CN100559188C - A kind of field calibration method of optical fibre gyroscope inertia measurement unit - Google Patents

Abstract

A kind of field calibration method of optical fibre gyroscope inertia measurement unit the present invention relates to a kind of scene and accurately demarcates optical fibre gyroscope inertia measurement unit (Fiber Optic Gyro Inertial Measurement Unit, the FIMU) method of error coefficient.This method can, accurately calibrate gyro constant multiplier, gyroscope constant value drift, gyro misalignment and accelerometer and often be worth biasing 15 error coefficients six locational 12 rotations by FIMU at the use scene that does not have precise rotating platform totally.The present invention has the high and characteristics simple to operate of precision, has not only improved the service precision of FIMU, has also improved the efficient of demarcating simultaneously greatly.

Description

A kind of field calibration method of optical fibre gyroscope inertia measurement unit

Technical field

The present invention relates to a kind ofly accurately demarcate optical fibre gyroscope inertia measurement unit at the scene (Fiber Optic GyroInertial Measurement Unit, the FIMU) method of error coefficient are used in and use the on-site proving optical fibre gyroscope inertia measurement unit.

Background technology

Optical fibre gyro has the precision height, it is fast to start, dynamic range is big, anti-vibrating and impact and low cost and other advantages, is the development trend in inertia type instrument field.In recent years, the fast development of fiber-optics gyroscope promoted fiber-optic gyroscope strapdown inertial navigation system in land, sea, air, the application in day field.Optical fibre gyroscope inertia measurement unit FIMU is the core component of fiber-optic gyroscope strapdown inertial navigation system, its error comprises ascertainment error and stochastic error two parts, wherein ascertainment error be systematic error account for total error about 90%, be the topmost error source of fiber-optic gyroscope strapdown inertial navigation system.Therefore, fiber-optic gyroscope strapdown inertial navigation must be determined every error coefficient of FIMU before use by rating test, in fiber-optic gyroscope strapdown inertial navigation system it is compensated.

It is to carry out in having the laboratory of precise rotating platform that traditional FIMU demarcates, and the method for demarcation comprises two kinds of static multiposition test method and angular speed test methods.Static multiposition test method can calibrate whole error coefficients of FIMU, but precision is not high.The angular speed stated accuracy is higher, but can only determine constant multiplier and 9 error coefficients of alignment error two classes of FIMU.Two kinds of methods are combined, not only can calibrate whole error coefficients of FIMU, but also have very high stated accuracy, therefore in engineering, obtained using widely.

But correlative study shows, every error coefficient of calibrating of chamber turntable test is not changeless by experiment, comprises gyroscope constant value error, constant multiplier, alignment error and accelerometer constant error, constant multiplier, alignment error etc.These parameters change along with the use of system or the passing of resting period, and especially gyroscopic drift and acceleration biasing starts all inequalityly at every turn, and the time interval, long more variation was big more.Therefore, need carry out half a year or three months periodic calibrating once to FIMU usually, and traditional scaling method based on precise rotating platform is very complicated, this has increased huge workload for applying unit.Therefore, at the use scene of FIMU it is carried out every error coefficient and demarcate, not only can reduce even cancel periodic calibrating, can also improve the service precision of fiber-optic gyroscope strapdown inertial navigation.But, there is not accurate turntable at the scene as test benchmark, can't carry out accurate pointing to fiber-optic gyroscope strapdown inertial navigation, thus traditional static multiposition test and speed trial based on precise rotating platform all can't implement, for the on-the-spot high-precision calibrating of FIMU has brought very big difficulty.Some technician have carried out the on-site proving research of flexible gyroscope IMU, mainly contain two positions method and three position methods.The two positions method is with IMU forward Rotate 180 ° at the uniform velocity, reverse Rotate 180 ° at the uniform velocity again, then according to the simplification error model of IMU, the constant multiplier, the gyroscope constant value sum of errors accelerometer that calibrate IMU often are worth biasing 9 error coefficients of totally 3 classes, this method can not calibrate gyrostatic alignment error, and stated accuracy is not high, and main shortcoming is to need an accurately rotary positioning mechanism of Rotate 180 °, has increased the complicacy of testing; Three location methods do not need rotary positioning mechanism, often be worth 6 error coefficients of biasing two classes but this method only can calibrate gyroscope constant value drift and accelerometer, and the precision of demarcating are lower.

