• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a navigational aid method for an inertial navigation unit comprising at least one inertial sensor (4) having a sensitive axis (XX), each inertial sensor (4) comprising an ASG gyroscope (8) capable of delivering an ASG signal representative of a rotation around the corresponding sensitive axis (XX), and a MEMS gyroscope (10) capable of delivering a MEMS signal representative of a rotation about the corresponding sensitive axis (XX), the method comprising the steps of: - between a first date and a third subsequent date, calculating a trajectory from the MEMS signals; from the third date, calculating the trajectory from the ASG signals; estimating a bias vector introduced by the MEMS gyroscopes (10), from the MEMS signals and the ASG signals; - at a fourth date after the third date, recalibration of the trajectory.
    公开号:FR3060114A1
    申请号:FR1662387
    申请日:2016-12-13
    公开日:2018-06-15
    发明作者:Sophie Morales；Augustin Palacios Laloy；Jean-Michel Leger；Georges Remillieux；Marc Gramliche；Etienne Brunstein
    申请人:Commissariat a lEnergie Atomique CEA；Safran SA；Commissariat a lEnergie Atomique et aux Energies Alternatives CEA；
    IPC主号:
    专利说明:

    Holder (s): COMMISSIONER FOR ATOMIC ENERGY AND ALTERNATIVE ENERGIES Public establishment, SAFRAN Société anonyme.
    Extension request (s)
    Agent (s): BREVALEX Limited liability company.
    NAVIGATION AID PROCESS, COMPUTER PROGRAM PRODUCT AND CONTROL UNIT
    ASSOCIATED INERTIAL NAVIGATION.
    FR 3 060 114 - A1 f5j> The invention relates to a navigation aid method for an inertial navigation unit comprising at least one inertial sensor (4) having a sensitive axis (Χ-Χ), each inertial sensor (4) comprising an ASG gyroscope (8) capable of delivering an ASG signal representative of a rotation about the corresponding sensitive axis (Χ-Χ), and a MEMS gyroscope (10) capable of delivering a MEMS signal representative of a rotation around the corresponding sensitive axis (Χ-Χ), the method comprising the steps of:
    - between a first date and a third later date, calculation of a trajectory from MEMS signals;
    - from the third date, calculation of the trajectory from ASG signals;
    - estimation of a bias vector introduced by MEMS gyroscopes (10), from MEMS signals and ASG signals;
    - on a fourth date later than the third date, readjustment of the trajectory.
    NAVIGATION AID METHOD, COMPUTER PROGRAM PRODUCT AND INERTIAL NAVIGATION CONTROL UNIT THEREFOR
    DESCRIPTION
    TECHNICAL AREA
    The present invention relates to a navigation aid method for an inertial navigation central stationary with respect to a solid.
    The invention also relates to a computer program product and an inertial navigation unit.
    The invention applies to the field of inertial navigation by gyroscopes, in particular by atomic spin gyroscopes (or ASG, from the English “atomic spin gyroscopes”), such as nuclear magnetic resonance gyroscopes, also called “NMR gyroscopes ", And the co-magnetometers.
    PRIOR STATE OF THE ART
    The use of ASG gyroscopes as inertial sensors for performing rotation measurements is known. Such ASG gyroscopes generally have a low angular drift noise ARW (from the English “Angulor Rondom Walk”) and a low drift. In addition, such gyroscopes are capable of being miniaturized (volume of the order of ten cubic centimeters) and are capable of being produced at low cost.
    ASG gyroscopes therefore represent an interesting alternative for the design of Inertial Navigation Power Plants (CIN) which are reliable, small and inexpensive, in particular for autonomous navigation applications without GPS (from the English "Global Positioning System") .
    However, such ASG gyroscopes are not entirely satisfactory.
    Indeed, the start-up time of such ASG gyroscopes, that is to say the duration, from their power-up, at the end of which such gyroscopes are in an operational operating phase, is likely to be too long for some autonomous navigation applications without GPS.
    The start time of an ASG gyroscope has two main limits:
    the first is a technical limit linked to the heating and stabilization of the control systems at start-up; the second, of a physical order, depends on an alkaline gas / noble gas couple present in the ASG gyroscope, and corresponds to the minimum time necessary to polarize the noble gas by spin exchange with the alkaline.
    Thus, the start-up time of an ASG gyroscope is likely to reach one or more minutes.
    However, the implementation of an inertial navigation unit requires the use of operational and efficient inertial sensors a maximum of a few seconds after the inertial navigation unit is powered up, so as to limit the duration of the phase d 'initialization of the inertial navigation unit, also called "alignment phase", before switching to a so-called "navigation" mode during which the inertial unit is operational and provides users with position, speed and attitude information.
    The alignment phase is broken down, for example, into the following steps:
    - a start-up of the inertial navigation unit (a few seconds to a few tens of seconds);
    - an initialization of the position and the speed (a few tenths of a second); and
    - an orientation of the navigation marker (a few minutes).
    The durations of the different stages are given above as an indication in the case of an alignment phase of gyrocompass type used for applications of air transport type for which the alignment phase therefore has a duration of a few minutes.
    It is understood that the use of gyroscopes of the ASG type in an inertial navigation unit results in an increase in the duration of the start-up step, therefore in the duration of the alignment phase of the inertial unit, due to the fact that ASG gyroscopes start up time is one or more minutes.
    This increase in the duration of the initialization phase of the inertial navigation unit is not desirable. Indeed, it is generally desirable that this initialization time be as short as possible, in particular for a navigation application without GPS.
    An object of the invention is therefore to propose an inertial navigation unit using an ASG gyroscope which is reliable, small in size and inexpensive while allowing rapid start-up.
