CN105180946A - Wideband measurement-based satellite high-precision attitude determination method and system thereof - Google Patents

Abstract

The invention discloses a wideband measurement-based satellite high-precision attitude determination method and a system thereof. The method comprises the following steps: completing output sampling of a gyro, a star sensor and an angle displacement sensor, and preprocessing sampled gyro data to eliminate high frequency noises output by the gyro; carrying out low pass filtering on the gyro data, and recurring current gyro pre-estimation attitude information according to the estimate of a last inertia measurement assembly; calculating the attitude estimate of the inertia measurement assembly; resolving an attitude correction signal after obtaining the attitude estimate of the inertia measurement assembly; and resolving an attitude matrix through combining the measurement value of the angle displacement sensor after obtaining the attitude correction signal, and resolving an attitude angle. The invention also discloses the system for realizing the method. The method and the system break limitations of an inertia attitude measurement system, so the satellite attitude measurement bandwidth is improved; and a multi-sensor information fusion technology is adopted, so the precision of the attitude determination system is improved.

Description

The satellite high-precision attitude determination method measured based on wideband and system

Technical field

The present invention relates to satellite wideband high-precision attitude determination technology, particularly, relate to a kind of satellite high-precision attitude determination method measured based on wideband.

Background technology

For ensureing the accurate control performance of satellite in orbit, need to utilize multiple sensor to complete the high-acruracy survey of satellite body relative attitude benchmark, to ensure the performance that attitude is determined, the result that attitude is determined the most at last exports, and the generation for controlled quentity controlled variable provides basis.For realizing correlation function, spaceborne computer needs the operation such as output sampling, Multi-source Information Fusion algorithm carrying out attitude measurement sensor.

At present in domestic high precision three axis stabilized satellite in orbit, the general high-precision attitude measuring system all adopting gyro and star sensor composition, state equation is set up by the Kinematics Law of satellite, by the output of gyro and star sensor by certain information fusion algorithm, obtain the attitude information of satellite.By the limitation of inertial sensor responsive bandwidth, when high frequency attitude disturbance be can not ignore, need to utilize the attitude measurement sensor with high-bandwidth response characteristic to provide attitude information.

In addition, domestic a kind of successful development with the attitude sensor part angular displacement sensor of high pass characteristic, compensate for the defect of Inertial Measurement Unit bandwidth deficiency, the measurement of high bandwidth attitude information is become a reality, but its shortcoming is to low frequency signal reproduction ability.

What require along with three axis stabilized satellite control accuracy improves constantly, and needs to determine the attitude measurement method of satellite steady-state operation in the design phase, studies a kind of wideband attitude determination method based on multi-sensor Theory of Information Fusion.

Summary of the invention

For the defect existed in above-mentioned prior art, the invention provides a kind of satellite high-precision attitude determination method measured based on wideband, by increasing the sensor of the responsive high frequency attitude information of a kind of energy, can on the basis of conventional inertia attitude measurement system, the attitude information realizing below 500Hz is measured, by information fusion method, improve the precision that three axis stabilized satellite attitude is determined.

For achieving the above object, the technical solution used in the present invention is as follows:

Based on the satellite high-precision attitude determination method that wideband is measured, comprise the steps:

Satellite, under equilibrium mode, completes and samples to the output of gyro, star sensor, angular displacement sensor, and the gyro data deduction constant value drift obtained sampling;

Carry out low-pass filtering to gyro data, eliminate the measurement noises that gyro exports, after gyro data process completes, utilize the attitude kinematics equations of satellite, clap the attitude of inertial measurement cluster estimation in conjunction with upper one, attitude information estimated by the current gyro of recursion;

Calculate the Attitude estimation value of current bat inertial measurement cluster;

After obtaining the Attitude estimation value of inertial measurement cluster, calculate attitude correction signal;

After obtaining attitude correction signal, in conjunction with the measured value of angular displacement sensor, calculate attitude matrix, and then calculate attitude angle.

Using the constant value drift increment of attitude misalignment hypercomplex number and gyro as quantity of state, star sensor is asked to export the attitude misalignment gone out with gyro recursion, with this deviation for measured value; According to the transfer law of attitude misalignment, the predicted value of predicted state amount; After this, according to the measurement model of gyro, star sensor, calculate the Attitude estimation value of current bat inertial measurement cluster.

