receive, from the electromechanical coupling instrumentation, a first signal for detecting a vibration of said resonator according to the primary resonance mode, and a second signal for detecting a vibration of said resonator according to the secondary resonance mode; each of the first and second detection signals comprising an in-phase component and a quadrature-phase component; O implement at least four control loops including: ■ a first loop for controlling the amplitude of the vibration of the resonator along the axis of the primary resonance mode, using a first regulator; ■ a second phase-locked loop generating in-phase and quadrature-phase reference harmonic signals of the primary mode and / or secondary mode, making it possible to separate said in-phase and quadrature-phase components from the detection signals, using a second regulator; ■ a third loop for compensating the Coriolis forces due to said rotation of the resonator exciting a resonance mode perpendicular to the vibration, using a third regulator; ■ a fourth loop for controlling the quadrature of said resonance mode perpendicular to the vibration of the resonator using a fourth regulator; O estimate said angular velocity and / or angular position based on the outputs of the regulators; characterized in that the vibration is constrained by the first and second excitation signals so as to present an azimuthal angle in the orthogonal modal base of the primary and secondary modes, the first regulator seeking to minimize the difference between the in-phase component of the first detection signal and the product, by a first coefficient depending on the azimuthal angle, of a setpoint amplitude of the vibration of the resonator, and the third regulator seeking to minimize the difference between the in-phase component of the second detection signal and the product, by a second coefficient depending on the azimuthal angle, of said setpoint amplitude of the vibration of the resonator, the first loop also being a Coriolis force compensation loop and the third loop also being an amplitude control loop of the vibration of the resonator along the axis of the secondary resonance mode.

[0015] The objects assigned to the invention are further achieved using a computer program product or program of a programmable logic circuit comprising code instructions for the execution of a method according to the invention, for measuring the angular speed of a rotation, or angular position, along a sensitive axis of an axisymmetric vibrating gyroscopic sensor, and allowing from these measures to deduce corrections for sensor errors applicable to vibration control loops to improve performance when said program is executed on a computer or programmable logic circuit within the sensor's implementation electronics.

[0016] The objects assigned to the invention are finally achieved using a storage means readable by computer equipment or a programmable logic circuit on which is recorded a computer program product or a program product of a programmable logic circuit comprising code instructions and fixed parameters stored for the execution of a method according to the invention for measuring the angular speed of a rotation, or for measuring the angular position, along a sensitive axis of an axisymmetric vibrating gyroscopic sensor, and making it possible, from these measurements, to deduce corrections to the errors of the sensor stored on a storage means readable by computer equipment or a programmable logic circuit applicable to the loops to improve the performances when said program is executed on a computer or a programmable logic circuit within the electronics implementing the sensor.

[0017] Other features and advantages of the invention will appear in more detail on reading the detailed description which follows, with reference to the appended drawings, given as purely illustrative and non-limiting examples, in which:

[0018] [Fig-1] illustrates, in a schematic perspective view, a gyroscopic sensor axisymmetric vibrating.

[0019] [Fig.2] illustrates, in a schematic view from above, a resonator of the vibrating gyroscopic sensor.

[0020] [Fig.3] represents a flowchart of a known FTR mode.

[0021] [Fig.4] illustrates how the commands generated by the control loops vary when the azimuthal electrical angle 0 is non-zero.

[0022] [Fig.5] represents a flowchart of an FTR mode, called FTR all attitude, in accordance with a preferred embodiment of the method according to the invention.

[0023] [Fig.6] represents a flowchart of an FTR mode in accordance with another preferred embodiment of the method according to the invention, this FTR mode being called FTR any continuous attitude, FTR TT.

[0024] [Fig.7] represents a logic diagram of a fifth control loop for a WA gyroscope mode in accordance with a preferred embodiment of the method according to the invention.

[0025] [Fig.8] illustrates the steps of a preferred embodiment of the method according to the invention.

[0026] Axisymmetric vibrating gyroscopic sensor

[0027] The invention relates to a method for measuring angular velocity and / or angular position, along an axis (called sensitive axis) by a vibrating gyroscopic sensor 1 preferably axisymmetric with respect to this sensitive axis, represented as an example in [Fig.l], which advantageously forms a vibrating gyroscopic sensor with Coriolis effect, that is to say a rotation sensor based on Coriolis forces. The vibrating gyroscopic sensor 1 is also called a vibrating inertial sensor of the CVG type (Coriolis Vibratory Gyro). We generally speak of a gyroscope when we seek to directly measure an angular position and of a gyrometer when we just seek to measure the angular speed, it being understood that said gyrometer can also indirectly determine an angle by integrating the angular speed measurements.However, in the case of axisymmetric CVGs, this integration can be done quite naturally, or directly, by the vibrating element itself (the resonator 3 described below), when the control electronics preferably implements a gyroscope operating mode, also called WA (Whole Angle), as explained below.

[0028] As illustrated in [Fig. 1], the vibrating gyroscopic sensor 1 generally comprises a base 2, which is for example metallic, or made of ceramic material, and provides a support function. The base 2 thus advantageously constitutes a base. The base 2 advantageously generally has the shape of a plate with two opposite faces, separated by the thickness of the plate, which are provided with convex and concave conformations forming housings whose functions will be specified in the following. The base 2 is advantageously provided with fixing means which may be for example in the form of a plurality of ears each crossed by an orifice in order to allow fixing by screwing the sensor 1 to a frame, for example to the frame of an inertial unit or any other equipment used for the navigation, guidance or control of a mobile object in space. As illustrated by [Fig.l], the fixing means 20, 21, 22, 23 of the base 2 can be designed to ensure a symmetrical fixing with respect to the main axis of symmetry of the resonator Z-Z' by being distributed equiangularly around this central axis which is advantageously normal to the mean plane in which the fixing lugs and / or the plate forming the base 2 extend. Preferably, the base 2 has in this case a discrete rotational symmetry of the fourth order along the central axis Z-Z'. Said central axis ZZ' corresponds to the sensitive axis along which the vibrating gyroscopic sensor 1 is designed to measure an angular speed or an angular rotation. Other cases are possible, such as for example a base 2 designed to ensure an isostatic fixing with three fixings distributed equiangularly around the central axis 7-71. Preferably, the base 2 in this case has a discrete rotational symmetry of the third order along the central axis Z-Z'.

[0029] Advantageously, the vibrating gyroscopic sensor 1 according to the invention comprises a bell-shaped cover 10, which covers the vibrating structure (the resonator 3 described below) of the sensor 1. Said cover 10 is made for example of a metallic material, and is advantageously intended to cooperate with the base 2 to delimit with the latter an internal space intended to accommodate the vibrating structure (the resonator 3 described below) of the vibrating gyroscopic sensor 1, in order to isolate it and protect it from the external environment. The housing thus formed by the base 2 and the cover 10 can advantageously be gas-tight, thanks to the implementation of appropriate sealing means arranged at the interface between the cover 10 and the base 2, which makes it possible to control the atmosphere prevailing within the housing in question.