    STATEMENT OF THE INVENTION
    To this end, the subject of the invention is a method of aid to navigation of the aforementioned type, the inertial unit comprising at least one inertial sensor having a sensitive axis, each inertial sensor comprising an ASG gyroscope and a MEMS gyroscope integral with it. one from the other, the ASG gyroscope being capable of delivering an ASG signal representative of a rotation around the corresponding sensitive axis, the MEMS gyroscope being suitable of delivering a MEMS signal representative of a rotation around the sensitive axis corresponding, the method comprising the steps:
    - calculation, between a first date and a third later date, of a trajectory and, for each inertial sensor, of a corresponding biased trajectory, from MEMS signals, assuming, for the biased trajectory, that the inertial sensor has a predetermined unit bias;
    - calculation, from the third date, of the trajectory and of each biased trajectory from the ASG signals, assuming, for the biased trajectory, that the inertial sensor has a predetermined unit bias;
    - estimation of a bias vector introduced by MEMS gyroscopes, from MEMS signals and ASG signals;
    - recalibration, on a fourth date subsequent to the third date, of the trajectory as a function of each biased trajectory, of the unit biases and of the estimated bias vector, to obtain a nominal trajectory which is not tainted by means of MEMS gyroscopes .
    Indeed, MEMS gyroscopes have a short start-up time, which makes the inertial unit quickly operational. The fusion between the signals collected from MEMS gyroscopes and ASG gyroscopes helps to compensate for errors linked to biases introduced by MEMS gyroscopes.
    The nominal trajectory thus obtained is no longer tainted by means of MEMS gyroscopes.
    In addition, unlike other types of instruments, MEMS and ASG gyroscopes are capable of providing continuous measurement, which makes it possible to use them in such an inertial unit.
    The navigation aid method which is the subject of the invention is therefore reliable and allows rapid start-up.
    According to other advantageous aspects of the invention, the inertial navigation unit comprises one or more of the following characteristics, taken in isolation or in any technically possible combination:
    - for each MEMS gyrometer, the corresponding component of the bias vector is equal to the average, between a second date and the fourth date, of the difference between an angular speed originating from the corresponding MEMS signal and an angular speed originating from the corresponding ASG signal, the second date being between the first date and the third date;
    - the nominal trajectory is obtained by subtracting an offset from the trajectory, the offset being a vector correction term calculated according to:
    dXn dXn dXïi ÔXn = b ° x 5DÔÏ + b ° y dDÔy + b ° z dDto where ÔXn is the shift;
    boi is the i-th component of the bias vector; and the size is calculated according to:
    3D0i dXn _XnDi (t rec ) -Xn (t rec ) dD0i _ DOi where XnDi (trec) is the i-th biased trajectory taken on the fourth date; and DOi is the predetermined unit bias associated with the component i;
    the method comprises a recovery step, the recovery step comprising:
    - between a second date and the third date, the second date being between the first date and the third date, a first phase for calculating the trajectory and each biased trajectory from the MEMS signal;
    - on the third date, a switch to calculate the trajectory and each biased trajectory from a corresponding angle increment, the angle increment being obtained, for each sensitive axis, by the relation:
    dôcom = 0ASG (t com) 0MEMs (tcom - Te) -Δθ where dùcom is the angle increment;
    0ASG (t C om) is a quantity equal to the cumulative increments of angle of rotation around the sensitive axis between the second date and the third date, calculated from the ASG signal at switching;
    0MEMs (tcom-T e ) is a quantity equal to the cumulative increments of angle of rotation around the sensitive axis between the second date and a duration T e before the third date, calculated from the MEMS signal;
    each increment being equal to the integral, between two successive instants, of the angular speed of rotation around a sensitive axis originating from the corresponding MEMS or ASG signal,
    ΔΘ is a predetermined angular correction term; and
    T e is a predetermined duration;
    - between the third date and the fourth date, a second phase to calculate the trajectory and each biased trajectory from the ASG signal;
    - the angular correction is equal to the average, between the second date and the third date, of the values taken over time by the quantity (0asg - 0mems), where 0asg is a quantity equal, at a given instant, to the cumulative since the second date until the given instant, increments of angle of rotation around the sensitive axis obtained from the ASG signal, for the inertial sensor 4 considered, and
    0mems is a quantity equal, at a given instant, to the accumulation, from the second date until said given instant, of the increments of angle of rotation around the sensitive axis, obtained from the MEMS signal;
    - the method includes, from the fourth date, the calculation of the nominal trajectory only from ASG signals.
    Furthermore, the subject of the invention is a computer program product comprising program code instructions which, when executed by a computer, implement the method as defined above.
    In addition, the subject of the invention is an inertial navigation unit, fixed relative to a solid, the inertial unit comprising at least one inertial sensor having a sensitive axis, each inertial sensor comprising an ASG gyroscope and a MEMS gyroscope integral with it. one from the other, the ASG gyroscope being capable of delivering an ASG signal representative of a rotation around the corresponding sensitive axis, the MEMS gyroscope being suitable of delivering a MEMS signal representative of a rotation around the sensitive axis corresponding, the inertial unit further comprising a computer configured to implement the navigation aid method as defined above.
    BRIEF DESCRIPTION OF THE DRAWINGS
    The invention will be better understood with the aid of the description which follows, given solely by way of nonlimiting example and made with reference to the appended drawings in which:
    - Figure 1 is a schematic representation of an inertial unit according to the invention;
    - Figure 2 is a schematic representation of an inertial sensor of the inertial unit of Figure 1; and
    FIG. 3 is a graph representing the evolution over time of an angular speed measured by the inertial unit of FIG. 1.
    DETAILED PRESENTATION OF PARTICULAR EMBODIMENTS
    In the following, the vector quantities will be noted in bold.
    An inertial navigation unit 2 according to the invention is shown in FIG. 1.
    The inertial unit 2 comprises at least one inertial sensor 4, a clock 5 and a computer 6.
    Each inertial sensor 4 is capable of detecting a displacement, such as a rotation about a corresponding predetermined axis, also called a “sensitive axis”, or a translation, or any combination of rotations and translations.
    For example, as illustrated in FIG. 1, the inertial unit 2 comprises three inertial sensors 4 having respectively a sensitive axis X-X, Y-Y and Z-Z. The inertial sensors 4 are fixed relative to each other.
