System and a method of analyzing and monitoring interfering movements of an inertial unit during a stage of static alignment

Abstract

A system and to a method of analyzing and monitoring interfering movements of an inertial unit of an aircraft during a stage of statically aligning the inertial unit. During the static alignment stage, measurements of the velocity of the aircraft relative to the ground are acquired, and states of a mirror process having a model that is close to the model of the process of aligning the inertial unit are estimated. The states of the mirror process are estimated from observations constituted by the measurements of velocity relative to the ground. Finally, the estimates of the states are compared with respective validation thresholds in order to validate or not validate said alignment of the inertial unit.

Claims

1. A method of analyzing and monitoring interfering movements of an inertial unit of an aircraft during a stage of statically aligning the inertial unit, the method being wherein during the stage of statically aligning the inertial unit, the following steps are performed: an acquisition step for acquiring measurements of the movement of the aircraft relative to the ground by a movement sensor; an estimation step for estimating states of a mirror process of a model that is simplified and/or approximated relative to the model of the static alignment process of the inertial unit, the states of the mirror process being estimated from observations constituted by the movement measurements; and an additional comparison step comprising comparing data supplied by the inertial unit to detect if interfering movements are present to cause the supplied data to be erroneous; if the supplied data is determined to be erroneous, either generate an alignment invalid signal or correct the supplied data in order to make the supplied data supplied usable.

2. The method according to claim 1, wherein validation thresholds for validating the static alignment are defined prior to the stage of statically aligning the inertial unit, each validation threshold corresponding to a respective one of the estimated states, and the additional comparison step comprises comparing the absolute value of at least one estimate of one of the states with at least one validation threshold.

3. The method according to claim 2, wherein during the additional comparison step, the absolute value of at least one estimate of the state is compared with the corresponding validation threshold, and an “alignment invalid” signal is activated whenever the absolute value of an estimate of a state is greater than the corresponding validation threshold.

4. The method according to claim 2, wherein during the additional comparison step, at least two estimates of the states are combined in order to form an estimated state combination, and an “alignment invalid” signal is activated when the estimated state combination is greater than a global threshold.

5. The method according to claim 4, wherein the estimated state combination is equal to a weighted sum of the squares of at least two estimates of the states.

6. The method according to claim 1, wherein the additional comparison step comprises an additional correction step that comprises the correcting data supplied by the inertial unit in order to make the supplied data by the inertial unit usable, the additional correction step using the estimates of the orientation and velocity errors of the inertial unit resulting from interfering movements during the alignment stage.

7. The method according to claim 6, wherein during the additional correction step, the correction is calculated from the estimated states, each estimated state being used as an initial value for a process of estimating errors of the inertial unit, the process of estimating errors being maintained throughout the duration of a stage of navigation of the inertial unit following the alignment stage, and the maintained estimate of the errors being subtracted from the data supplied by the inertial unit.

8. The method according to claim 1, wherein during the acquisition step, the measurements of movements of the aircraft relative to the ground are measurements of a velocity ({right arrow over (v)}.sub.g) of the aircraft relative to the ground.

9. The method according to claim 1, wherein during the acquisition step, the measurements of movement of the aircraft relative to the ground are measurements of a position ({right arrow over (x)}.sub.g) of the aircraft relative to the ground.

10. The method according to claim 1, wherein the estimation step of states of the mirror process consist in estimating the coefficients of at least one polynomial function of time close to the movement measurements.

11. The method according to claim 10, wherein the movement measurements are measurements of velocity {right arrow over (v)}.sub.g relative to the ground, and the model of the mirror process is suitable for generating a second degree polynomial function of time for the North/South component of velocity relative to ground, and a first degree polynomial function of time for the East/West component of velocity relative to the ground.

12. The method according to claim 10, wherein the estimation step of states of the mirror process is performed by the least squares method in which the coefficients for identification are those of the at least one polynomial function of time.

13. The method according to claim 1, wherein the estimation step of states of the mirror process consist in Kalman filtering for which the states are those of the mirror process and for which the observations are the measurements of movement relative to the ground.

14. The method according to claim 1, wherein the movement sensor comprises at least one of a velocity sensor capable of measuring velocity of the aircraft relative to the ground and a position sensor capable of measuring position of the aircraft relative to the ground.

15. The method according to claim 14, wherein the movement sensor comprises a GNSS receiver or a Doppler effect radar.

16. A system for analyzing and monitoring interfering movements of an inertial unit of an aircraft during a stage of statically aligning the inertial unit, the system comprising: a movement sensor for sensing movement of the aircraft and supplying measurements of the movement of the aircraft relative to the ground; an estimator for estimating a mirror process close to the static alignment process of the inertial unit, the estimator being provided with at least one calculator and with at least one memory storing thresholds for validating the static alignment of the inertial unit and calculation instructions, the estimator serving to estimate states of the mirror process having a model that is close to the model of the static alignment process of the inertial unit, the estimation step of the states of the mirror process being performed on the basis of observations constituted by measurements of movement supplied by the movement sensor, wherein data supplied by the inertial unit is used to generate an alignment invalid signal when an estimated state combination is greater than or equal to a global threshold and causing restarting statically aligning the inertial unit.