[0030] The vibrating gyroscopic sensor 1 comprises, as explained, a resonator 3 which forms a test body intended to vibrate in response to an excitation. The resonator 3 includes, as illustrated in the figures, a central foot 30 by which the resonator 3 is attached to said base 2. In other words, the vibrating gyroscopic sensor 1 comprises mechanical connection means which ensure a fixing of the central foot 30 to the base 2. Advantageously, said mechanical connection means ensure a built-in connection between the central foot 30 and the base 2, so as to immobilize the central foot 30 relative to the base 2. The mechanical connection means have, for example, a definitive character, that is to say that they cannot be dismantled, and may consist, for example, of an assembly by gluing, welding or brazing of the foot 30 and the base 2.

[0031] The resonator 3 has a distal portion of shape in particular hemispherical, cylindrical, annular or polygonal with 8xk sides of identical lengths equi-distributed angularly with k a non-zero integer.

[0032] Conventionally, the vibrating gyroscopic sensor 1 comprises an electromechanical coupling instrumentation 31, attached to a proximal part of the resonator 3 (around the foot 30) itself comprising: - a plurality of excitation devices, called “forcers”, for exciting said resonator 3 into vibration, and in particular said symmetrical second-order vibration modes, primary and secondary, of elliptical deformation, of said distal portion of the resonator 3. - a plurality of detection devices, called “pickoffs”, for detecting vibrations of the resonator 3, and in particular the vibrations of said distal portion, excited by the excitation devices.

[0033] This positioning of the excitation and detection devices at the proximal portion makes it possible in particular to limit any damping effects caused by the electromechanical coupling instrumentation 31 and likely to generate measurement errors.

[0034] Said excitation devices and said detection devices may be of different natures, electromagnetic, electrostatic, optical, or for example piezoelectric elements. These are generally referred to as electrodes. The elements 31 in question are thus designed to, on the one hand, impart, via the proximal portion to which they are fixed, vibrations to the distal portion of the resonator 3, to excite in particular the symmetrical vibration resonance modes of order two, primary and secondary, of elliptical deformation, and on the other hand to detect, again via the proximal portion to which they are fixed, the vibrations of the distal portion. The elements 31 thus ensure a dual function, of vibration excitation on the one hand and of vibration detection on the other hand. The nature of the excitation elements 31 may be different from the nature of the detection elements. They participate fully in the control of the vibration.

[0035] Regardless of the number of elements 31 used for implementing and controlling the primary and secondary resonance modes of the resonator, it is known that for vibrating axisymmetric gyroscopic sensors, four signals are sufficient to carry out this control. Two signals are excitation signals, denoted E1 and E2 in [Fig. 3], and two signals are detection signals, denoted D1 and D2 in [Fig. 3]. The distribution of these signals from or to the resonator 3 from the implementation electronics can be carried out in different ways, and these embodiments define the number of electromechanical coupling elements 31 used. Preferably, the number of electromechanical coupling elements 31 is an integer multiple of eight as illustrated in [Fig. 1] and [Fig. 2].But this number can more simply become a multiple of two, or equal to two, when lower performance is acceptable, in particular when multiplexing of the detection and excitation signals is acceptable.

[0036] With reference to [Fig. 2], the elements 31 (denoted 31X, 31X', 31Y and 31Y', see below) are preferably distributed regularly on the proximal circumference of the resonator 3. For example, if there are 8 elements as shown, they are spaced 45° apart, i.e. placed at 0° and 180° for the pair [31X], 45° and 225° for the pair [31Y], 90° and 270° for the pair [31X'], and finally 135° and 315° for the pair [31Y'].

[0037] The vibrating gyroscopic sensor 1 further comprises data processing means 11 (i.e. the signal) connected to the electromechanical coupling elements 31 via interface and signal conditioning electronics, in particular digital electronics known as DGB, Digital Board, a proximity electronic card PB, and an interface card IB. The DGB means may include a processor, an FPGA (Field-Programmable Gate Array), a microcontroller, an ASIC (Application-Specific Integrated Circuit), etc., as well as the associated memories and clocks.

[0038] Conventionally, the PB interface card (or proximity card, or Proximity Board) is a card transforming the detection signals from the elements 31 of high impedance electromechanical coupling DX and DY attached to the resonator into low impedance signals which can thus travel to the DGB electronics without disturbance, and transferring the excitation signals EX and EY to the excitation elements 31 attached to the resonator. The interface board IB (Interface Board) is a board receiving the detection signals from PB and adapting and / or filtering them before analog-to-digital conversion by analog to digital converters (Analog to Digital Converter, ADC) at the input of the DGB (we denote DI and D2 the adapted / filtered versions of DX and DY), and adapting and / or filtering the excitation signals (denoted El and E2) received from the digital to analog converters (Digital to Analog Converter, DAC) at the output of DGB before transmitting them to the PB electronics and then to the resonator (then denoted EX and EY as explained).For the impedance adaptation of the detection signals, we can also have operational amplifiers (AGP) on the detection signals DX, DY in PB.

[0039] FTR Mode

[0040] We know an operating mode of a vibrating gyroscopic sensor 1, called FTR (Force To Rebalance) mode, in which the vibration is constrained in the reference direction defined by the elements 31 of the resonator 3. The control system associated with this mode is represented in [Fig.3]. The vibration is then aligned with the primary mode. As explained, the primary and secondary modes have the property of being modally orthogonal so that they can form an orthogonal modal base defining a modal plane (two-dimensional modal space) of the system. By denoting 0 the azimuthal angle defining the axis of the vibration in this orthogonal modal base of the primary and secondary modes (i.e. with respect to the abscissa axis of the modal plane), we have here 0=0°.

[0041] With reference to Figures 2 and 3, in a configuration with eight electromechanical coupling elements, in the physical domain, the orientation of the major axis of the vibration for the primary mode is aligned with four first elements, including a pair 3IX at 180° from each other, and another pair 3IX' also at 180° from each other, the first pair being spaced from the second by 90° in physical space, and the two pairs being positioned along the directions denoted X and X' in [Fig.2]. The secondary mode is deduced from the primary mode by a rotation of 45° in physical space; this corresponding to a rotation of 90° in the modal plane (i.e. 0=90°). This mode is also aligned with four other electromechanical coupling elements, a 31 Y pair, and a 31Y' pair, the elements of one pair being located 180° from each other, the two pairs being separated by 90°, and each of the pairs being aligned along the directions denoted Y and Y'.The electromechanical coupling instrumentation 31 receives the first and second excitation signals EX and EY (inputs, respectively via the elements 3IX and 31 Y, and produces the first and second detection signals DX and . DY (outputs or outputs, respectively via the elements 3IX' and 31 Y') which can be interpreted by the signal and data processing means 11 after normalization and filtering in the proximity card PB and / or the interface card IB.