    Advantageously, the sensitive axes X-X, Y-Y and Z-Z of each of the inertial sensors 4 are two by two non-parallel, for example orthogonal.
    The clock 5 is configured to deliver a clock signal representative of the passage of time.
    The computer 6 is configured to calculate the trajectory over time Xn (t) of a solid 7 fixed relative to the inertial unit 2. Such a solid 7 is, for example, an aircraft on board the inertial unit 2.
    In particular, the computer 6 is configured to calculate the trajectory Xn (t) of the solid 7 as a function of the clock signal and of signals coming from the inertial sensors 4 and described later.
    By “trajectory”, it is understood, within the meaning of the present invention, the data of the position, the speed and the attitude of the solid 7 in a predetermined reference trihedron linked to the Earth.
    By “attitude”, it is understood, within the meaning of the present invention, the data of the roll, pitch and heading angles formed by predetermined axes of the solid 7 and the axes of the predetermined reference trihedron. In this case, the trajectory Xn (t) of the solid 7 is a vector with nine components, that is to say three components of position, three components of speed and three angles of attitude.
    Each point of the trajectory Xn (t) is associated with a date, also called "instant", given by the clock signal coming from the clock 5.
    Each inertial sensor 4 comprises an ASG gyroscope 8, a MEMS gyroscope 10 and an accelerometer 11.
    The ASG gyroscope 8 has a sensitive axis defining the sensitive axis of the inertial sensor 4.
    The ASG gyroscope 8 is capable of delivering an ASG signal representative of a rotation of the inertial sensor 4 around the corresponding sensitive axis.
    The ASG 8 gyroscope is, for example, an NMR gyroscope or a comagnetometer, conventionally known.
    The ASG 8 gyroscope is associated with a start time Td, also called "start time". As soon as the ASG 8 gyroscope, also called “start-up”, is powered up, the ASG 8 gyroscope is only operational after a duration equal to the start-up duration Td.
    For example, the starting time Td is typically of the order of a minute.
    A gyroscope is said to be “operational”, within the meaning of the present invention, when it is in a nominal operating mode.
    The MEMS 10 gyroscope (acronym for "microelectromechanical Systems") is a gyroscope with an electromechanical microsystem conventionally known.
    The MEMS 10 gyroscope is integral with the ASG 8 gyroscope.
    The MEMS gyroscope 10 has a sensitive axis coincident with the sensitive axis of the ASG 8 gyroscope.
    The MEMS gyroscope 10 is capable of delivering a MEMS signal representative of a rotation of the inertial sensor 4 around the corresponding sensitive axis.
    The MEMS 10 gyroscope is operational at most a few seconds after it is started.
    For each inertial sensor 4, the corresponding MEMS gyroscope 10 is assumed to have an average bias bo over the starting time Td. Such a bias is homogeneous at an angular speed. The drift of the value of the bias bo over the starting time Td is assumed to be negligible compared to the value of the bias bo and compatible with the needs of the inertial navigation unit 2.
    The bias values for all three sensitive axes XX, YY, ZZ form a bias vector Bo with three components. The three components of the bias vector Bo, denoted box, bo y , boz, are respectively associated with the sensitive axes XX, YY and ZZ.
    In addition, for each inertial sensor 4, the corresponding ASG gyroscope 8 is assumed to present, at the end of the start time Td, a bias whose value and drift are negligible compared to the value of the bias bo of the associated MEMS gyroscope 10 .
    The accelerometer 11 has a sensitive axis, preferably coincident with the sensitive axis of the corresponding inertial sensor 4.
    The accelerometer 11 is capable of delivering an acceleration signal representative of the non-gravitational acceleration, also called "specific force", of the inertial sensor 4 along the corresponding sensitive axis.
    Preferably, the accelerometer 11 is integral with the ASG gyroscope 8 and the MEMS gyroscope 10.
    The accelerometer 11 is operational at most a few seconds after it is started.
    The computer 6 is connected to the ASG gyroscope 8 to receive the ASG signal. The computer 6 is also connected to the MEMS gyroscope 10 to receive the MEMS signal.
    The computer 6 is also connected to the accelerometer 11 to receive the acceleration signal. The computer 6 is also connected to the clock 5 to receive the clock signal.
    The computer 6 has a memory 12 and a processor 14.
    The memory 12 has a configuration location 16 and a storage location 18.
    The memory 12 is, moreover, configured to store navigation software 20, calculation software 22 and correction software 24.
    The configuration location 16 is configured to store the start-up time Td, an overlap time T re c, and a switching time T C om.
    For example, the switching time T CO m is typically of the order of a few seconds. The switching time T CO m is less than or equal to the recovery time T re c.
    The configuration location 16 is also configured to store, for each MEMS gyroscope 10, a predetermined arbitrary constant unit bias. For each sensitive axis X-X, Y-Y and Z-Z, the unit biases are respectively denoted DOx, DOy and DOz. The value of the unit biases DOx, DOy, DOz is preferably less than a few tenths of degrees per hour (° / h), for example DOx = DOy = DOz = 0.01 ° / h. Such a value minimizes linearization errors, as will be described later.
    The storage location 18 is configured to store the bias vector Bo.
    The recording location 18 is also configured to store the trajectory Xn (t) of the solid 7.
    The recording location 18 is, moreover, configured to store three biased trajectories XnDi (t) (i equal to x, y or z) of the solid 7 and an offset ÔXn, defined later.
    Each biased trajectory XnDi (t) is a trajectory calculated by supposing that the inertial unit 2 is, for the axis i (i equal to x, y or z), tainted with the corresponding unitary bias DOi.
    The navigation software 20 is configured to calculate, for each sensitive axis X-X, Y-Y, Z-Z, the evolution over time of the angular speed ω around the sensitive axis, illustrated by the curve 26 in FIG. 3.