Summary of the invention

Technology of the present invention is dealt with problems and is: overcome the deficiencies in the prior art, a kind of optical fibre gyroscope inertia measurement unit field calibration method is provided, this method can be at the scene that does not have precise rotating platform by FIMU six locational 12 rotations, accurately calibrate gyro constant multiplier, gyroscope constant value drift, gyro misalignment and accelerometer and often be worth biasing 15 error coefficients of totally four classes.The present invention has the high and characteristics simple to operate of precision, has not only improved the service precision of FIMU, has also improved the efficient of demarcating simultaneously greatly.

Technical solution of the present invention is: a kind of field calibration method of optical fibre gyroscope inertia measurement unit, it is characterized in that FIMU is rotated 12 times on six positions, and can calibrate 15 error coefficients of FIMU, specifically may further comprise the steps:

(1) FIMU is placed on the surface level, the Z that makes FIMU axially on, start FIMU, wait for that the FIMU preheating finishes;

(2) FIMU is rotated counterclockwise 360 ° on this surface level, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output in record rotation and the static process;

(3) FIMU is turned clockwise on this surface level 360 °, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output in record rotation and the static process;

(4) make the Z of FIMU axially place revision test step (2) and (3) down;

(5) make on the X axis of FIMU respectively, place downwards, at each position revision test step (2) and (3);

(6) make FIMU and Y-axis make progress, place, respectively downwards at each position revision test step (2) and (3);

(7) utilize 12 groups of data that 12 times are rotated on six positions, error model according to FIMU, the gyro constant multiplier, gyroscope constant value drift, gyro misalignment and the accelerometer that adopt FIMU error coefficient computing formula to calibrate FIMU often are worth biasing 15 error coefficients of totally four classes, the error model of described FIMU comprises angular velocity channel error model and acceleration channel error model, and is as follows:

SU＝Mω+D

kN＝If+B

Wherein, S is the constant multiplier of optical fibre gyro, U is the output of fibre optic gyroscope, and M is the alignment error of optical fibre gyro, and ω is the input angular velocity of optical fibre gyro, D is the constant value drift of optical fibre gyro, k is the constant multiplier of accelerometer, and N is the output of accelerometer, and I is the alignment error of accelerometer, f is the input acceleration of accelerometer, and B is the normal value biasing of accelerometer;

U＝[U _x?U _y?U _z] ^T _1×3，ω＝[ω _x?ω _y?ω _z] ^T _1×3，D＝[D _x?D _y?D _z] ^T _1×3， $S={\left[\begin{array}{ccc}{S}_x& & \\ & S_y& \\ & & S_z\end{array}\right]}_{3\×3},$ $M={\left[\begin{array}{ccc}1& M_{\mathrm{xy}}& M_{\mathrm{xz}}\\ M_{\mathrm{yx}}& 1& M_{\mathrm{yz}}\\ M_{\mathrm{zx}}& M_{\mathrm{zy}}& 1\end{array}\right]}_{3\×3},$ N＝[N _x?N _y?N _z] ^T _1×3，f＝[f _x?f _y?f _z] ^T _1×3，B＝[B _x?B _y?B _z] ^T _1×3， $k={\left[\begin{array}{ccc}{k}_x& & \\ & k_y& \\ & & k_z\end{array}\right]}_{3\×3},$ $I={\left[\begin{array}{ccc}1& I_{\mathrm{xy}}& I_{\mathrm{xz}}\\ I_{\mathrm{yx}}& 1& I_{\mathrm{yz}}\\ I_{\mathrm{zx}}& I_{\mathrm{zy}}& 1\end{array}\right]}_{3\×3}.$