[0042] In [Fig.2], the shape taken by resonator 3 when the vibration is constrained so as to be aligned with the primary mode, i.e. 0=0° in the modal plane (for convenience we will speak of vibration constrained in the “0=0° direction”), i.e. the main axis of the vibration coincides with the reference direction XX of resonator 3 in physical space, is shown in dotted lines.

[0043] When the vibrating gyroscopic sensor 1 rotates around its sensitive axis, the resulting Coriolis forces acting on the resonator excite the secondary mode.

[0044] A closed loop brings the secondary resonance mode back to zero, and the force required to cancel this mode is proportional to the angular velocity (denoted Q), perpendicular to the plane of [Fig.2], or else, collinear with the sensitive axis of the resonator 3, hence the name “force to rebalance”.

[0045] It may be noted here that the principle described above would also apply to the case where each pair would be replaced by a single electromechanical coupling element. In this case, only four electromechanical coupling elements would be used. Also, when the digital implementation elements operate at a sufficiently high frequency and well above the frequency of the primary sustained mode, the electromechanical coupling elements could be used in time sharing, alternately in excitation mode then in detection mode. In this case, theoretically, two electromechanical coupling elements would be sufficient, instead of four, or eight, and one would be for example 31X and the other 31Y.

[0046] With reference to [Fig. 3], there are thus at least 4 control loops in FTR mode, each comprising a regulator. In each loop, the input detection signals are demodulated so as to separate an in-phase component and a quadrature-phase component. They are denoted respectively Dlp [in-phase component of the first detection signal, i.e., in-phase component of the vibration according to the first mode], Dlq [quadrature-phase component of the first detection signal, i.e., quadrature-phase component of the vibration according to the primary mode], D2p [in-phase component of the second detection signal, i.e., in-phase component of the vibration according to the secondary mode], and D2q [quadrature-phase component of the second detection signal, i.e., quadrature-phase component of the vibration according to the secondary mode].Symmetrically, the outputs of the regulators are inversely remodulated to obtain the excitation signals El and E2, or EX and EY after filtering.

[0047] The first loop aims at amplitude control (setpoint amplitude noted A) on the in-phase component of the primary mode deduced from the first detection data DX. It implements a regulator RI whose output is the command noted CA (amplitude control) determining the first excitation signal EX.

[0048] The second loop is a phase-locked loop (PLL) with a regulator R2 locking the system to the primary mode frequency and generating, from the phase quadrature component of the primary mode deduced from the first detection data DX, reference harmonic signals SIN and COS, respectively in phase and in phase quadrature with the primary mode. These reference signals are used in all the control loops for the modulation of the detection signals and the modulation of the control forces applied to the resonator.

[0049] The third loop is the one that puts into practice the compensation of the Coriolis force on the in-phase component of the secondary mode deduced from the second detection data DY (this control aims to cancel this phase of the secondary mode) with a regulator R3 which produces at output the command noted CP (precession control). This command is proportional to the angular speed Q.

[0050] The fourth loop aims at quadrature control on the phase quadrature component of the secondary mode deduced from the second detection data DY (this control aims at cancelling this phase quadrature of the secondary mode) so that the oscillation of the resonator 3 remains a standing wave which is represented as a straight line in the mode space, that is to say in an orthogonal reference frame having the primary mode as its abscissa axis, and the secondary mode as its ordinate axis. It implements a regulator R4 whose output is the command denoted CQ (control of the quadrature of the vibration).

[0051] The second excitation signal is determined by the sum of the CP and CQ commands.

[0052] In order to orient the vibration in a direction different from the azimuthal electrical direction 0=0°, i.e. with a non-zero angle in the modal base of the primary and secondary modes, it is known to use rotation matrices applied to both the detection signals and the excitation signals. These matrices are implemented in the digital processing, carried out in DGB for example, preferably after the ADCs for detection, and before the DACs for excitation, to minimize the computational load by avoiding mathematical multiplication operations involving harmonic signals at the frequency of the resonator vibration. The set points of the regulators remain constant.

[0053] The present method implements a new FTR mode called any attitude, designated FTR TT, which allows, compared to the state of the art described previously, a significant reduction in the computing load, and therefore in the consumption, in the size of the components and the overall cost of the electronics. The expression "any attitude" means that the azimuthal angle 0 can take any desired value between 0 and 90°, or even 360°, without using complex mathematical operations. To reformulate, the main axis of the vibration can be moved away from an azimuthal angle 0 with respect to the modal reference direction preferentially defined by the primary mode of the resonator 3, without complex calculations. A preferred embodiment of this FTR TT mode is illustrated by [Fig.5], which will be described below.

[0054] A limiting case is obtained when the azimuthal angle is 90°, i.e. 0=90°. The vibration (primary mode) is then aligned with the elements 31 Y, 31 Y' and the secondary mode is aligned with the elements 3IX, 3IX'.

[0055] This “à la carte” FTR mode proves to be very effective and in particular makes it possible to implement various techniques for identifying sensor errors, with ultimately the calibration of these errors according to a mode known as bias self-calibration.

[0056] Principle of FTR mode any attitude

[0057] With reference to [Fig.4] which represents in the modal plane the orientation of the forces generated by the elements 31, the idea is based on the observation that whatever the orientation of the vibration defined by the azimuthal electrical angle 0, CA is always in temporal phase with CP and geometrically orthogonal to CP, and CQ is always in temporal phase quadrature with CA, and geometrically collinear with CP.

[0058] We can then generate each force as the action of 2 vectors applied in the orthogonal directions of the primary and secondary modes, with for example for CA a vector aligned on the direction X-X' with an amplitude AxCl, and another vector aligned on the direction Y-Y' with an amplitude AxC2, where Cl and C2 are two coefficients which are functions of 6, called first coefficient and second coefficient, preferably such that Cl2+C22=l, which then very cleverly leads to controlling the amplitude of the vibration in the direction 0 such that Cl=cos(0) and C2=sin(0).