The Attitude estimation value of inertial measurement cluster and the attitude output valve of attitude and heading reference system are asked deviation, by this deviation signal successively by low-pass filter, controller, produces attitude correction signal;

With the measured value of attitude correction signal and angular displacement sensor for input, calculate attitude matrix, and then calculate attitude angle.

Compared with prior art, its advantage and beneficial effect are the method that the present invention adopts:

By a kind of satellite high-precision attitude determination method measured based on wideband, breach the limitation of inertial posture measuring system, the bandwidth of satellite attitude measurement is improved; By multi-sensor information fusion method, improve the precision of attitude and heading reference system.

The inventive method is particularly useful for super steady remote sensing of the earth and super quiet inertial space is stablized in sensing task, on the basis of conventional inertia attitude measurement system, increase a kind of high-precision angular displacement sensor, complete the measurement to high frequency attitude information, control with the high precision high stability degree realizing spacecraft.

Accompanying drawing explanation

By reading the detailed description done non-limiting example with reference to the following drawings, other features, objects and advantages of the present invention will become more obvious:

Fig. 1 is data flow of the present invention and calculation flow chart;

Fig. 2 is that Attitude Calculation flow process estimated by gyro data process and gyro;

Fig. 3 is inertial posture measuring system Attitude Calculation flow process;

Fig. 4 is LFC low-frequency correction signal calculation process;

Fig. 5 is that wideband attitude information estimates flow process.

Accompanying drawing variable-definition:

Q _g: attitude quaternion estimated by gyro;

Q _s: star sensor exports attitude quaternion;

B: gyroscope constant value drift;

Q: inertial measurement cluster estimates attitude quaternion;

θ _g: inertial measurement cluster estimates Eulerian angle;

θ _cm: attitude correction signal;

θ: wideband Attitude estimation Eulerian angle.

Embodiment

Below in conjunction with specific embodiment, the present invention is described in detail.Following examples will contribute to those skilled in the art and understand the present invention further, but not limit the present invention in any form.It should be pointed out that to those skilled in the art, without departing from the inventive concept of the premise, some distortion and improvement can also be made.These all belong to protection scope of the present invention.

As shown in Figure 1, for the concrete steps that the present invention adopts, when satellite is in stable state, in each execution cycle, first the output of gyro, star sensor and angular displacement sensor is sampled, then pre-service is carried out to the data of gyro, eliminate the high frequency noise that gyro exports, the gyro to measure angular velocity integration after process is obtained the attitude quaternion information that gyro is estimated; Calculate star sensor afterwards to export and estimate the deviation hypercomplex number of attitude with gyro, in conjunction with the deviation hypercomplex number of one-step prediction and the yield value that calculates based on gyro and star sensor measurement model, calculate the optimal estimation value of low frequency attitude; Subsequently the difference between the current low frequency optimal estimation signal that obtains and previous step wideband attitude optimal estimation value is processed successively, obtain the wideband attitude correction signal with low frequency characteristic, in conjunction with the high frequency attitude information that angular displacement sensor exports, calculate attitude matrix; Last according to attitude matrix, calculate wideband attitude angle.

As shown in Figure 2, the gyro data sequence x of collection _in (), after deduction gyroscopic drift b, the sequence obtained is X _in (), the sequence exported afterwards is after filtering Y _n, filtering recurrence equation is:

$Y_i\left(n\right)=\frac{T}{\mathrm{\τ}}\left(X_i\right(n)-Y_i(n-1\left)\right)+Y_i(n-1)$

In formula, T gets 0.2, τ and gets 0.1, n=1, and 2 ..., i=x, y, z.After this by Y _in (), calculates gyro and estimates attitude q _g,j(n) (j=0,1,2,3), accounting equation is as follows:

$\left(\begin{array}{c}{q}_{G,0}\left(n\right)\\ q_{G,1}\left(n\right)\\ q_{G,2}\left(n\right)\\ q_{G,3}\left(n\right)\end{array}\right)=\left(\begin{array}{c}{q}_{G,0}(n-1)+T\frac{Y_z(n-1)}{2}{q}_{G,0}(n-1)-T\frac{Y_y(n-1)}{2}{q}_{G,2}(n-1)+T\frac{Y_z(n-1)}{2}{q}_{G,3}(n-1)\\ -T\frac{Y_z(n-1)}{2}{q}_{G,0}(n-1)+q_{G,1}+q_{G,1}(n-1)+T\frac{Y_x(n-1)}{2}{q}_{G,2}(n-1)+T\frac{Y_y(n-1)}{2}{q}_{G,3}(n-1)\\ T\frac{Y_y(n-1)}{2}{q}_{G,0}(n-1)-T\frac{Y_x(n-1)}{2}{q}_{G,1}(n-1)+q_{G,2}(n-1)+T\frac{Y_z(n-1)}{2}{q}_{G,3}(n-1)\\ -T\frac{Y_x(n-1)}{2}{q}_{G,0}(n-1)-T\frac{Y_y(n-1)}{2}{q}_{G,1}(n-1)-T\frac{Y_z(n-1)}{2}{q}_{G,2}(n-1)+q_{G,3}(n-1)\end{array}\right)$

As shown in Figure 3, the output sequence of star sensor is designated as q _s,kn (), (k=1,2,3), definition attitude misalignment hypercomplex number is Δ q _e,k(n) (k=0,1,2,3), the observed reading Z of deviation hypercomplex number _l(n), (l=1,2,3):

$Z_l\left(n\right)=q_{S,k}\left(n\right)\⊗q_{G,j}^{-1}\left(n\right)$

With following formula recursion Δ q _e,kthe vector section of (n) and the constant value drift increment of gyro:

$\left[\begin{array}{c}\mathrm{\Δ}{q}_{e,1}\left(n\right)\\ \mathrm{\Δ}{q}_{e,2}\left(n\right)\\ \mathrm{\Δ}{q}_{e,3}\left(n\right)\\ \mathrm{\Δ}{b}_x\left(n\right)\\ \mathrm{\Δ}{b}_y\left(n\right)\\ \mathrm{\Δ}{b}_z\left(n\right)\end{array}\right]=\left[\begin{array}{c}\mathrm{\Δ}{q}_{e,1}(n-1)\\ \mathrm{\Δ}{q}_{e,2}(n-1)\\ \mathrm{\Δ}{q}_{e,3}(n-1)\\ \mathrm{\Δ}{b}_x(n-1)\\ \mathrm{\Δ}{b}_y(n-1)\\ \mathrm{\Δ}{b}_z(n-1)\end{array}\right]+\left[\begin{array}{c}0.1(Z_1\left(n\right)-\mathrm{\Δ}{q}_{e,1}\left(n-1\right))\\ 0.1(Z_2\left(n\right)-\mathrm{\Δ}{q}_{e,2}\left(n-1\right))\\ 0.1(Z_3\left(n\right)-\mathrm{\Δ}{q}_{e,3}\left(n-1\right))\\ 0.01(Z_1\left(n\right)-\mathrm{\Δ}{q}_{e,1}\left(n-1\right))\\ 0.01(Z_2\left(n\right)-\mathrm{\Δ}{q}_{e,2}\left(n-1\right))\\ 0.01(Z_3\left(n\right)-\mathrm{\Δ}{q}_{e,3}\left(n-1\right))\end{array}\right]$

Δ q _e,kn the scalar of () is:

${\mathrm{\Δq}}_{e,0}\left(n\right)=\sqrt{1-{\left({\mathrm{\Δq}}_{e,1}\right(n\left)\right)}^2-{\left({\mathrm{\Δq}}_{e,2}\right(n\left)\right)}^2-{\left({\mathrm{\Δq}}_{e,3}\right(n\left)\right)}^2}$

The attitude of satellite hypercomplex number q then only gone out by gyro and the quick data estimation of star is:

$\left(\begin{array}{c}{q}_0\left(n\right)\\ q_1\left(n\right)\\ q_2\left(n\right)\\ q_3\left(n\right)\end{array}\right)=\left(\begin{array}{c}{q}_{G,3}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)+q_{G,0}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)+q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)+q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)\\ q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)+q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)+q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)-q_{G,0}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)\\ q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)+q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)+q_{G,0}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)-q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)\\ q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)-q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)+q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)-q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)\end{array}\right)$

Gyroscope constant value drift update equation:

$\left[\begin{array}{c}{b}_x\left(n\right)\\ b_y\left(n\right)\\ b_z\left(n\right)\end{array}\right]=\left[\begin{array}{c}{b}_x(n-1)\\ b_y(n-1)\\ b_z(n-1)\end{array}\right]+\left[\begin{array}{c}\mathrm{\Δ}{b}_x\left(n\right)\\ \mathrm{\Δ}{b}_y\left(n\right)\\ \mathrm{\Δ}{b}_z\left(n\right)\end{array}\right]$