    The navigation software 20 is also configured to calculate over time, and for each sensitive axis X-X, Y-Y, Z-Z, the value of a corresponding angle increment d0. For each of the sensitive axes XX, YY, ZZ, in the case of discretization of the numerical calculations allowing the calculation of Xn (t), the increment d0 is equal to the integral, between two successive instants, of the angular speed ω from the corresponding MEMS or ASG signal. The increment is noted ôOmems, respectively d0ASG, if it is obtained from the MEMS signal, respectively from the ASG signal.
    The navigation software 20 is also configured to calculate the trajectory over time Xn (t) of the solid 7, from the MEMS signal and / or the ASG signal and the acceleration signal supplied by each inertial sensor 4. In particular , the navigation software 20 is configured to calculate the path Xn (t) from the increments Ô0mems and / or d0ASG, and from each acceleration signal.
    In addition, the navigation software 20 is configured to calculate the three biased paths XnDi (t) over time of the solid 7 from the unit biases DOx, DOy, DOz stored in the configuration location 16, and from the MEMS signal. and / or the ASG signal, and the acceleration signal supplied by each inertial sensor 4. In particular, the navigation software 20 is configured to calculate the biased trajectories XnDi (t) from the unit biases DOx, DOy, DOz, increments ôOmems, ôOasg, and each acceleration signal.
    The calculation software 22 is configured to calculate the bias vector Bo.
    The calculation software 22 is also configured to calculate an angular correction ΔΘ between the MEMS signal and the ASG signal, defined later.
    The correction software 24 is configured to calculate the offset ÔXn.
    The processor 14 is adapted to execute each of the navigation software 20, the calculation software 22 and the correction software 24 stored in the memory 12 of the computer 6.
    The operation of the inertial navigation unit 2 will now be described, with reference to FIG. 3.
    During a start-up step, the ASG gyroscope 8, the MEMS gyroscope 10 and the accelerometer 11 of each inertial sensor 4 of the inertial unit 2 are started, that is to say powered up, at an instant t = 0 corresponding to the start of the start-up step.
    The start-up step has a duration equal to the start-up time Td stored in the configuration location 16. During the start-up step, the ASG 8 gyroscope is not operational.
    During the start-up step, the navigation software 20 calculates the trajectory over time Xn (t) of the solid 7 from the MEMS signal and the acceleration signal from each inertial sensor 4, that is that is to say that the navigation software 20 calculates the trajectory Xn (t) as a function of the increments d0MEMs and of each acceleration signal. Such a calculation is conventionally known.
    In addition, the navigation software 20 calculates the biased trajectories over time XnDi (t) of the solid 7.
    The calculation of the biased trajectories XnDi (t) differs from the calculation of the trajectory Xn (t) only in that the increments dÔMEMs obtained from the MEMS signal are increased by an increment of angle δθ.
    For example, in the case where the MEMS signal and the ASG signal are each of the discrete signals obtained by sampling, at a sampling frequency f e , of a corresponding continuous signal, the angle increment δθ is equal, for each sensitive axis XX, YY, ZZ, to the result of the division of the unit bias DOx, DOy, DOz corresponding by the sampling frequency f e , expressed in the appropriate unit.
    During the start-up step, the navigation software 20 writes, in the recording location 18, the trajectory Xn (t) and the biased trajectories XnDi (t) calculated.
    The step following the start-up step is a recovery step.
    During the recovery step, for each inertial sensor 4, each of the two gyroscopes ASG 8 and MEMS 10 is in an operational operating phase, the two gyroscopes ASG 8 and MEMS 10 being used jointly.
    During the recovery step, the ASG and MEMS signals are compared with each other in order to switch from the MEMS gyroscope 10 to the ASG gyroscope 8.
    By “switching”, it is understood, within the meaning of the present invention, the transition from a calculation of the trajectory Xn (t) from the MEMS signal to a calculation of the trajectory Xn (t) from the signal ASG.
    In addition, during the recovery step, the two signals ASG and MEMS are also used to estimate the bias vector Bo associated with each MEMS gyroscope 10. The start of the recovery step corresponds to an instant t = Td, also noted. td.
    The recovery step has a duration equal to the recovery time T rec stored in the configuration location 16, so that the recovery step ends at time t = Td + T re c, also noted t re c.
    The recovery step is broken down into a first phase, called switching, and a second phase.
    The first phase has a duration equal to the switching time Tœm stored in the configuration location 16. The first phase begins at the start of the recovery step, at time td, and ends at time t = Td + T com , also noted t CO m.
    Switching takes place at time t CO m.
    The second phase begins at time t CO m, and ends at the end of the recovery step, that is to say at time t re c.
    During the recovery step, the calculation software 22 calculates, for the sensitive axis i of each inertial sensor 4, a corresponding bias boi, equal to the average, preferably over the entire recovery step, of the difference between the angular speed from the MEMS signal and the angular speed from the corresponding ASG 8 gyroscope. As the bias of the ASG 8 gyroscope is deemed to be small compared to the bias of the MEMS gyroscope, the bias difference between the two MEMS 10 and ASG 8 gyroscopes is attributed to the MEMS 10 gyroscope.
    Then, for each component of the bias vector Bo associated with a sensitive axis XX, YY, ZZ, the calculation software 22 writes, in the recording location 18, the bias boi (i being equal to x, y or z), calculated for the MEMS gyroscope 10 of the corresponding inertial sensor 4. The duration T rec is chosen to allow filtering of the white noises ARW of the two gyroscopes MEMS 10 and ASG 8 in order to best estimate the bias vector Bo.
    The precision of estimation of each component of the bias vector Bo is given by formula (1):
    σ (δΜ = k RWm2 + q ARWr2 (1) Trec where o (ôb 0 ) is the standard deviation of the error of estimation of the bias of the MEMS gyroscope 10 (in ° / h);
    qARWm is the power spectral density of the white noise of drift of the MEMS 10 gyroscope (in ° / Vh); and qARWr is the power spectral density of the white drift noise of the ASG 8 gyroscope (in ° / Vh).