[0059] Therefore, while the first loop had the setpoint A and the third loop virtually the setpoint 0, these setpoints become respectively AxCl and AxC2. This leads the regulator RI to minimize AxCl-Dlp, and the regulator R3 to minimize AxC2-D2p. We understand that: - the first loop, like the third loop, plays a role in compensating for the Coriolis forces due to the rotation of resonator 3 exciting the resonance mode perpendicular to the vibration (the secondary mode if 0 = 0°, the primary mode if 0 = 90° and a combination of the two otherwise); - the third loop presents like the first loop a role of control in amplitude of the vibration, for the first loop according to the axis of the secondary mode of resonance (projection on the axis of the abscissas of the modal base by the multiplication

[0060]

[0061] by cos(0)), and for the third loop the axis of the secondary resonance mode (projection on the abscissa axis of the modal base by multiplication by sin(0)), see [Fig.3]. These set points do not require any specific processing and we can simply pre-calculate values ​​of Cl and C2 and store them in a table 12 (Look-up Table, or LUT of [Fig.5]) which receives as input an external setpoint to DGB noted CE defining a desired angle 0. This external setpoint can have a permanent character of programming of the LUT 12 (i.e. 0 is predefined) which then returns fixed Cl and C2 coefficients called "factory". This external setpoint can also control the LUT to obtain variable Cl and C2 coefficients, which would cause a precession of the vibration. This external setpoint can also itself directly provide the values ​​of Cl and C2, the LUT then only transferring these values ​​to the regulators.Preferably, this external instruction CE requires the LUT to return stored values ​​of coefficients Cl and C2 calculated as sine and cosine functions of an angle 0 which then becomes the azimuthal angle taken by the vibration. As the R3 regulator now plays an identical role to that of the RI regulator, in this new FTR mode any attitude, the RI regulator and the R3 regulator can be designed with the same parameters (i.e. gain, bandwidth, frequencies), and for Figure 5, the R3 regulator can then be identical to the RL regulator. Thus, to obtain an azimuthal angle 0=90°, the first loop takes the first coefficient Cl=0 (the first loop then has a pure role of compensation of the Coriolis forces) and the third loop takes the second coefficient C2=l, i.e. the setpoint AxC2=A (the third loop then has a pure role of amplitude control). The vibration, with respect to the forces generated by loops 1 and 3 then takes an orientation of 45° in the physical frame, i.e. oriented along the Y and Y' axes of Figure 2. Any intermediate value is possible, such as for example ^3 and X—• -1 ““ -x C2=l / 2 which would lead to an electrical angle of 30°.

[0062] Concerning the regulator R2 of the PLL, when for example Cl=0 and the azimuthal angle takes a value of 90°, taking into account that the initial secondary mode (0=0°), then virtually takes the role of the primary mode, it becomes necessary to establish the input signal of the regulator R2 from the quadrature signal D2q of the second detection channel. For this purpose, a switch is used upstream of R2 selecting either the signal Dlq, or the signal D2q, to ​​establish the input signal D12q of the regulator R2. The control of this switch is also a third coefficient C3 stored in the LUT which like Cl and C2, results either from a factory-fixed programming, or is defined by a DGB input signal denoted CE. Preferably, C3 makes it possible to obtain a maximum amplitude for the signal D12q. Preferably, C3 makes it possible to obtain a maximum amplitude for the signal D12q. Alternatively, C3 toggles the switch when the azimuth angle 0 takes the value 45°, with for azimuth angle values ​​between 0° and 45°, the equality D12q=Dlq, and for azimuth electrical angle values ​​between 45° and 90°, the equality D12q=D2q.

[0063] Concerning the quadrature control regulator R4, similarly to the PLL regulator, it is also necessary to establish its input signal by selecting either the Dlq signal or the D2q signal. Again a switch is used and the control of this switch is also a fourth coefficient C4 stored in the LUT which, like C1 and C2, results either from a factory-fixed programming or is defined by a variable DGB input signal denoted CE. This fourth coefficient C4 can be a coefficient different from C3. Preferably, this coefficient makes it possible to obtain a maximum amplitude for the D21q input signal of the regulator R4. Preferably, C4 toggles the switch when the azimuth angle 0 takes the value 45°, with for the azimuth angle values ​​between 0° and 45°, the equality D21q=D2q, and for the azimuth electrical angle values ​​between 45° and 90°, the equality D21q=Dlq.

[0064] The output OR4 of the quadrature regulator R4 remains conventionally summed to the output OR3 of the regulator R3 when the azimuth electrical angle 0 is equal to 0°. On the other hand, when the azimuth electrical angle 0 is equal to 90° and the secondary mode virtually becomes the primary mode, the output OR4 of the quadrature regulator R4 must then be summed to the output OR1 of the regulator RL. This is again achieved using a switch. The control of this switch is advantageously identical to the control of the input switch of R4 (fourth coefficient C4) and can therefore be a coefficient different from C3.Alternatively, this coefficient is identical to C3 with a switch flip for an azimuth electrical angle of 45° so that when the azimuth electrical angle takes values ​​between 0° and 45°, this switch sends OR4 to a summer whose other input signal is output OR3, and when the azimuth electrical angle takes values ​​between 45° and 90°, this switch sends OR4 to a summer whose other input signal is output OR1, this time with a change of sign to take into account the real orientation of the quadrature force in the modal frame.

[0065] To conclude, in the presence of a rotation along the measurement axis, this FTR mode in any attitude leads to a modulation of the outputs of the regulators RI and R3 by the angular velocity Q. In other words, the output of the regulators RI and R3 contains the information Q since a Coriolis force then appears in the direction opposite to the vector Cp on [Fig.4] with a non-zero projection on the axes of the primary and secondary modes, that is to say with a component proportional to -C2 in the direction of the primary mode, and a component proportional to +C1 in the direction of the secondary mode. secondary. To extract the angular velocity Q from the signals coming from these regulators, it is then necessary to multiply the output OR1 of the regulator RI by the coefficient -C2, to multiply the output OR3 of the regulator R3 by the coefficient Cl, and to calculate the difference OR3xCl-ORlxC2, the setpoints Cl and C2 being preferentially chosen such that Cl2+C22=l.

[0066] To summarize: - As a starter: O In the first loop, the RI regulator always takes Dlp as input, i.e. the in-phase component of the primary mode, but seeks to minimize AxCl-Dlp, with Cl a setpoint defined by a LUT 12 controlled or programmed by an external signal to DGB noted CE, rather than just A-Dlp, AxCl corresponding as explained to the projection of A on the axis of the primary vibration mode when Cl=cos(0). O In the second loop, the regulator R2 seeks to minimize D12q with D12q taking the output value of a switch driven by a setpoint C2 defined by a LUT 12 controlled or programmed by an external signal to DGB denoted CE. This switch selects either Dlq or D2q, depending on the value of the azimuthal electrical angle, i.e. the phase quadrature component of either the primary mode or the secondary mode, rather than just Dlq, O In the third loop, the regulator R3 still takes D2p as input, i.e. the in-phase component of the secondary mode, but seeks to minimize AxC2-D2p, with C2 a setpoint defined by a LUT 12 controlled or programmed by an external signal to DGB noted CE and preferably such that Cl2+C22=l, rather than just D2p, AxC2 corresponding as explained to the projection of A on the axis of the secondary mode of vibration when C2=sin(0), O In the fourth loop, the regulator R4 seeks to minimize D21q with D21q taking the output value of a switch driven by a setpoint C4 defined by a LUT 12 controlled or programmed by an external signal to DGB denoted CE. This switch selects either D2q or Dlq, depending on the value of the azimuthal electrical angle, i.e. the quadrature phase component of either the secondary mode or the primary mode, rather than just D2q. - On exit: O The outputs OR1 and OR3 of the regulators RI and R3 both contain part of the angular velocity information. To reconstruct this information, the difference OR3xCl- ORlxC2 must be calculated, O We always have at the output of the regulator R2 the reference harmonic signals SIN and COS, respectively the phase and the quadrature of the phase of the primary mode when the azimuthal angle is 0°, or of the secondary mode when the azimuthal angle is 90°, or of a combination of the primary mode and the secondary mode representing the state of the vibration when the azimuthal angle is any between 0° and 90°, O Whereas previously we always had the output OR4 of the regulator R4 summed to the output of the regulator R3, this output OR4, thanks to a switch controlled by the fourth coefficient C4 defined by a LUT controlled or programmed by an external signal to DGB noted CE, is either summed to OR3, or to OR1, depending on the value taken by the azimuthal angle.