Convert attitude quaternion to attitude angle by 312 orders, have:

θ _X＝arcsin(2(q ₁q ₂+q ₀q ₃))

${\mathrm{\θ}}_Y=arctan\left(\frac{2(q_1{q}_3-q_0{q}_2)}{-q_0^2-q_1^2+q_2^2+q_3^2}\right)$

${\mathrm{\θ}}_Z=arctan\left(\frac{2(q_2{q}_3-q_0{q}_1)}{-q_0^2+q_1^2-q_2^2+q_3^2}\right)$

As shown in Figure 4, estimated value and upper one estimated value of clapping wideband attitude information of inertial posture measuring system ask deviation, are designated as Δ θ _m(n), wherein m=X, Y, Z, n=1,2,3 ..., then correction signal following formula is estimated:

θ _Cm(n)＝0.5662Δθ _m(n-1)+0.09154Δθ _m(n-2)-0.1261Δθ _m(n-3)-0.02205Δθ _m(n-4)+3.39θ _Cm(n-1)-4.336θ _Cm(n-2)+2.479θ _Cm(n-3)

In formula, the meaning of i is the same, n=5 during recursion, and 6,7 ..., before recursion, the moment needs the amount of assignment all to fix tax 0.1, and recursion certain hour can be restrained.

As shown in Figure 5, the attitude matrix of current time is designated as C _a, the output of angular displacement sensor is designated as θ _hmn (), wherein the meaning of m is with the same.C _aform be:

$C_A\left(n\right)=\left(\begin{array}{ccc}{C}_{A11}\left(n\right)& C_{A12}\left(n\right)& C_{A13}\left(n\right)\\ C_{A21}\left(n\right)& C_{A22}\left(n\right)& C_{A23}\left(n\right)\\ C_{A31}\left(n\right)& C_{A32}\left(n\right)& C_{A33}\left(n\right)\end{array}\right)$

Then the accounting equation of attitude matrix is as follows:

C _A11(n)＝C _A11(n-1)+

C _A21(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A31(n-1)×(θ _HY(n)-θ _HY(n-1)+θ _CY(n)-θ _CY(n-1))

C _A12(n)＝C _A12(n-1)+

C _A22(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A32(n-1)×(θ _HY(n)-θ _HY(n-1)+θ _CY(n)-θ _CY(n-1))

C _A13(n)＝C _A13(n-1)+

C _A23(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A33(n-1)×(θ _HY(n)-θ _HY(n-1)+θ _CY(n)-θ _CY(n-1))

C _A21(n)＝C _A11(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A21(n-1)+

C _A31(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CX(n)+θ _CX(n-1))

C _A22(n)＝C _A12(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+

C _A22(n-1)+C _A32(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CX(n)+θ _CX(n-1))

C _A23(n)＝C _A13(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+

C _A23(n-1)+C _A33(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CX(n)+θ _CX(n-1))

C _A31＝C _A11(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CY(n)+θ _CY(n-1))+

C _A21(n-1)×(θ _HX(n)-θ _HX(n-1)+θ _CX(n)-θ _CX(n-1)+C _A31(N-1))

C _A32＝C _A12(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CY(n)+θ _CY(n-1))+

C _A22(n-1)×(θ _HX(n)-θ _HX(n-1)+θ _CX(n)-θ _CX(n-1)+C _A32(N-1))

C _A33＝C _A13(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CY(n)+θ _CY(n-1))+

C _A23(n-1)×(θ _HX(n)-θ _HX(n-1)+θ _CX(n)-θ _CX(n-1)+C _A33(N-1))

When attitude angle is a small amount of, calculate attitude angle by 312 turns of sequences as follows:

${\mathrm{\θ}}_X=\frac{1}{2}(C_{A23}-C_{A32})$

${\mathrm{\θ}}_Y=\frac{1}{2}(C_{A31}-C_{A13})$

${\mathrm{\θ}}_Z=\frac{1}{2}(C_{A12}-C_{A21})$

The invention provides a kind of satellite high-precision attitude determination method measured based on wideband, make three axis stabilized satellite in high precision high stability degree controls, rely on the conventional inertia such as gyro, star sensor attitude measurement sensor, the low frequency attitude information of satellite is estimated, rely on angular displacement sensor, the high frequency attitude information of satellite is measured.On the basis of low frequency attitude prediction, based on the feedback principle of closed-loop control, generate attitude correction signal, in conjunction with the high frequency attitude information that angular displacement sensor is measured, the wideband high-precision attitude realizing satellite is determined.Compared with prior art, its beneficial effect is: the limitation breaching inertial posture measuring system, and the bandwidth of satellite attitude measurement is improved; By multi-sensor information fusion method, improve the precision of attitude and heading reference system.