    For example, with a power spectral density qARWm of the white drift noise of the MEMS 10 gyroscope equal to 10 _3 ° / Vh, a power spectral density qARWr of the white drift noise of the ASG 8 gyroscope equal to 10 _3 ° / Vh, a duration recovery time T rec being 60 sec, the standard deviation of the error on the estimate o (ôb 0 ) through the MEMS gyroscope 10 is 0.011 ° / h.
    In addition, during the recovery step, the navigation software 20 calculates the trajectory Xn (t) and the biased trajectories XnDi (t) of the solid 7.
    More precisely, during the first phase, the navigation software 20 calculates the trajectory Xn (t) of the solid 7 from the MEMS signal and the acceleration signal from each inertial sensor 4. In particular, the navigation software 20 calculates the trajectory Xn (t) from the angle increment ôOmems from each MEMS signal, and from the acceleration signal from each inertial sensor 4.
    In addition, during the first phase, the navigation software 20 calculates the biased trajectories XnDi (t) of the solid 7 from the MEMS signal and the acceleration signal from each inertial sensor 4, and from the unit biases. More precisely, the navigation software 20 calculates the biased trajectories XnDi (t) from the angle increment d0MEMs from each MEMS signal, from the angle increment δθ determined from the unit biases DOi and from each acceleration signal.
    In addition, during the first phase, the calculation software 22 calculates, for each inertial sensor 4, a corresponding angular correction ΔΘ. The angular correction ΔΘ is equal to the average, over the switching time T CO m, of the values taken over time by the quantity (0 A sg - 0mems), where 0mems is a quantity calculated from the MEMS signal and equal , at a given instant, to the accumulation, from the instant td to the said instant, of the increments d0MEMs, and where 0asg is a quantity calculated from the signal ASG and equal, at a given instant, to the accumulation, from the instant td until audit given instant, increments d0 A sG, for the inertial sensor 4 considered. The angular correction ΔΘ is intended to correct the error induced, during switching, by the angular white noise on the measurements of the ASG and MEMS gyroscopes.
    Then, during the switching, the calculation software 22 transmits to the navigation software 20 the angular correction ΔΘ obtained at the end of the first phase, so as to ensure continuity between the measurements based on the MEMS gyroscopes 10 and the measurements based on ASG 8 gyroscopes.
    In addition, the navigation software 20 calculates the point of the path Xn (tcom) of the solid 7, at the instant t CO m, from an angle increment d0 CO m, and from each acceleration signal. .
    For a given sensitive axis, the corresponding angle increment d0 CO m is obtained by equation (2):
    d0com = 0asg (î com) 0MEMs (tcom - Te) -ΔΘ (2) where 0ASG (t C om) is the value taken by 0 A sc at time t C om;
    0MEMs (tcom-T e ) is the value taken by 0mems a sampling period before the instant tcom ' t and
    T e is the sampling period, equal to the inverse of the sampling frequency.
    In addition, during the switching, the correction software 24 rewrites, in the configuration location 16, the value of each unit drift DOx, DOy, DOz to assign it a zero value. This is due to the fact that, starting from the switching, the calculation of the trajectory Xn (t) and of the biased trajectories XnDi (t) is carried out from the ASG signals, the drift of the ASG 8 gyroscopes being assumed to be negligible compared to the drift MEMS 10 gyroscopes.
    For each sensitive axis, the navigation software 20 calculates the point of the biased trajectory XnD (t C om) of the solid 7, at the instant t CO m, from the angle increment dOcom and from each signal d acceleration, the value of each unit drift DOx, DOy, DOz having been fixed zero during switching.
    Switching to a calculation of the path Xn (t) (and the biased paths XnDi (t)) from the ASG signal instead of the MEMS signal is possible because the error linked to the switching depends mainly on the angular white noise on the ASG 8 and MEMS 10 gyroscope measurements.
    The standard deviation of the angular error committed linked to the switching is given by the relation (3):
    σ (θ) = / q BAm2 + q BAr2 (g) y T C om where qBAm is the power spectral density of the angular white noise of the MEMS 10 gyroscope (in prad / VHz);
    qBArest the power spectral density of the angular white noise of the ASG 8 gyroscope (in prad / VHz);
    σ (θ) is the standard deviation of the angular error due to switching.
    For example, for a power spectral density of the angular white noise equal to 1 prad / VHz for each of the two gyroscopes ASG 8 and MEMS 10, and a switching duration T CO m equal to 5 s, the standard deviation of the error angular σ (θ) linked to the switching is 0.63 prad.
    Then, during the second phase, the navigation software 20 calculates the trajectory Xn (t) of the solid 7 only from the signal ASG and the acceleration signal from each inertial sensor 4.
    The navigation software 20 also calculates the biased paths XnDi (t) of the solid 7 only from the ASG signal and the acceleration signal from each inertial sensor 4, the value of each unit drift DOx, DOy, DOz having been set to zero when switching.
    During the recovery step, the navigation software 20 writes, in the recording location 18, the trajectory Xn (t) and the biased trajectories XnDi (t) calculated.
    In summary, during the first phase, the navigation software 20 uses the angle increments d0MEMs originating from the MEMS signal; at the time of switching, the navigation software 20 uses the increment d0 CO m; then, during the second phase, the navigation software 20 uses the increments d0ASG originating from the signal ASG.
    The step following the recovery step is a correction step intended to correct the angular errors introduced by means of MEMS gyroscopes during the use of MEMS gyroscopes during the start-up step and the first phase of the recovery step.
    The correction step takes place on the date t re c.
    During the correction step, the navigation software 20 calculates the trajectory Xn (t) of the solid 7 from the signal ASG and the acceleration signal from each inertial sensor 4.
    In addition, during the correction step, the correction software 24 shifts the trajectory Xn (t) of the solid 7 by the value ÔXn at the instant t re c. The trajectory thus shifted is the trajectory which would have been calculated by the navigation software 20 if the ASG gyroscopes 8 had been operational from the start-up of the inertial unit 2.