Principle of the present invention is: when the Z of FIMU carried out counterclockwise on axially and turns clockwise, the error equation of FIMU was as follows:

S _zU _z1 ⁺＝(ω _z1 ⁺+ω _ez1)+D _z+M _zy(ω _y1 ⁺+ω _ey1)+M _zx(ω _x1 ⁺+ω _ex1)(1)

S _xU _x1 ⁺＝(ω _x1 ⁺+ω _ex1)+D _x+M _xy(ω _y1 ⁺+ω _ey1)+M _xz(ω _z1 ⁺+ω _ez1)(2)

S _yU _y1 ⁺＝(ω _y1 ⁺+ω _ey1)+D _y+M _yx(ω _x1 ⁺+ω _ex1)+M _yz(ω _z1 ⁺+ω _ez1)(3)

S _zU _z1 ^-＝(ω _z1 ^-+ω _ez1)+D _z+M _zy(ω _y1 ^-+ω _ey1)+M _zx(ω _x1 ^-+ω _ex1)(4)

S _xU _x1 ^-＝(ω _x1 ^-+ω _ex1)+D _x+M _xy(ω _y1 ^-+ω _ey1)+M _xz(ω _z1 ^-+ω _ez1)(5)

S _yU _y1 ^-＝(ω _y1 ^-+ω _ey1)+D _y+M _yx(ω _x1 ^-+ω _ex1)+M _yz(ω _z1 ^-+ω _ez1)(6)

k _zN _z1＝f _z1+B _z+I _zyf _y1+I _zxf _x1(7)

Wherein, " 1 " in subscript expression the 1st position, i.e. the position of Z on axially, subscript "+" and "-" represent respectively counterclockwise with turn clockwise ω _Il ^k(i=x, y, z; K=+,-) angular velocity of rotation of the expression FIMU relative earth is at the projection of i axle, ω _Ei1 ^k(i=x, y, z; K=+,-) FIMU represented when the 1st inverse position (suitable) hour hands rotate, rotational-angular velocity of the earth ω _eProjection at the i axle.

Because optical fibre gyro system has all rotated 360 degree clockwise and counterclockwise, with equation (1)～(7) integration, then can get:

$S_x{\∫}_0^t{U_{x1}}^{+}\mathrm{d\τ}=0+{\∫}_0^t[D_x+M_{\mathrm{xz}}({{\mathrm{\ω}}_{z1}}^{+}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(9\right)$

$S_y{\∫}_0^t{U_{y1}}^{+}\mathrm{d\τ}=0+{\∫}_0^t[D_y+M_{\mathrm{yz}}({{\mathrm{\ω}}_{z1}}^{+}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(10\right)$

$S_x{\∫}_0^t{U_{x1}}^{-}\mathrm{d\τ}=0+{\∫}_0^t[D_x+M_{\mathrm{xz}}({{\mathrm{\ω}}_{z1}}^{-}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(12\right)$

$S_y{\∫}_0^t{U_{y1}}^{-}\mathrm{d\τ}=0+{\∫}_0^t[D_y+M_{\mathrm{yz}}({{\mathrm{\ω}}_{z1}}^{-}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(13\right)$

$k_z{N}_{z1}\·t={\∫}_0^t(f_{z1}+B_z)\mathrm{d\τ}---\left(14\right)$

For the 2nd～6 position (be Z axially down, on the X axis, under the X axis, Y-axis upwards and downward 5 positions of Y-axis), according to said method, can set up each locational 7 equation, the system of equations simultaneous solution is set up in the 1st～6 position, and the computing formula that can obtain the FIMU error coefficient is suc as formula shown in (15)～formula (18).