[0067] Principle of FTR mode in any continuous attitude

[0068] The main disadvantage of the all-attitude FTR mode described above is the interruption of the continuous operation of the regulators R2 and R4 caused when the switches are toggled. Even if this interruption can be achieved in a very short time, less than one period of the harmonic signal corresponding to the sustained vibration of the resonator, a disturbance of the vibration control system can be created. Another aspect is also that of the variation in the amplitude of the input signals of the regulators R2 and R4 while the vibration takes different angular positions in the modal plane. This leads to a variation in performance.

[0069] To remedy this, a variant of the principles set out above is described by [Fig.6] where the three switches acting on the regulators R2 and R4 are replaced by calculation elements using other fifth, sixth, seventh and eighth coefficients C5, C6, C7, C8, which, like Cl and C2, the principles of which presented above remain unchanged, are preferably tabulated, i.e. stored in a LUT 12 which receives as input an external instruction to DGB denoted CE. This external instruction can have a permanent character of programming of the LUT which then returns fixed coefficients Cl, C2, C5, C6, C7 and C8 called "factory". This external instruction CE can lead the LUT to define variable coefficients stored in the LUT, or directly transferred from CE to the regulators, via the LUT for example.Preferably, this external instruction CE requires the LUT 12 to return stored values ​​of coefficients Cl, C2, C5, C6, C7 and C8 calculated as sine and cosine functions of an angle 6 which then becomes the azimuthal angle taken by the vibration. As a result, the FTR mode in any attitude becomes continuous.

[0070] For this new all-attitude FTR implementation architecture, R3 still plays a similar role to the RI regulator, and the RI regulator and the R3 regulator can be designed, preferably, with the same parameters (i.e., gain, bandwidth, frequencies). The RI regulator receives the AxCl setpoint and always works to minimize the AxCl-Dlp magnitude. The R3 regulator receives the AxC2 setpoint and minimizes the AxC2-D2p magnitude, these 2 RI and R3 regulators controlling both the vibration amplitude and canceling the effect of the Coriolis forces. When the condition Cl2+C22=l is met, the Cl and C2 coefficients can be represented as harmonic functions of the angle 0 which defines the angular position of the vibration in the modal plane.

[0071] Concerning the regulator R2 of the PLL, two new fifth and sixth coefficients C5 and C6 are used to act on the signal D12q that this regulator seeks to minimize. The coefficients C5 and C6 are stored in the LUT 12 and as for Cl and C2, they result either from a fixed “factory” programming, or are defined by a DGB input signal which assigns them variable values. The external signal controlling the LUT is noted CE. The regulator then minimizes the quantity D12q=DlqxC5+D2qxC6. Preferably, C52+C62=l. Preferably, C5=C1 and C6=C2.

[0072] The outputs of regulator 2 are unchanged and correspond to the reference harmonic signals SIN and COS, respectively the phase and the phase quadrature of the primary mode when the azimuthal angle is 0°, or of the secondary mode when the azimuthal angle is 90°, or of a combination of the primary mode and the secondary mode representing the state of the vibration when the azimuthal angle is any between 0° and 90°.

[0073] Concerning the quadrature control regulator R4, similarly to the PLL regulator, two new seventh and eighth coefficients C7 and C8 are used to act on the signal D21q that this regulator seeks to minimize. The coefficients C7 and C8 are stored in the LUT and as for Cl, C2, C5 and C6, they result either from a fixed “factory” programming, or are defined by an input signal CE of DGB which assigns them variable values. The regulator then minimizes the quantity D21q=DlqxC7+D2qxC8. Preferably, C72+C82=l. Preferably, C7=C2 and C8=C1.

[0074] The output OR4 of the regulator R4 is indirectly added to the outputs of the regulators RI and R3 with a weighting -OR4xC7 for the quantity added to the output of the regulator RI, and a weighting OR4xC8 for the quantity added to the output of the regulator R3. This principle makes it possible to restore the vector components of the quadrature force taking into account the angular position of the vibration represented by the azimuthal angle 6.

[0075] The estimation of the angular velocity in this continuous mode is identical to the case of any attitude using switches described previously. In other words, the output of the regulators RI and R3 contains the information Q with a non-zero projection on the axes of the primary and secondary modes, that is to say with a component proportional to -C2 in the direction of the primary mode, and a component proportional to +C1 in the direction of the secondary mode. To extract the angular velocity Q from the signals coming from these regulators, it is then necessary to multiply the output OR1 of the regulator RI by a setpoint -C2, to multiply the output OR3 of the regulator R3 by a setpoint Cl, and to calculate the difference OR3xCl-ORlxC2.

[0076] WA mode

[0077] One of the main known drawbacks of the FTR mode, whether standard, or any attitude, or any continuous attitude as described above, is that the rotation measurement range is limited. Indeed, the more the amplitude of the inertial rotation applied collinearly to the sensor measurement axis increases, the more the Coriolis force control electronics, including electromechanical coupling elements, must provide signals of high values ​​opposing the effect of the Coriolis forces, which in practice is limited by the sensor supply voltage.

[0078] To overcome this limitation, the full attitude FTR mode, or the continuous full attitude FTR mode described above, can be transformed into a gyroscope mode, called WA (Whole Angle) mode for which, theoretically, there is no limitation for the rotation measurement domain.

[0079] With reference to [Fig.7], a fifth control loop of said azimuthal angle 0 is implemented, using a fifth regulator R5 seeking to minimize the difference OR 13 of the product by C2 of the output OR1 of the first regulator RI and the product by Cl of the output OR3 of the third regulator R3, where OR 13 corresponds to the amplitude of the force CP to be provided to oppose the action of the Coriolis forces on the sustained vibration, and therefore proportional to the angular speed Q along the measurement axis of the vibrating gyroscopic sensor 1.