    As described above, the correction software 24 calculates, during the start-up step and the recovery step, the three biased trajectories XnDi (t) (with i taking the value x, y or z) corresponding to the output data of the navigation algorithm when the MEMS 10 data is biased by a constant unit bias DO, (with i taking the value x, y or z). For example, XnDx (t) is the trajectory calculated by the navigation algorithm of the inertial navigation unit when the nominal measurements of the inertial sensor 4 of sensitive axis XX are offset by an additional unit bias D0 x . This bias D0, (stored in configuration location 16) takes two values depending on the instant considered:
    - from t = 0 to tcom, the value of this unit bias is fixed at a value which must be low to minimize linearization errors. Typically D0 x = D0 y = D0 z = 0.01 ° / h;
    - then, from tcom to trec, the biases are fixed at 0 because the switching leads to continuing navigation using the data from ASG gyroscopes considered without bias.
    At the start of the correction step, the correction software 24 calculates partial derivatives (with i taking the value x, y or z). Each partial derivative (with i taking the value x, y or z) is the derivative of the trajectory Xn (t) with respect to the unit drift DO, of the corresponding MEMS gyroscope 10, calculated with the following relation (4):
    JXd_XllDi (trec) Xh (t rec ),.,
    J Di - DOj '' where DOj is the unit bias associated with axis i.
    The correction software 24 then calculates the vectorial offset nXn from the estimate of the partial derivatives and from the estimate of the bias vector Bo of the MEMS gyroscope 10 according to formula (5):
    3X5
    ÔXn = b
    Ox
    3D0 Y + b oy
    3X5
    3D0 v + b oz - (5) Uz 3D0 7 ' 1 boi being the component i of the bias vector Bo.
    ÔXn is therefore a vector with nine dimensions.
    Then the correction software 24 readjusts the trajectory Xn (t) at the instant t re c by subtracting the correction term ÔXn from the trajectory Xn (t) according to the relation (6):
    Xn (after registration) = Xn (before registration) - ÔXn (6)
    In this way, the initial error due to the use of MEMS gyroscopes is corrected.
    The readjusted trajectory Xn (t) is called “nominal trajectory”.
    Once the trajectory Xn (t) has been reset, the calculation of the biased trajectories is interrupted because it is useless. Thereafter, the navigation software 20 continues the calculation of the trajectory Xn (t) of the solid 7 from the angle increment coming solely from the ASG signal and from the acceleration signal from each inertial sensor 4.
    The method for correcting the bias of the MEMS 10 gyroscope exposed above leads to linearizing Xn with respect to the three unit biases DOx, DOy and DOz. This imposes a bias value b Oi (with i taking the value x, y or z) which does not exceed a few tenths of degrees per hour, to avoid an excessively high navigation error leading to too high non-linearities rendering the correction formula 5 above.
    The trajectory Xn (t) calculated by the navigation software 20 after the switching is tainted by means of the MEMS gyroscopes 10, which introduces navigation errors in the calculation of the trajectory during the use of the MEMS gyroscopes 10 from the start-up of the inertial unit (t = 0) until the end of the first phase (t C om), these errors propagating until the end of the second phase of the recovery step (t re c). Thanks to such an inertial unit 2, such navigation errors are compensated for and a quick start of the inertial unit is possible.
    The method of correcting the errors induced by means of the MEMS gyroscopes which has been described above has the advantage, unlike a method which would consist in recalculating the whole trajectory from the beginning with measurements of the corrected MEMS gyroscopes. de Bo, to be simple to implement in real time, and not to require the storage of a large volume of data in a very short time.
    Such a method makes it possible to correct at the instant t re c the errors induced during the navigation from the instant t = 0 to the instant t CO m by the drift of the MEMS gyroscope, and this without requiring the replay (this is i.e. recalculation) of navigation from the start with gyroscope measurements corrected for the value of the bias.
    The use of MEMS and ASG gyroscopes allows continuous operation over time, such gyroscopes being capable of providing continuous measurement over time. This property makes it possible to use such gyroscopes in an inertial unit. Indeed, in particular for safety reasons, a discontinuity over time in the rotation angle or rotation speed measurements is not tolerable. The use of such gyroscopes is therefore advantageous compared to the use, for example, of material wave gyroscopes, which have the drawback of having a low bandwidth and of providing measurements that are discontinuous over time.
    In addition, the small dimensions and production costs of gyroscopes
    ASG and MEMS make inertial unit 2 inexpensive.
    权利要求:
    Claims (8)
    [1" id="c-fr-0001]
    1. A navigation aid method for an inertial unit (2) for fixed navigation relative to a solid (7), the inertial unit (2) comprising at least one inertial sensor (4) having a sensitive axis (XX, YY , ZZ), each inertial sensor (4) comprising an ASG gyroscope (8) and a MEMS gyroscope (10) integral with one another, the ASG gyroscope (8) being capable of delivering an ASG signal representative of a rotation around the corresponding sensitive axis (XX, YY, ZZ), the MEMS gyroscope (10) being capable of delivering a MEMS signal representative of a rotation around the corresponding sensitive axis (XX, YY, ZZ), the process comprising the steps:
    - calculation, between a first date (t = 0) and a third date (t CO m) later, of a trajectory (Xn (t)) and, for each inertial sensor (4), of a biased trajectory (XnDi (t)) corresponding, from MEMS signals, assuming, for the biased trajectory (XnDi (t)), that the inertial sensor (4) has a predetermined unit bias (DOx, DOy, DOz);
    - calculation, from the third date (t CO m), of the trajectory (Xn (t)) and of each biased trajectory (XnDi (t)) from the ASG signals, assuming, for the biased trajectory (XnDi (t)), that the inertial sensor (4) has a predetermined unit bias (DOx, DOy, DOz);
    - estimation of a bias vector (Bo) introduced by MEMS gyroscopes (10), from MEMS signals and ASG signals;
    - recalibration, on a fourth date (t rec ) subsequent to the third date (tcom), of the trajectory (Xn (t)) as a function of each biased trajectory (XnDi (t)), of unit biases (DOx, DOy, DOz) and the estimated bias vector (Bo), to obtain a nominal trajectory (Xn (t)) which is not tainted by MEMS gyroscopes (10).