$D_i=S_i{\∫}_0^t({U_{\mathrm{ij}}}^{+}+{U_{\mathrm{ij}}}^{-}+{U_{i(j+1)}}^{+}+{U_{i(j+1)}}^{-})\mathrm{d\τ}/4t---\left(16\right)$

$B_i=\frac{1}{2}{k}_i(N_{\mathrm{ij}}+N_{i(j+1)})---\left(18\right)$

Wherein, n, i=x, y, z, n ≠ i; J represents j position, and during i=z, j=1; During i=x, j=3; During i=y, j=5, subscript+and subscript-expression is rotated counterclockwise and turns clockwise respectively, t is a data acquisition time.

The present invention's advantage compared with prior art is: the present invention utilizes it six locational 12 rotations at the use scene of optical fibre gyroscope inertia measurement unit, can accurately calibrate the gyro constant multiplier S of FIMU _x, S _y, S _z, gyroscope constant value drift D _x, D _y, D _z, gyro misalignment M _Xy, M _Xz, M _Yx, M _Yz, M _Zx, M _Zy, accelerometer often is worth biasing B _x, B _y, B _z, 15 error coefficients of totally four classes have improved the service precision of FIMU greatly, have improved the efficient of demarcating.

Description of drawings

Fig. 1 is the scene rotation rating test conceptual scheme of optical fibre gyroscope inertia measurement unit of the present invention, and wherein Fig. 1 a, Fig. 1 b, Fig. 1 c, Fig. 1 d, Fig. 1 e and Fig. 1 f are respectively the 1st to the 6th position of rating test;

Fig. 2 is the field calibration method process flow diagram of optical fibre gyroscope inertia measurement unit of the present invention.

Embodiment

As shown in Figure 1 and Figure 2, concrete implementation step of the present invention is as follows:

1, the preparation of optical fibre gyroscope inertia measurement unit (FIMU)

FIMU is placed on (do not require this surface level abswolute level, tiltable ± 5 are spent) on the surface level, makes the Z of FIMU axially go up placement, as shown in Figure 1a, start FIMU, and wait for that the FIMU preheating finishes.

2, FIMU is rotated counterclockwise 360 ° on this plane, kept then the static 1-3 of FIMU minute, write down the angular velocity and the acceleration of FIMU output in rotation and the static process, set up the error model of FIMU, shown in (19)～formula (20):

SU＝Mω+D (19)

kN＝If+B (20)

Set up the error equation of FIMU when the Z axle is rotated counterclockwise, suc as formula (21)～formula (24):

S _zU _z1 ⁺＝(ω _z1 ⁺+ω _ez1)+D _z+M _zy(ω _y1 ⁺+ω _ey1)+M _zx(ω _x1 ⁺+ω _ex1)(21)

S _xU _x1 ⁺＝(ω _x1 ⁺+ω _ex1)+D _x+M _xy(ω _y1 ⁺+ω _ey1)+M _xz(ω _z1 ⁺+ω _ez1)(22)

S _yU _y1 ⁺＝(ω _y1 ⁺+ω _ey1)+D _y+M _yx(ω _x1 ⁺+ω _ex1)+M _yz(ω _z1 ⁺+ω _ez1)(23)

k _zN _z1＝f _z1+B _z+I _zyf _y1+I _zxf _x1(24)

Wherein, S is the constant multiplier of optical fibre gyro, U is the output of fibre optic gyroscope, and M is the alignment error of optical fibre gyro, and ω is the input angular velocity of optical fibre gyro, D is the constant value drift of optical fibre gyro, k is the constant multiplier of accelerometer, and N is the output of accelerometer, and I is the alignment error of accelerometer, f is the input acceleration of accelerometer, and B is the normal value biasing of accelerometer.