Above specific embodiments of the invention are described.It is to be appreciated that the present invention is not limited to above-mentioned particular implementation, those skilled in the art can make various distortion or amendment within the scope of the claims, and this does not affect flesh and blood of the present invention.

Claims (10)

1., based on the satellite high-precision attitude determination method that wideband is measured, it is characterized in that, comprise the steps:

Complete and the output of gyro, star sensor, angular displacement sensor is sampled, and pre-service is carried out to the gyro data obtained of sampling, eliminate the high frequency noise that gyro exports;

Carry out low-pass filtering to gyro data, clap the estimated value of inertial measurement cluster according to upper one, attitude information estimated by the current gyro of recursion;

Calculate the Attitude estimation value of current bat inertial measurement cluster;

After obtaining the Attitude estimation value of inertial measurement cluster, calculate attitude correction signal;

After obtaining attitude correction signal, in conjunction with the measured value of angular displacement sensor, calculate attitude matrix, and then calculate attitude angle.

2. the satellite high-precision attitude determination method measured based on wideband according to claim 1, it is characterized in that, the method that attitude information estimated by the current gyro of recursion is: the gyro data sequence x establishing collection _in (), after deduction gyroscopic drift b, the sequence obtained is X _in (), the sequence exported afterwards is after filtering Y _n, filtering recurrence equation is:

$Y_i\left(n\right)=\frac{T}{\mathrm{\τ}}\left(X_i\right(n)-Y_i(n-1\left)\right)+Y_i(n-1)$

In formula, T gets 0.2, τ and gets 0.1, n=1, and 2 ..., i=x, y, z, after this by Y _in (), calculates gyro and estimates attitude q _g,j(n) j=0,1,2,3, accounting equation is as follows:

$\left(\begin{array}{c}{q}_{G,0}\left(n\right)\\ q_{G,1}\left(n\right)\\ q_{G,2}\left(n\right)\\ q_{G,3}\left(n\right)\end{array}\right)=\left(\begin{array}{c}{q}_{G,0}(n-1)+T\frac{Y_z(n-1)}{2}{q}_{G,0}(n-1)-T\frac{Y_y(n-1)}{2}{q}_{G,2}(n-1)+T\frac{Y_z(n-1)}{2}{q}_{G,3}(n-1)\\ -T\frac{Y_z(n-1)}{2}{q}_{G,0}(n-1)+q_{G,1}+q_{G,1}(n-1)+T\frac{Y_x(n-1)}{2}{q}_{G,2}(n-1)+T\frac{Y_y(n-1)}{2}{q}_{G,3}(n-1)\\ T\frac{Y_y(n-1)}{2}{q}_{G,0}(n-1)-T\frac{Y_x(n-1)}{2}{q}_{G,1}(n-1)+q_{G,2}(n-1)+T\frac{Y_z(n-1)}{2}{q}_{G,3}(n-1)\\ -T\frac{Y_x(n-1)}{2}{q}_{G,0}(n-1)-T\frac{Y_y(n-1)}{2}{q}_{G,1}(n-1)-T\frac{Y_z(n-1)}{2}{q}_{G,2}(n-1)+q_{G,3}(n-1)\end{array}\right).$

3. the satellite high-precision attitude determination method measured based on wideband according to claim 1, it is characterized in that, the method calculating the Attitude estimation value of current bat inertial measurement cluster is: the output of current bat star sensor and current gyro are estimated deviation between attitude as measured value, the attitude quaternion that the inertial measurement cluster calculating star sensor and gyro composition exports, calculates attitude angle by attitude quaternion.