    [2" id="c-fr-0002]
    2. Method according to claim 1, in which, for each MEMS gyrometer (10), the corresponding component (box, bo y , boz) of the bias vector (Bo) is equal to the average, between a second date (td) and the fourth date (t rec ), of the difference between an angular speed coming from the corresponding MEMS signal and an angular speed coming from the corresponding ASG signal, the second date (td) being between the first date (t = 0) and the third date (t C om).
    [3" id="c-fr-0003]
    3. Method according to claim 1 or 2, in which the nominal trajectory (Xn (t)) is obtained by subtracting an offset (ÔXn) from the trajectory (Xn (t)), the offset being a calculated vector correction term. according to:
    dXn dXn dXn ÔXn = b ° x 5DÔÏ + b ° y ôDÔÿ + b ° z dDto where ÔXn is the shift;
    boi is the i-th component of the bias vector (Bo); and the size is calculated according to:
    3D0i dXn _XnDi (t rec ) -Xn (t rec ) dD0i _ DOi where XnDi (trec) is the i-th biased trajectory (XnDi (t)) taken on the fourth date (t re c); and
    DOi is the predetermined unit bias associated with component i.
    [4" id="c-fr-0004]
    4. Method according to any one of claims 1 to 3, comprising a recovery step, the recovery step comprising:
    - between a second date (td) and the third date (t CO m), the second date (td) being between the first date (t = 0) and the third date (t CO m), a first phase to calculate the trajectory (Xn (t)) and each biased trajectory (XnDi (t)) from the MEMS signal;
    - on the third date (t CO m), a switch to calculate the trajectory (Xn (t)) and each biased trajectory (XnDi (t)) from a corresponding angle increment (dO C om), l 'angle increment (d0 CO m) being obtained, for each sensitive axis (XX, YY, ZZ), by the relation:
    dOcom = 0ASG (t com) - 0MEMs (tcorrrTe) -Δθ where dOcom is the angle increment;
    OASG (tcom) is a quantity equal to the cumulative increments (d0ASG) of angle of rotation around the sensitive axis (XX, YY, ZZ) between the second date (td) and the third date (t C om) , calculated from the ASG signal at switching;
    OMEMs (tcom-Te) is a quantity equal to the cumulative increments (d0MEMs) of angle of rotation around the sensitive axis (XX, YY, ZZ) between the second date (td) and a duration T e before the third date (t C om), calculated from the MEMS signal;
    each increment (d0MEMs, d0ASG) being equal to the integral, between two successive instants, of the angular speed (ω) of rotation around a sensitive axis (X-X, YY, Z-Z) coming from the corresponding MEMS or ASG signal,
    ΔΘ is a predetermined angular correction term; and
    T e is a predetermined duration;
    - between the third date (t C om) and the fourth date (t re c), a second phase to calculate the trajectory (Xn (t)) and each biased trajectory (XnDi (t)) from the ASG signal.
    [5" id="c-fr-0005]
    5. Method according to claim 4, in which the angular correction (ΔΘ) is equal to the average, between the second date (td) and the third date (t com), of the values taken over time by the quantity (0asg - 0mems), where 0asg is a quantity equal, at a given instant, to the accumulation from the second date (td) until said given instant, increments (d0ASG) of angle of rotation around the sensitive axis (XX, YY, ZZ) obtained from the ASG signal, for the inertial sensor 4 considered, and
    0mems is a quantity equal, at a given instant, to the accumulation, from the second date (td) until said given instant, of the increments (d0MEMs) of angle of rotation around the sensitive axis (XX, YY, ZZ) , obtained from the MEMS signal.
    [6" id="c-fr-0006]
    6. Method according to any one of claims 1 to 5, comprising, from the fourth date (t re c), the calculation of the nominal trajectory (Xn (t)) only from ASG signals.
    [7" id="c-fr-0007]
    7. Computer program product comprising program code instructions which, when executed by a computer, implement the navigation aid method according to any one of claims 1 to 6.
    5 8. Inertial unit (2) for fixed navigation relative to a solid (7), the inertial unit (2) comprising at least one inertial sensor (4) having a sensitive axis (XX, YY, ZZ), each inertial sensor (4) comprising an ASG gyroscope (8) and a MEMS gyroscope (10) integral with one another, the ASG gyroscope (8) being suitable for delivering an ASG signal representative of a rotation around the sensitive axis (XX, YY, ZZ)
    [8" id="c-fr-0008]
    10 corresponding, the MEMS gyroscope (10) being capable of delivering a MEMS signal representative of a rotation around the corresponding sensitive axis (XX, YY, ZZ), the inertial unit (2) further comprising a computer ( 6) configured to implement the navigation aid method according to any one of claims 1 to 6.