17. A method of analyzing and monitoring interfering movements of an inertial unit of an aircraft during statically aligning the inertial unit, the method comprising: acquiring measurements of the movement of the aircraft relative to the ground by a movement sensor; estimating states of a mirror process of a model that is simplified and/or approximated relative to the model of the static alignment process of the inertial unit, the states of the mirror process being estimated from observations constituted by the movement measurements; and comparing data supplied by the inertial unit to a threshold to determine if the supplied data is useable; if the supplied data is useable, using the supplied data to statically align the inertial unit, if the supplied data is not useable, correcting the data or restarting statically aligning the inertial unit in order to make the data supplied by the inertial unit usable.

18. The method according to claim 17, wherein if an estimated state combination is greater than or equal to a global threshold, an alignment invalid signal is generated.

19. A method of analyzing and monitoring interfering movements of an inertial unit of an aircraft during a stage of statically aligning the inertial unit, the method being wherein during the stage of statically aligning the inertial unit, the following steps are performed: acquiring measurements of the movement of the aircraft relative to the ground by a movement sensor; and estimating states of a mirror process of a model that is simplified and/or approximated relative to the model of the static alignment process of the inertial unit, the states of the mirror process being estimated from observations constituted by the movement measurements; and comparing data supplied by the inertial unit to detect if interfering movements are present to cause the supplied data to be erroneous to determine if the stage of statically aligning the inertial unit with the data supplied by the inertial unit should continue, if the supplied data is useable, statically aligning the inertial unit, if the supplied data is not useable, making the data supplied by the inertial unit usable.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention and its advantages appear in greater detail from the context of the following description of implementations given by way of illustration and with reference to the accompanying figures, in which:

(2) FIG. 1 shows an aircraft having a system for analyzing and monitoring interfering movements of an inertial unit;

(3) FIG. 2 shows a system for analyzing and monitoring interfering movements of an inertial unit;

(4) FIG. 3 is a diagram summarizing a method of analyzing and monitoring interfering movements of an inertial unit;

(5) FIG. 4 shows a model of an alignment process of an inertial unit;

(6) FIG. 5 shows a mirror process; and

(7) FIG. 6 shows measurements of movement of the aircraft.

(8) Elements present in more than one of the figures are given the same references in each of them.

DETAILED DESCRIPTION OF THE INVENTION

(9) FIG. 1 shows a rotary wing aircraft 20. The aircraft 20 has an inertial unit 1 and a system 10 for analyzing and monitoring interfering movements of the inertial unit 1 during a stage of aligning the inertial unit 1. The system 10 is shown in detail in FIG. 2 and it includes a movement sensor 2 for sensing movement of the aircraft 20 and an estimator 5 having a calculator 3 and a memory 4.

(10) The movement sensor 2 is a GNSS receiver and it supplies accurate measurements of movement of the aircraft 20 relative to the ground, which measurements may be velocity or position measurements, these measurements being based by way of example on phase increments of carrier waves of signals transmitted by satellites in at least one GNSS system. The memory 4 stores calculation instructions and possibly thresholds for validating the alignment of the inertial unit 1. The calculator 3 makes use of these calculation instructions, of the movement measurements, and where applicable of the thresholds for validating the alignment of the inertial unit 1 in order to perform a method of analyzing and monitoring interfering movements of the inertial unit 1, as shown in the summary diagram of FIG. 3.

(11) The movement sensor 2 is connected to the estimator 5 in order to supply it with measurements of the movement of the aircraft 20 relative to the ground. The inertial unit 1 is connected to the system 10 in order to supply it with a start signal t.sub.0 and an end signal t.sub.1 marking the start and the end of the alignment stage.

(12) The method of analyzing and monitoring interfering movements of the inertial unit 1 during an alignment stage comprises two steps.

(13) An acquisition step 110 for acquiring measurements of the movement of the aircraft 20 relative to the ground is performed by means of the movement sensor 2 during the stage of aligning the inertial unit 1.

(14) An estimation step 120 of estimating states of a mirror process is also performed from the observations constituted by the movement measurements. The model of the mirror process has a structure that is close to the structure of the model of the alignment process of the inertial unit 1.

(15) The model of the mirror process could equally well be rigorously identical to the model of the alignment process of the inertial unit 1.

(16) During the alignment stage, the alignment process seeks to estimate the vertical direction, by zeroing the two angles of inclination of the inertial unit 1 about North/South and East/West geographical axes, to estimate the direction of North by zeroing the misalignment angle of the inertial unit 1 about the vertical axis, and finally to estimate the components of the velocity of the aircraft 20 relative to the ground. By way of example, the alignment process of an inertial unit 1 is a system having five states, comprising: i) the three angular differences between the axes of the virtual platform of the inertial unit 1 and the corresponding directions of the local geographical axes; and ii) the two horizontal components of the velocity of the inertial unit 1 relative to the ground.

(17) If the inertial unit 1 is genuinely stationary during the alignment stage, these angular differences and horizontal velocity components converge towards zero values and the inertial unit 1 is then correctly initialized.