4. the satellite high-precision attitude determination method measured based on wideband according to claim 3, it is characterized in that, method attitude quaternion being calculated attitude angle is: establish the output sequence of star sensor to be designated as q _s,k(n), k=0,1,2,3, definition attitude misalignment hypercomplex number is Δ q _e,k(n), k=0,1,2,3, the observed reading Z of deviation hypercomplex number _l(n), l=0,1,2,3:

$Z_l\left(n\right)=q_{S,k}\left(n\right)\⊗q_{G,j}^{-1}\left(n\right)$

With following formula recursion Δ q _e,kthe vector section of (n) and the constant value drift increment of gyro:

$\left[\begin{array}{c}\mathrm{\Δ}{q}_{e,1}\left(n\right)\\ \mathrm{\Δ}{q}_{e,2}\left(n\right)\\ \mathrm{\Δ}{q}_{e,3}\left(n\right)\\ \mathrm{\Δ}{b}_x\left(n\right)\\ \mathrm{\Δ}{b}_y\left(n\right)\\ \mathrm{\Δ}{b}_z\left(n\right)\end{array}\right]=\left[\begin{array}{c}\mathrm{\Δ}{q}_{e,1}(n-1)\\ \mathrm{\Δ}{q}_{e,2}(n-1)\\ \mathrm{\Δ}{q}_{e,3}(n-1)\\ \mathrm{\Δ}{b}_x(n-1)\\ \mathrm{\Δ}{b}_y(n-1)\\ \mathrm{\Δ}{b}_z(n-1)\end{array}\right]+\left[\begin{array}{c}0.1(Z_1\left(n\right)-\mathrm{\Δ}{q}_{e,1}\left(n-1\right))\\ 0.1(Z_2\left(n\right)-\mathrm{\Δ}{q}_{e,2}\left(n-1\right))\\ 0.1(Z_3\left(n\right)-\mathrm{\Δ}{q}_{e,3}\left(n-1\right))\\ 0.01(Z_1\left(n\right)-\mathrm{\Δ}{q}_{e,1}\left(n-1\right))\\ 0.01(Z_2\left(n\right)-\mathrm{\Δ}{q}_{e,2}\left(n-1\right))\\ 0.01(Z_3\left(n\right)-\mathrm{\Δ}{q}_{e,3}\left(n-1\right))\end{array}\right]$

Δ q _e,kn the scalar of () is:

${\mathrm{\Δq}}_{e,0}\left(n\right)=\sqrt{1-{\left({\mathrm{\Δq}}_{e,1}\right(n\left)\right)}^2-{\left({\mathrm{\Δq}}_{e,2}\right(n\left)\right)}^2-{\left({\mathrm{\Δq}}_{e,3}\right(n\left)\right)}^2}$

The attitude of satellite hypercomplex number q then only gone out by gyro and the quick data estimation of star is:

$\left(\begin{array}{c}{q}_0\left(n\right)\\ q_1\left(n\right)\\ q_2\left(n\right)\\ q_3\left(n\right)\end{array}\right)=\left(\begin{array}{c}{q}_{G,3}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)+q_{G,0}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)+q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)+q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)\\ q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)+q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)+q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)-q_{G,0}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)\\ q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)+q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)+q_{G,0}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)-q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)\\ q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,3}\left(n\right)-q_{G,3}\left(n\right){\mathrm{\Δq}}_{e,0}\left(n\right)+q_{G,1}\left(n\right){\mathrm{\Δq}}_{e,1}\left(n\right)-q_{G,2}\left(n\right){\mathrm{\Δq}}_{e,2}\left(n\right)\end{array}\right)$

Gyroscope constant value drift update equation:

$\left[\begin{array}{c}{b}_x\left(n\right)\\ b_y\left(n\right)\\ b_z\left(n\right)\end{array}\right]=\left[\begin{array}{c}{b}_x(n-1)\\ b_y(n-1)\\ b_z(n-1)\end{array}\right]+\left[\begin{array}{c}\mathrm{\Δ}{b}_x\left(n\right)\\ \mathrm{\Δ}{b}_y\left(n\right)\\ \mathrm{\Δ}{b}_z\left(n\right)\end{array}\right]$

Convert attitude quaternion to attitude angle:

θ _X＝arcsin(2(q ₁q ₂+q ₀q ₃))

${\mathrm{\θ}}_Y=arctan\left(\frac{2(q_1{q}_3-q_0{q}_2)}{-q_0^2-q_1^2+q_2^2+q_3^2}\right).$

${\mathrm{\θ}}_Z=arctan\left(\frac{2(q_2{q}_3-q_0{q}_1)}{-q_0^2+q_1^2-q_2^2+q_3^2}\right).$

5. the satellite high-precision attitude determination method measured based on wideband according to claim 3, it is characterized in that, the method resolving attitude correction signal is: the Attitude estimation value of inertial measurement cluster and upper one are clapped between wideband Attitude estimation value and ask deviation, by this deviation signal successively by low-pass filter, controller, produce the attitude correction signal with low frequency characteristic.