    S.59145
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3339805B1|2019-04-17|Method for assisting navigation, associated computer program product and inertial navigation unit
    CA2461595C|2013-01-15|Hybrid inertial navigation system with improved integrity
    EP0875002B1|1999-12-29|Aircraft piloting aid system using a head-up display
    EP2921923B1|2018-07-11|Method for tracking a transfer orbit or a phase of placing a space vehicle in orbit, in particular a vehicle with electric drive, and apparatus for implementing such a method
    EP2361368B1|2019-03-20|Method for determining a heading in the direction of true north using an inertial measurement unit
    EP2938965B1|2017-02-15|Method for comparing two inertial units integral with a same carrier
    EP3436779B1|2020-09-23|Device for measuring rotation, associated method and inertial navigation unit
    EP2223047B1|2011-05-25|Method for independent alignment of an inertial unit for an onboard instrument of an aircraft
    FR2789172A1|2000-08-04|APPARATUS FOR GYROMETERS AND ACCELEROMETERS FOR DETERMINING ATTITUDES OF AN AERODYNE
    EP3447654A1|2019-02-27|Method for determining the trajectory of a moving object, program and device for implementing said method
    EP1934558A1|2008-06-25|Device and method for correcting aging effects of a measurement sensor
    EP2558819B1|2014-07-16|Gyroscopic measurement in a navigation system
    FR3007840A1|2015-01-02|METHOD FOR DETECTING A FAILURE OF AT LEAST ONE SENSOR PRESENTED ON AN AIRCRAFT USING AN ANEMO-INERTIAL LOOP AND ASSOCIATED SYSTEM
    WO2009083374A1|2009-07-09|Method for independent alignment of an inertial unit for an onboard instrument to be mounted e.g. in an aircraft, and onboard instrument using such method
    FR2980570A1|2013-03-29|VIBRATION GYROSCOPE CALIBRATION
    EP2558820B1|2014-12-24|Gyroscopic measurement by a gyroscope vibrating in precession
    FR2892504A1|2007-04-27|Gyroscope e.g. hemispherical resonator gyroscope, controlling system, has gyroscopes outputting intermediate frequency signals with same frequency, and processor processing frequency signals to output directional signal
    EP2410293A1|2012-01-25|Method and system for harmonising an angular positioner reference in relation to an Earth-based reference frame
    EP2232273B1|2018-05-16|Method of correcting the gain of a capacitive unit and device for implementing same
    EP3896398A1|2021-10-20|Method for identifying a static phase of a vehicle
    EP3726183A1|2020-10-21|Method for determining the position and orientation of a vehicle
    EP2006707A1|2008-12-24|Method for determining a protection limit with compensation for calculation delays
    EP2373965B1|2016-06-22|Method for estimating a jet temperature in a jet engine
    FR3109991A1|2021-11-12|Method of estimating a physical quantity
    FR2958759A1|2011-10-14|METHOD AND DEVICE FOR RECLAIMING THE POSITION OF AN AIRCRAFT ON A FLIGHT
    同族专利:
    公开号 | 公开日
    US10401177B2|2019-09-03|
    EP3339805A1|2018-06-27|
    EP3339805B1|2019-04-17|
    US20180164102A1|2018-06-14|
    FR3060114B1|2019-05-17|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP2685212A2|2012-07-12|2014-01-15|Honeywell International Inc.|Multi-axis atomic inertial sensor system|
    WO2016107806A1|2014-12-30|2016-07-07|Thales|Hybrid inertia sensor employing cold atoms and mems and associated inertial platform|
    US9810549B2|2011-01-06|2017-11-07|University Of Utah Research Foundation|Systems, methods, and apparatus for calibration of and three-dimensional tracking of intermittent motion with an inertial measurement unit|
    US9026263B2|2011-11-30|2015-05-05|Alpine Electronics, Inc.|Automotive navigation system and method to utilize internal geometry of sensor position with respect to rear wheel axis|
    US9702674B2|2014-08-07|2017-07-11|The United States Of America As Represented By The Secretary Of The Army|Method and apparatus for GPS-denied navigation of spin-stabilized projectiles|
    FR3026193B1|2014-09-19|2016-12-23|Commissariat Energie Atomique|MAGNETOMETER WITHOUT ASSEMBLY AND COMPENSATION OF LOW FIELD RESONANCE SLOPE FLUCTUATIONS, MAGNETOMETER NETWORK AND MEASURING METHOD|
    FR3038730B1|2015-07-08|2017-12-08|Commissariat Energie Atomique|MAGNETOMETER ALL OPTICAL AND ISOTROPIC|
    US10097757B1|2017-03-24|2018-10-09|Fotonation Limited|Method for determining bias in an inertial measurement unit of an image acquisition device|
    GB2565264B|2017-05-23|2022-03-09|Atlantic Inertial Systems Ltd|Inertial navigation system|FR3049702B1|2016-03-31|2018-04-13|Commissariat A L'energie Atomique Et Aux Energies Alternatives|ROTATION MEASUREMENT DEVICE, METHOD AND INERTIAL NAVIGATION CENTER THEREFOR|
    FR3077884B1|2018-02-12|2021-01-01|Commissariat Energie Atomique|ELLIPTICAL POLARIZATION VECTOR MAGNETOMETER|
    JP2019168287A|2018-03-22|2019-10-03|トヨタ自動車株式会社|Attitude angle computing device, moving device, attitude angle computing method, and program|
    FR3083876B1|2018-07-16|2020-10-16|Commissariat Energie Atomique|ALIGNMENT VECTOR MAGNETOMETER WITH TWO DIFFERENTLY POLARIZED PROBE BEAMS|
    FR3093360B1|2019-02-28|2021-03-19|Commissariat Energie Atomique|Isotropic and all-optical scalar magnetometer|
    FR3093816B1|2019-03-12|2021-04-16|Commissariat Energie Atomique|Zero-field slave magnetometer with low-frequency filtering of the compensation field|
    US11268810B2|2019-07-10|2022-03-08|United States Of America As Represented By The Secretary Of The Navy|Angle random walk minimization for frequency modulated gyroscopes|
    法律状态:
    2018-01-02| PLFP| Fee payment|Year of fee payment: 2 |
    2018-06-15| PLSC| Search report ready|Effective date: 20180615 |
    2019-12-31| PLFP| Fee payment|Year of fee payment: 4 |
    2021-09-10| ST| Notification of lapse|Effective date: 20210806 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1662387A|FR3060114B1|2016-12-13|2016-12-13|NAVIGATION ASSISTANCE METHOD, COMPUTER PROGRAM PRODUCT, AND INERTIAL NAVIGATION CENTER|
    FR1662387|2016-12-13|FR1662387A| FR3060114B1|2016-12-13|2016-12-13|NAVIGATION ASSISTANCE METHOD, COMPUTER PROGRAM PRODUCT, AND INERTIAL NAVIGATION CENTER|
    EP17206281.2A| EP3339805B1|2016-12-13|2017-12-08|Method for assisting navigation, associated computer program product and inertial navigation unit|
    US15/840,139| US10401177B2|2016-12-13|2017-12-13|Navigational aid method, computer program product and inertial navigation system therefor|
    [返回顶部]