(18) A block diagram of an example of a model of the process of aligning the inertial unit 1 is shown in FIG. 4. For this alignment process, θ.sub.n is the orientation error about the North/South axis, θ.sub.e is the orientation error about the East/West axis, θ.sub.d is the orientation error about a vertical axis, v.sub.n is velocity along the North/South axis, and v.sub.e is the velocity along the East/West axis.

(19) This model of the alignment process takes account of the known latitude ϕ of the aircraft 20 and uses both the modulus of the acceleration of terrestrial gravity g and a vector representing the angular velocity {right arrow over (Ω)}.sub.E of the earth about its axis. Two projections Ω.sub.n, Ω.sub.d of this angular velocity vector {right arrow over (Ω)}.sub.E, respectively onto the North/South axis and onto the vertical axis and depending on the latitude ϕ are calculated as follows:
Ω.sub.n=Ω.sub.E.Math.cos ϕ and Ω.sub.d=Ω.sub.E.Math.sin ϕ

(20) In FIG. 4, the symbol .Math. represents an adder and the symbol represents an integrator. The acceleration errors γ.sub.n along the North/South axis and γ.sub.e along the East/West axis are also marked.

(21) An estimator of the states of the alignment process may then be a Kalman filter having these five states. In other examples of the alignment process, one or more original states may optionally be used. For example, the latitude ϕ of the aircraft 20, should it be unknown, may constitute an additional state of the alignment process and may then be determined by the estimator.

(22) The mirror process model used by the method of the invention may be a model that is rigorously identical to the alignment process model shown in FIG. 4. Under such circumstances, both models have the same number of states and the same matrices defining the relationship between those states.

(23) It is also possible for the mirror process model to be simplified or indeed approximated relative to the alignment process model. For example, the 24-hour mode associated with rotation of the earth has specifically been ignored in order to set up the mirror process shown in the form of a block diagram in FIG. 5.

(24) When operating in an open loop, the model of this FIG. 5 mirror process generates time velocity profiles on its output having the form of two time polynomials:
v.sub.e(t)=θ.sub.n0.Math.g.Math.t+V.sub.e0
and
v.sub.n(t)=−½.Math.(θ.sub.d0.Math.Ω.sub.n+θ.sub.n0.Math.Ω.sub.d).Math.g.Math.t.sup.2+θ.sub.e0.Math.g.Math.t+V.sub.n0

(25) From these equations, it can be deduced that the movements having an effect on the accuracy of the alignment of the inertial unit 1 are movements consisting in: i) a speed ramp along the East/West axis, said ramp being defined by the coefficients θ.sub.n0 and V.sub.e0; and ii) a velocity parabola along the North/South axis, said parabola being defined by the coefficients θ.sub.d0, θ.sub.n0, θ.sub.e0, and V.sub.n0.

(26) By using the movement measurements, which in this example are measurements of the speed of the aircraft 20 relative to the ground during the stage of aligning the inertial unit 1, the method of the invention makes it possible to identify the coefficients of these two polynomial functions. FIG. 6 shows these measurements of the velocity of the aircraft 20 relative to the ground along the North/South axis obtained during the alignment stage, i.e. between the start t.sub.0 and the end t.sub.1 of the alignment stage, together with a representation of the polynomial function corresponding to this speed along the North/South axis.

(27) The coefficients of these polynomial functions are directly associated with the states of the mirror process. The estimation step 120 of estimating the states of the mirror process consists either in estimating these coefficients for each polynomial function, from which coefficients these states are deduced, or else in estimating the states directly. The states of the mirror process can be estimated from observations constituted by the movement measurements by using known mathematical methods such as the non-recursive least squares method or indeed the recursive least squares method, or else by using a Kalman filter.

(28) The method of the invention may make use during an additional step 130, 140 of these estimated errors for the orientation and the velocities of the inertial unit 1 due to the interfering movements during the alignment stage.

(29) In a first variant of the invention, an additional comparison step 130 of the method comprises comparing the absolute value of at least one estimate of the orientation and velocity errors of the inertial unit 1 with at least one validation threshold, and activating an “alignment invalid” signal if at least one of the thresholds is exceeded.

(30) Activating this “alignment invalid” signal then indicates that the accuracy of the data supplied by the inertial unit 1 is deemed to be insufficient, and that the stage of aligning the inertial unit 1 must be restarted, for example, or that the aircraft must be operated in a mode that does not rely on inertial measurements.

(31) In a second variant of the invention, an additional correction step 140 of the method consists in correcting the data supplied by the inertial unit 1. This correction 140 makes use of the estimated orientation and velocity errors of the inertial unit 1 that results from interfering movements during the alignment stage and that have already been calculated, in order to improve the data supplied by the inertial unit 1 and make that data sufficiently accurate to be usable.

(32) Naturally, the present invention may be subjected to numerous variations as to its implementation. Although several implementations are described, it will readily be understood that it is not conceivable to identify exhaustively all possible implementations.

(33) For example, the movement sensor may be arranged outside the system 10 and may provide its movement measurements to the system 10 so that the system can use them.

(34) Naturally, it is possible to envisage replacing any of the means described by equivalent means without going beyond the ambit of the present invention.