With formula (22)～formula (25) integration, can get:

$S_x{\∫}_0^t{U_{x1}}^{+}\mathrm{d\τ}=0+{\∫}_0^t[D_x+M_{\mathrm{xz}}({{\mathrm{\ω}}_{z1}}^{+}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(26\right)$

$S_y{\∫}_0^t{U_{y1}}^{+}\mathrm{d\τ}=0+{\∫}_0^t[D_y+M_{\mathrm{yz}}({{\mathrm{\ω}}_{z1}}^{+}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(27\right)$

$k_z{N}_{z1}\·t={\∫}_0^t(f_{z1}+B_z)\mathrm{d\τ}---\left(28\right)$

3, FIMU is turned clockwise on this plane 360 °, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output are set up the error equation of FIMU when the Z axle turns clockwise according to the error model of FIMU in record rotation and the static process, suc as formula (29)～formula (31):

S _zU _z1 ^-＝(ω _z1 ^-+ω _ez1)+D _z+M _zy(ω _y1 ^-+ω _ey1)+M _zx(ω _x1 ^-+ω _ex1)(29)

S _xU _x1 ^-＝(ω _x1 ^-+ω _ex1)+D _x+M _xy(ω _y1 ^-+ω _ey1)+M _xz(ω _z1 ^-+ω _ez1)(30)

S _yU _y1 ^-＝(ω _y1 ^-+ω _ey1)+D _y+M _yx(ω _x1 ^-+ω _ex1)+M _yz(ω _z1 ^-+ω _ez1)(31)

With formula (29)～formula (31) integration, can get:

$S_x{\∫}_0^t{U_{x1}}^{-}\mathrm{d\τ}=0+{\∫}_0^t[D_x+M_{\mathrm{xz}}({{\mathrm{\ω}}_{z1}}^{-}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(33\right)$

$S_y{\∫}_0^t{U_{y1}}^{-}\mathrm{d\τ}=0+{\∫}_0^t[D_y+M_{\mathrm{yz}}({{\mathrm{\ω}}_{z1}}^{-}+{\mathrm{\ω}}_{\mathrm{ez}1})]\mathrm{d\τ}+0---\left(34\right)$

4, the Z of FIMU is axially placed down, shown in Fig. 1 b, FIMU is rotated counterclockwise 360 ° on this plane, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output are rotated counterclockwise 360 ° with FIMU then in record rotation and the static process on this plane, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output are set up 7 error equations according to the error model of FIMU, shown in (35)～(41) in record rotation and the static process;

$S_x{\&Integral;}_0^t{U_{x2}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_x+M_{\mathrm{xz}}({{\mathrm{\&omega;}}_{\mathrm{<figure-callout\; id="2"\; label="z"\; filenames="C20071006478400131.png"\; state="\{\{state\}\}">z</figure-callout>}2}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ez}2})]\mathrm{d\&tau;}+0---\left(36\right)$

$S_y{\&Integral;}_0^{\mathrm{<figure-callout\; id="2"\; label="t\; U\; y"\; filenames="C20071006478400131.png"\; state="\{\{state\}\}">t</figure-callout>}}{U_y}^{}{2_{}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_y+M_{\mathrm{yz}}({{\mathrm{\&omega;}}_{\mathrm{<figure-callout\; id="2"\; label="z"\; filenames="C20071006478400131.png"\; state="\{\{state\}\}">z</figure-callout>}2}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ez}2})]\mathrm{d\&tau;}+0---\left(37\right)$

$S_x{\&Integral;}_0^t{U_{x2}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_x+M_{\mathrm{xz}}({{\mathrm{\&omega;}}_{z2}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ez}2})]\mathrm{d\&tau;}+0---\left(39\right)$

$S_y{\&Integral;}_0^t{U_{y2}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_y+M_{\mathrm{yz}}({{\mathrm{\&omega;}}_{z2}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ez}2})]\mathrm{d\&tau;}+0---\left(40\right)$