[0080] More precisely, the integral of the value of OR13 is proportional to the angular position of the vibrating gyroscopic sensor 1, and therefore the regulator R5 plays the role of an integrator of the angular velocity of the vibrating gyroscopic sensor 1 and must provide a command CI which causes the LUT 12 to send coefficients Cl and C2 such that the force CP is zero. An output of the regulator R5 is the physical rotation angle q> of the vibrating gyroscopic sensor 1, proportional to the azimuthal angle 0 (in practice half).

[0081] In this case, the angular position of the vibration is no longer controlled in relation to the resonator, which therefore begins to rotate around the vibration plane. This corresponds to the gyroscope operating mode.

[0082] The coefficients Cl and C2, and C5, C6, C7, C8 are preferentially, as explained, harmonic coefficients, with Cl=C5=C8=cos(0), C2=C6=C7=sin(0). The angle 0 is the azimuthal angle defining the angular position of the vibration maintained in the modal plane.

[0083] Note that the feedback branch of the fifth loop is not shown in [Fig.7]. This branch comes from the coefficients C1 and C2, as well as C5, C6, C7 and C8, distributed to the different regulators shown in [Fig.5] or [Fig.6].

[0084] This new WA mode has many advantages over the state of the art based on principles presented in IEEE Guide 1431-2004, entitled "IEEE Standard Specification Format Guide and Test Procedure for Coriolis Vibratory Gyros”: - It is derived directly from the FTR all attitude or FTR all attitude continuous mode, - The estimation of the physical angle of rotation of the vibrating gyroscopic sensor 1 results directly from the output of a regulator and therefore does not require complex trigonometric calculations, such as the arc tangent of the azimuthal angle 0, - It is possible to combine the FTR mode in any attitude and the WA mode by implementing a switch controlled by a ninth coefficient C9 represented in [Fig.7] such that, beyond a specified value of angular velocity Q, the WA mode is activated and the FTR mode is deactivated. Preferably, the coefficient C9 results from the output of a comparator (not represented in [Fig.7]) having as input OR13 and a fixed setpoint value programmed in the factory. Preferably, the coefficient C9 can result from a command external to the device 11.

[0085] Method

[0086] According to a first aspect, with reference to [Fig. 8], the present invention proposes a method for measuring the angular speed of a rotation, or of an angular position, along a sensitive axis of the vibrating gyroscopic sensor 1, which can begin with an optional step (i) of initial selection of an operating mode of the vibrating gyroscopic sensor 1, either FTR all attitude gyrometer (FTR TT), or FTR continuous TT, or Whole Angle gyroscope, WA, or a combination of these operating modes, FTR TT and WA.

[0087] When the FTR TT operating mode (continuous or non-continuous) is selected, the system moves to step (a) which comprises signal and data processing means 11, from the electromechanical coupling instrumentation 31 of the resonator 3, of the first detection signal DX of a vibration of said resonator 3 according to a primary resonance mode, and of the second detection signal DY of a vibration of said resonator 3 according to the secondary resonance mode, and generates using four control loops a first excitation signal EX of the primary resonance mode and a second excitation signal EY of the secondary resonance mode, the physical vibration taking an orientation not necessarily colinear to the primary and secondary modes.

[0088] Originally, during this step (a), the vibrating gyroscopic sensor 1 preferentially relies on data storage means 12 comprising a table of precalculated values ​​C1 to C9 or variables defined by an external command CE for use by the data processing means 11 making it possible to orient the vibration axis of the resonator 3 by an azimuthal angle 0 in the modal plane relative to a reference direction defined by the primary mode (abscissa axis of the plane modal). The data processing means implement at least four control loops including: O a first loop for controlling the amplitude of the vibration of the resonator 3 projected along the axis of the primary resonance mode and also compensating for the effect of Coriolis forces, using a first regulator RI, which seeks to minimize the difference between the component Dlp in phase of the first detection signal and the product AxCl of an amplitude setpoint A of the vibration of the resonator 3 and a first coefficient Cl. The first excitation signal EX is generated as a function of the outputs of the first and fourth regulator RI, R4. O a second phase-locked loop generating reference harmonic signals in phase and in phase quadrature with the sustained vibration, making it possible to separate said in-phase and quadrature components of the detection signals, using a second regulator R2 (more precisely, step (b) comprises producing said in-phase component and phase quadrature component from the first and second detection signals by demodulating the first and second detection signals as a function of the reference harmonic signals in phase and in phase quadrature with the sustained vibration). In particular, the second regulator R2 seeks to minimize the sum D12q between the product of the phase quadrature component Dlq of the first detection signal by a fifth coefficient C5, and the product of the phase quadrature component D2q of the second detection signal by a sixth coefficient C6, O a third loop for controlling the amplitude of the vibration of resonator 3 projected onto the axis of the secondary mode and compensating for the effect of Coriolis forces, using a third regulator R3, which seeks to minimize the difference between the component D2p in phase of the second detection signal and the product AxC2 of an amplitude setpoint A of the vibration of resonator 3 and a second coefficient C2 from table 12. The output of regulators RI and R3 contains the information Q. To extract the angular velocity Q from the signals from these regulators, it is then necessary to multiply the output OR1 of regulator RI by -C2, to multiply the output OR3 of regulator R3 by the first coefficient Cl, and to calculate the difference OR3xCl-ORlxC2, O a fourth loop for controlling the quadrature of the vibration of the resonator 3 according to either the secondary resonance mode, or the primary mode, or a combined resonance mode, using a fourth regulator R4, which seeks to minimize the difference D21q between the product of the quadrature component D2q of the second detection signal by an eighth coefficient C8, and the product of the quadrature component Dlq of the first detection signal by a seventh coefficient C7. Conventionally, the second excitation signal EY is generated as a function of the outputs of the third and fourth regulator R3, R4.

[0089] Preferably, the operating mode can automatically switch from FTR TT to WA and vice versa. In particular, the FTR TT operating mode switches to the WA operating mode if the estimated angular velocity is greater than a first threshold (denoted Threshold 1), and / or the WA operating mode switches to the FTR TT operating mode if the angular velocity is less than a second threshold (denoted Threshold 2), preferably close to or identical to the first threshold.

[0090] More specifically, in this preferred embodiment, when the measured angular velocity reaches and exceeds the predefined measurement threshold, Threshold 1, the output signal of the implementation electronics 11 of the vibrating gyroscopic sensor 1, and comprising different signal adaptation and processing functions, either saturates if the FTR TT mode was selected during step (i), or automatically switches to WA mode for which a fifth control loop is activated using a switch. In this case, the natural output of the system switches from an angular velocity measurement to a measurement of a rotation angle (angular position), or an angle increment.The knowledge of the angle increments clocked by a clock internal to the device 11, is a quantity proportional to the angular speed which when it falls below the second predefined measurement threshold, Threshold 2, causes a switch from the WA operating mode to the FTR TT mode (continuous or non-continuous). Threshold 2 can be triggered for an angular speed corresponding to the first threshold, Threshold 1.

[0091] If during the initial selection step the operating mode chosen is the WA mode, the 5 regulators are activated from the start, with the regulator R5 controlling 0 and therefore selecting the first and second coefficients C1 and C2 used by the Amplitude / Coriolis regulators R1 and R3, such that these regulators do not generate forces opposing the Coriolis forces. The vibration is then free, that is to say it is no longer driven by the rotating resonator, and the output of the regulator 5 is angular rotation information q> along the measurement axis of the vibrating gyroscopic sensor 1.

[0092] In a step (c), the data processing means 11 estimate said angular speed and / or said angular position, as a function of the outputs of the regulators. This step (c) is in practice implemented simultaneously with the other steps.

[0093] As represented by [Fig.7]: - the angular speed Q can be estimated from the outputs OR1 and OR3 of the first and third regulators RI, R3, in particular it is then necessary to multiply the output OR1 of the regulator RI by -C2, to multiply the output OR3 of the regulator R3 by the first coefficient Cl, and to calculate the difference OR3xCl-ORlxC2; and / or - if the fifth regulator is used, the angular position q> can be estimated at from the output OR5 of the fifth regulator R5; in practice it is appropriate to take half of the output OR5 of the fifth regulator R5 since this corresponds to the azimuthal angle 6.

[0094] Vibrating gyroscopic sensor

[0095] According to a second aspect, the invention proposes the vibrating gyroscopic sensor 1 for implementing the method according to the first aspect (i.e. for measuring an angular speed of a rotation, or an angular position, along a sensitive axis).

[0096] It typically includes: - a resonator 3 having electromechanical coupling instrumentation 31 for setting said resonator into vibration according to a first detection signal and a first excitation signal of a primary resonance mode, and a second detection signal and a second excitation signal of a secondary resonance mode; - data processing means 11 making it possible to define the orientation of the vibration in the mode space possibly from an external CE setpoint, and to establish information proportional either to the rotation speed projected onto the measurement axis of the vibrating gyroscopic sensor 1, or to the angular rotation of the sensor; - Possibly data storage means 12 (for storing in particular the table of values ​​of the coefficients C1 and C2, ie cos(0) and sin(0), as well as possibly C3, C4, C5, C6, C7, C8 and C9).

[0097] As explained, it is assumed that the main axis of vibration according to the primary resonance mode can be deviated by an azimuthal angle 0 relative to a reference direction in the mode space defined by the primary and secondary modes of the resonator 3, and thus the means 11 are configured to: O receive, from the electromechanical coupling instrumentation 31, a first signal for detecting a vibration of said resonator 3 according to the primary resonance mode, and a second signal for detecting a vibration of said resonator 3 according to the secondary resonance mode; each of the first and second detection signals comprising an in-phase component and a quadrature-phase component; O implement at least four control loops including: ■ a first loop for controlling the amplitude of the vibration of the resonator 3 projected onto the axis of the primary resonance mode, and also compensating for the Coriolis forces due to said rotation of the resonator exciting the resonance mode perpendicular to the vibration, using a first RI regulator, the first RI regulator seeking to minimize the difference between the in-phase component of the first detection signal and the product by cos(0) of a setpoint amplitude A of the vibration of the resonator 3; ■ a second phase-locked loop generating phase reference harmonic signals and the phase quadrature of the primary mode making it possible to separate said in-phase and phase quadrature components of the detection signals, using a second regulator R2; ■ a third loop for controlling the amplitude of the vibration of the resonator projected onto the axis of the secondary resonance mode, and also compensating for the Coriolis forces due to said rotation of the resonator 3 exciting the secondary resonance mode, using a third regulator R3, the third regulator R3 seeking to minimize the difference between the in-phase component of the second detection signal and the product by sin(0) of said setpoint amplitude A of the vibration of the resonator 3; ■ a fourth loop for controlling the quadrature of a resonance mode perpendicular to the vibration of the resonator, that is to say, either the primary mode, or the secondary mode, or a combined resonance mode, and using a fourth regulator R4; ■ a possible fifth loop acting as an integrator of the angular velocity of the vibrating gyroscopic sensor 1 and providing a command controlling the setpoints of the regulators RI, R3 such that no force opposes the Coriolis forces. An output of the regulator R5 is the physical rotation angle q> of the vibrating gyroscopic sensor 1; O estimate said angular speed, and / or said angular position, as a function of at least the outputs of the first regulator, the third regulator, and where appropriate the fifth regulator;

[0098] Computer program product or program product of a logic circuit pro grammable

[0099] According to a third and a fourth aspect, the invention relates to a computer program product or program of a programmable logic circuit comprising code instructions for the execution (in particular on the data processing means 11) of a method according to the first aspect of the invention for measuring the angular speed of a rotation, or an angular position, along a sensitive axis of a vibrating gyroscopic sensor 1, as well as storage means readable by computer equipment on which this computer program product or program of a programmable logic circuit is found.

Claims

1. Claims Method for measuring an angular speed of a rotation and / or an angular position, along a sensitive axis of a vibrating axisymmetric gyroscopic sensor (1), comprising the implementation, by data processing means (11) of the vibrating axisymmetric gyroscopic sensor (1), of steps of: (a) Receiving, from an electromechanical coupling instrumentation (31) of a resonator (3) of the vibrating axisymmetric gyroscopic sensor (1), a first signal for detecting a vibration of said resonator (3) according to a primary resonance mode, and a second signal for detecting a vibration of said resonator (3) according to a secondary resonance mode; each of the first and second detection signals comprising an in-phase component and a quadrature-phase component; said resonator (3) being vibrated via said electromechanical coupling instrumentation (31) according to a first excitation signal of the primary resonance mode and a second excitation signal of the secondary resonance mode; (b) Implementation of at least four control loops including: O a first control loop of the amplitude of the vibration of the resonator (3) along the axis of the primary resonance mode, using a first regulator (RI); O a second phase-locked loop generating in-phase and quadrature-phase reference harmonic signals of the primary mode and / or secondary mode, making it possible to separate said in-phase and quadrature-phase components from the detection signals, using a second regulator (R2); O a third loop for compensating the Coriolis forces due to said rotation of the resonator (3) exciting a resonance mode perpendicular to the vibration, using a third regulator (R3); O a fourth quadrature control loop of said resonance mode perpendicular to the vibration of the resonator using a fourth regulator (R4); (c) Estimation of said angular velocity and / or said angular position, based on the outputs of the regulators; characterized in that the vibration is constrained by the first and second excitation signals so as to present an azimuthal angle 0 in the orthogonal modal base of the primary and secondary modes, the first regulator (RI) seeking to minimize the difference between the in-phase component of the first detection signal and the product, by a first coefficient Cl depending on the azimuthal angle 0, of a setpoint amplitude of the vibration of the resonator (3), and the third regulator (R3) seeking to minimize the difference between the in-phase component of the second detection signal and the product, by a second coefficient C2 depending on the azimuthal angle 0, of said setpoint amplitude of the vibration of the resonator (3), the first loop also being a Coriolis force compensation loop and the third loop also being an amplitude control loop of the vibration of the resonator (3) along the axis of the secondary resonance mode.