$k_{\mathrm{<figure-callout\; id="2"\; label="z\; N\; z"\; filenames="C20071006478400131.png"\; state="\{\{state\}\}">z</figure-callout>}}{N}_z{2}_{}\&CenterDot;t={\&Integral;}_0^t({\mathrm{<figure-callout\; id="2"\; label="f\; z"\; filenames="C20071006478400131.png"\; state="\{\{state\}\}">f</figure-callout>}}_z+B_z)\mathrm{d\&tau;}---\left(41\right)$

5, make respectively on the X axis of FIMU, place downwards, shown in Fig. 1 c and Fig. 1 d, at each position, FIMU is rotated counterclockwise 360 ° on this plane, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output in record rotation and the static process, then FIMU is rotated counterclockwise 360 ° on this plane, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output in record rotation and the static process, 7 error equations are set up in each position of error model according to FIMU, and two positions are totally 14 error equations, shown in (42)～(55);

Shown in (42)～(55);

$S_z{\&Integral;}_0^t{U_{z3}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zx}}({{\mathrm{\&omega;}}_{x3}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ex}3})]\mathrm{d\&tau;}+0---\left(43\right)$

$S_y{\&Integral;}_0^t{U_{y3}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_y+M_{\mathrm{yx}}({{\mathrm{\&omega;}}_{x3}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ex}3})]\mathrm{d\&tau;}+0---\left(44\right)$

$S_z{\&Integral;}_0^t{U_{z3}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zx}}({{\mathrm{\&omega;}}_{x3}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ex}3})]\mathrm{d\&tau;}+0---\left(46\right)$

$S_y{\&Integral;}_0^t{U_{y3}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_y+M_{\mathrm{yx}}({{\mathrm{\&omega;}}_{x3}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ex}3})]\mathrm{d\&tau;}+0---\left(47\right)$

$k_x{N}_{z3}\&CenterDot;t={\&Integral;}_0^t(f_{x3}+B_x)\mathrm{d\&tau;}---\left(48\right)$

$S_z{\&Integral;}_0^t{U_{z4}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zx}}({{\mathrm{\&omega;}}_{x4}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ex}4})]\mathrm{d\&tau;}+0---\left(50\right)$

$S_y{\&Integral;}_0^t{U_{y4}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_y+M_{\mathrm{yx}}({{\mathrm{\&omega;}}_{x4}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ex}4})]\mathrm{d\&tau;}+0---\left(51\right)$

$S_z{\&Integral;}_0^t{U_{z4}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t\left[D_z{+M}_{\mathrm{zx}}({{\mathrm{\&omega;}}_{x4}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ex}4})\right]\mathrm{d\&tau;}+0---\left(53\right)$

$S_y{\&Integral;}_0^t{U_{y4}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_y+M_{\mathrm{yx}}({{\mathrm{\&omega;}}_{x4}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ex}4})]\mathrm{d\&tau;}+0---\left(54\right)$

$k_x{N}_{x4}\&CenterDot;t={\&Integral;}_0^t(f_{x4}+B_z)\mathrm{d\&tau;}---\left(55\right)$

6, the Y-axis of FIMU is made progress, place downwards, shown in Fig. 1 e and Fig. 1 f, at each position, FIMU is rotated counterclockwise 360 ° on this plane, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output in record rotation and the static process, then FIMU is rotated counterclockwise 360 ° on this plane, kept then the static 1-3 of FIMU minute, the angular velocity and the acceleration of FIMU output in record rotation and the static process, 7 error equations are set up in each position of error model according to FIMU, and two positions are totally 14 error equations, shown in (56)～(69);

$S_x{\&Integral;}_0^t{U_{x5}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_x+M_{\mathrm{xy}}({{\mathrm{\&omega;}}_{y5}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ey}5})]\mathrm{d\&tau;}+0---\left(57\right)$

$S_y{\&Integral;}_0^t{U_{z5}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zy}}({{\mathrm{\&omega;}}_{y5}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ey}5})]\mathrm{d\&tau;}+0---\left(58\right)$