2. Method according to the preceding claim characterized in that the first and second coefficients Cl and C2 verify Cl2+C22=l.

3. Method according to the preceding claim characterized in that the first and second coefficients Cl and C2 verify Cl=cos(0), and C2=sin(0).

4. Method according to any one of the preceding claims, characterized in that the first regulator (RI) and the third regulator (R3) are identical.

5. Method according to any one of the preceding claims, characterized in that the second regulator (R2) seeks to minimize the sum between the product, by a fifth coefficient C5 depending on the azimuthal angle 0, of the phase quadrature component of the first detection signal, and the product, by a sixth coefficient C6 depending on the azimuthal angle 0, of the phase quadrature component of the second detection signal.

6. Method according to the preceding claim characterized in that the fifth and sixth coefficients C5 and C6 verify C52+C62=1.

7. Method according to the preceding claim characterized in that the fifth and sixth coefficients C5 and C6 verify C5=C1, and C6=C2.

8. Method according to any one of the preceding claims, characterized in that the fourth regulator (R4) seeks to minimize the difference between the product, by a seventh coefficient C7 depending on the azimuthal angle 0, of the phase quadrature component of the first detection signal, and the product, by an eighth coefficient C8 depending on the azimuthal angle 0, of the phase quadrature component of the second detection signal.

9. Method according to the preceding claim characterized in that the seventh and eighth coefficients C7 and C8 verify C72+C82=1.

10. Method according to the preceding claim characterized in that the seventh and eighth coefficients C7 and C8 verify C7=C2, and C8=C1.

11. Method according to any one of the preceding claims, characterized in that said angular velocity is estimated in step (c) by calculating the difference OR 13 of the product by the second coefficient C2 of the output of the first regulator (RI) and of the product by the first coefficient Cl of the output of the third regulator (R3).

12. Method according to any one of the preceding claims characterized in that the vibrating axisymmetric gyroscopic sensor (1) comprises data storage means (12) storing a table of precalculated values ​​of the coefficients C1, C2, C3, C4, C5, C6, C7 and C8 used by the data processing means (11).

13. Method according to any one of the preceding claims, characterized in that the electromechanical coupling instrumentation (31) consists of piezoelectric elements, in particular eight, arranged regularly around the resonator (3).

14. Method according to any one of the preceding claims, characterized in that step (b) comprises the implementation of a fifth control loop of said azimuthal angle 0 providing a command controlling the setpoints of the regulators (RI, R3) of the first and third loops, using a fifth regulator (R5).

15. Method according to claim 14, characterized in that said angular position is estimated in step (c) as a function of the output of the fifth regulator (R5).

16. Method according to any one of claims 14 and 15, characterized in that the fifth regulator (R5) seeks to minimize the difference OR 13 of the product by the second coefficient C2 of the output OR1 of the first regulator (RI) and of the product by the first coefficient Cl of the output OR3 of the third regulator (R3).

17. Method according to the preceding claim, characterized in that it comprises a prior step (i) of selecting an operating mode from at least one mode called Force-To-Rebalance any attitude, FTR TT, in which said azimuthal angle 0 has a manually selected value, while the fifth control loop is not implemented, and a mode called Whole Angle, WA, in which the fifth control loop is implemented.

18. Method according to the preceding claim characterized in that said selection of an operating mode is carried out as a function of the result of the comparison between the estimated angular speed and at least one threshold.

19. Method according to the preceding claim characterized in that: - The FTR TT operating mode switches to the WA operating mode if the estimated angular speed is greater than a first threshold; - The WA operating mode switches to the FTR TT operating mode if the angular speed is less than a second threshold, preferably close to or identical to the first threshold.

20. Vibrating axisymmetric gyroscopic sensor (1) for measuring an angular velocity or a rotation along a sensitive axis, comprising: - a resonator (3) having electromechanical coupling instrumentation (31) for vibrating said resonator according to a first excitation signal from a primary resonance mode and a second excitation signal from a secondary resonance mode; - data processing means (11) configured to: O receive, from the electromechanical coupling instrumentation (31), a first signal for detecting a vibration of said resonator (3) according to the primary resonance mode, and a second signal for detecting a vibration of said resonator (3) according to the secondary resonance mode; each of the first and second detection signals comprising an in-phase component and a quadrature-phase component;O implement at least four control loops including: ■ a first loop for controlling the amplitude of the vibration of the resonator (3) along the axis of the primary resonance mode, using a first regulator (RI); ■ a second phase-locked loop generating reference harmonic signals in phase and in phase quadrature of the primary mode and / or of the secondary mode, making it possible to separate said in-phase and phase quadrature components from the detection signals, using a second regulator (R2); ■ a third loop for compensating the Coriolis forces due to said rotation of the resonator (3) exciting a resonance mode perpendicular to the vibration, using a third regulator (R3); ■ a fourth loop for controlling the quadrature of said resonance mode perpendicular to the vibration of the resonator using a fourth regulator (R4); O estimate said angular velocity and / or angular position based on the outputs of the regulators; characterized in that the vibration is constrained by the first and second excitation signals so as to present an azimuthal angle 0 in the orthogonal modal base of the primary and secondary modes, the first regulator (RI) seeking to minimize the difference between the in-phase component of the first detection signal and the product, by a first coefficient Cl depending on the azimuthal angle 0, of a setpoint amplitude of the vibration of the resonator (3), and the third regulator (R3) seeking to minimize the difference between the in-phase component of the second detection signal and the product, by a second coefficient C2 depending on the azimuthal angle 0, of said setpoint amplitude of the vibration of the resonator (3), the first loop also being a Coriolis force compensation loop and the third loop also being an amplitude control loop of the vibration of the resonator (3) along the axis of the secondary resonance mode.

21. Computer program product comprising code instructions for executing a method according to one of claims 1 to 18 for measuring the angular velocity and / or an angular position along a sensitive axis of a vibrating axisymmetric gyroscopic sensor (1), when said program is executed on a computer.

22. Storage means readable by computer equipment on which is recorded a computer program product comprising code instructions for the execution of a method according to one of claims 1 to 14 for measuring the angular speed and / or an angular position along a sensitive axis of a vibrating axisymmetric gyroscopic sensor (1).