$S_x{\&Integral;}_0^t{U_{x5}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_x+M_{\mathrm{xy}}({{\mathrm{\&omega;}}_{y5}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ey}5})]\mathrm{d\&tau;}+0---\left(60\right)$

$S_z{\&Integral;}_0^t{U_{z5}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zy}}({{\mathrm{\&omega;}}_{y5}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ey}5})]\mathrm{d\&tau;}+0---\left(61\right)$

$k_y{N}_{y5}\&CenterDot;t={\&Integral;}_0^t(f_{y5}+B_y)\mathrm{d\&tau;}---\left(62\right)$

$S_x{\&Integral;}_0^t{U_{x6}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_x+M_{\mathrm{xy}}({{\mathrm{\&omega;}}_{y6}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ey}6})]\mathrm{d\&tau;}+0---\left(64\right)$

$S_z{\&Integral;}_0^t{U_{z6}}^{+}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zy}}({{\mathrm{\&omega;}}_{y6}}^{+}+{\mathrm{\&omega;}}_{\mathrm{ey}6})]\mathrm{d\&tau;}+0---\left(65\right)$

$S_x{\&Integral;}_0^t{U_{x6}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_x+M_{\mathrm{xy}}({{\mathrm{\&omega;}}_{y6}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ey}6})]\mathrm{d\&tau;}+0---\left(67\right)$

$S_z{\&Integral;}_0^t{U_{z6}}^{-}\mathrm{d\&tau;}=0+{\&Integral;}_0^t[D_z+M_{\mathrm{zy}}({{\mathrm{\&omega;}}_{y6}}^{-}+{\mathrm{\&omega;}}_{\mathrm{ey}6})]\mathrm{d\&tau;}+0---\left(68\right)$

$k_y{N}_{y6}\&CenterDot;t={\&Integral;}_0^t(f_{y6}+B_y)\mathrm{d\&tau;}---\left(69\right)$

7, utilize on six positions 12 groups of data of 12 rotations, according to 7 equations setting up on each position, six positions are totally 42 equations, draw the error coefficient computing formula of FIMU, shown in (70)～formula (73),

$D_i=S_i{\&Integral;}_0^t({U_{\mathrm{ij}}}^{+}+{U_{\mathrm{ij}}}^{-}+{U_{i(j+1)}}^{+}+{U_{i(j+1)}}^{-})\mathrm{d\&tau;}/4t---\left(71\right)$

$B_i=\frac{1}{2}{k}_i(N_{\mathrm{ij}}+N_{i(j+1)})---\left(73\right)$

Wherein, n, i=x, y, z, n ≠ i; J represents j position, and during i=z, j=1; During i=x, j=3; During i=y, j=5, subscript+and subscript-expression is rotated counterclockwise and turns clockwise respectively, t is a data acquisition time.

Utilize formula (70)～formula (73) to calculate the gyro constant multiplier S of FIMU _x, S _y, S _z, gyroscope constant value drift D _x, D _y, D _z, gyro misalignment M _Xy, M _Xz, M _Yx, M _Yz, M _Zx, M _Zy, accelerometer often is worth biasing B _x, B _y, B _z, 15 error coefficients of totally four classes, so far, the demarcation of FIMU is finished.

Claims (2)

1, a kind of field calibration method of optical fibre gyroscope inertia measurement unit is characterized in that may further comprise the steps:

(1) optical fibre gyroscope inertia measurement unit is placed on the surface level, the Z that makes optical fibre gyroscope inertia measurement unit axially on, start optical fibre gyroscope inertia measurement unit, wait for that the optical fibre gyroscope inertia measurement unit preheating finishes;

(2) optical fibre gyroscope inertia measurement unit is rotated counterclockwise 360 ° on this surface level, keeps optical fibre gyroscope inertia measurement unit static then, the angular velocity and the acceleration of optical fibre gyroscope inertia measurement unit output in record rotation and the static process;

(3) optical fibre gyroscope inertia measurement unit is turned clockwise on this surface level 360 °, keep optical fibre gyroscope inertia measurement unit static then, the angular velocity and the acceleration of optical fibre gyroscope inertia measurement unit output in record rotation and the static process;

(4) make the Z of optical fibre gyroscope inertia measurement unit axially place revision test step (2) and (3) down;

(5) make on the X axis of optical fibre gyroscope inertia measurement unit respectively, place downwards, at each position revision test step (2) and (3);

(6) Y-axis of optical fibre gyroscope inertia measurement unit made progress, place downwards, at each position revision test step (2) and (3);

(7) utilize six locational 12 groups of data, error model according to optical fibre gyroscope inertia measurement unit, the gyro constant multiplier, gyroscope constant value drift, gyro misalignment and the accelerometer that adopt optical fibre gyroscope inertia measurement unit error coefficient computing formula to calibrate optical fibre gyroscope inertia measurement unit often are worth biasing 15 error coefficients of totally four classes, the error model of described optical fibre gyroscope inertia measurement unit comprises angular velocity channel error model and acceleration channel error model, and is as follows:

SU＝Mω+D

kN＝If+B

Wherein, S is the constant multiplier of optical fibre gyro, U is the output of fibre optic gyroscope, and M is the alignment error of optical fibre gyro, and ω is the input angular velocity of optical fibre gyro, D is the constant value drift of optical fibre gyro, k is the constant multiplier of accelerometer, and N is the output of accelerometer, and I is the alignment error of accelerometer, f is the input acceleration of accelerometer, and B is the normal value biasing of accelerometer;

U＝[U _x?U _y?U _z] ^T _1×3，ω＝[ω _x?ω _y?ω _z] ^T _1×3，D＝[D _x?D _y?D _z] ^T _1×3， $S={\left[\begin{array}{ccc}{S}_x& & \\ & S_y& \\ & & S_z\end{array}\right]}_{3\&times;3},$ $M={\left[\begin{array}{ccc}1& M_{\mathrm{xy}}& M_{\mathrm{xz}}\\ M_{\mathrm{yx}}& 1& M_{\mathrm{yz}}\\ M_{\mathrm{zx}}& M_{\mathrm{zy}}& 1\end{array}\right]}_{3\&times;3},$ N＝[N _x?N _y?N _z] ^T _1×3，f＝[f _x?f _y?f _z] ^T _1×3，B＝[B _x?B _y?B _z] ^T _1×3， $k={\left[\begin{array}{ccc}{k}_x& & \\ & k_y& \\ & & k_z\end{array}\right]}_{3\&times;3},$ $I={\left[\begin{array}{ccc}1& I_{\mathrm{xy}}& I_{\mathrm{xz}}\\ I_{\mathrm{yx}}& 1& I_{\mathrm{yz}}\\ I_{\mathrm{zx}}& I_{\mathrm{zy}}& 1\end{array}\right]}_{3\&times;3};$

Described optical fibre gyroscope inertia measurement unit error coefficient computing formula is:

$D_i=S_i{\&Integral;}_0^t({U_{\mathrm{ij}}}^{+}+{U_{\mathrm{ij}}}^{-}+{U_{i(j+1)}}^{+}+{U_{i(j+1)}}^{-})\mathrm{d\&tau;}/4t$

$B_i=\frac{1}{2}{k}_i(N_{\mathrm{ij}}+N_{i(j+1)})$

Wherein, n, i=x, y, z, n ≠ i; J represents j position, and during i=z, j=1; During i=x, j=3; During i=y, j=5, subscript+and subscript-expression is rotated counterclockwise and turns clockwise respectively, t is a data acquisition time.

2, the field calibration method of optical fibre gyroscope inertia measurement unit according to claim 1 is characterized in that: be 1-3 minute the maintenance optical fibre gyroscope inertia measurement unit rest time in described step (2) or the step (3).

Images