6. the satellite high-precision attitude determination method measured based on wideband according to claim 5, it is characterized in that, described low-pass filter is 3 rank Butterworth filters, and described controller is scale-up factor is 0.1, integral coefficient is 10, differential coefficient is the controller of 0.01.

7. the satellite high-precision attitude determination method measured based on wideband according to claim 5, is characterized in that, the Attitude estimation value of inertial attitude assembly and upper one estimated value of clapping wideband attitude information ask the method for deviation to be: deviation is designated as Δ θ _m(n), wherein m=X, Y, Z, n=1,2,3 ..., correction signal following formula is estimated:

θ _Cm(n)＝0.5662Δθ _m(n-1)+0.09154Δθ _m(n-2)-0.1261Δθ _m(n-3)-0.02205Δθ _m(n-4)+3.39θ _Cm(n-1)-4.336θ _Cm(n-2)+2.479θ _Cm(n-3)

Wherein i=x, y, z, n=5 during recursion, 6,7 ..., before recursion, the moment needs the amount of assignment all to fix tax 0.1, and recursion certain hour is restrained.

8. the satellite high-precision attitude determination method measured based on wideband according to claim 5, it is characterized in that, with the high-frequency signal of the attitude correction signal and angular displacement sensor with low frequency characteristic for inputting, according to the metastatic rule of attitude matrix, recursion goes out the attitude matrix of current bat, according to attitude matrix, calculate attitude angle.

9. the satellite high-precision attitude determination method measured based on wideband according to claim 8, it is characterized in that, according to attitude matrix, the method calculating attitude angle is: the attitude matrix of current time is designated as C _a, the output of angular displacement sensor is designated as θ _hm(n), wherein m=X, Y, Z, n=1,2,3 ..., C _aform be:

$C_A\left(n\right)=\left(\begin{array}{ccc}{C}_{A11}\left(n\right)& C_{A12}\left(n\right)& C_{A13}\left(n\right)\\ C_{A21}\left(n\right)& C_{A22}\left(n\right)& C_{A23}\left(n\right)\\ C_{A31}\left(n\right)& C_{A32}\left(n\right)& C_{A33}\left(n\right)\end{array}\right)$

Then the accounting equation of attitude matrix is as follows:

C _A11(n)＝C _A11(n-1)+

C _A21(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A31(n-1)×(θ _HY(n)-θ _HY(n-1)+θ _CY(n)-θ _CY(n-1))

C _A12(n)＝C _A12(n-1)+

C _A22(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A32(n-1)×(θ _HY(n)-θ _HY(n-1)+θ _CY(n)-θ _CY(n-1))C _A13(n)＝C _A13(n-1)+

C _A23(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A33(n-1)×(θ _HY(n)-θ _HY(n-1)+θ _CY(n)-θ _CY(n-1))

C _A21(n)＝C _A11(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+C _A21(n-1)+

C _A31(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CX(n)+θ _CX(n-1))

C _A22(n)＝C _A12(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+

C _A22(n-1)+C _A32(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CX(n)+θ _CX(n-1))

C _A23(n)＝C _A13(n-1)×(θ _HZ(n)-θ _HZ(n-1)+θ _CZ(n)-θ _CZ(n-1))+

C _A23(n-1)+C _A33(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CX(n)+θ _CX(n-1))

C _A31＝C _A11(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CY(n)+θ _CY(n-1))+

C _A21(n-1)×(θ _HX(n)-θ _HX(n-1)+θ _CX(n)-θ _CX(n-1)+C _A31(N-1))

C _A32＝C _A12(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CY(n)+θ _CY(n-1))+

C _A22(n-1)×(θ _HX(n)-θ _HX(n-1)+θ _CX(n)-θ _CX(n-1)+C _A32(N-1))

C _A33＝C _A13(n-1)×(-θ _HY(n)+θ _HY(n-1)-θ _CY(n)+θ _CY(n-1))+

C _A23(n-1)×(θ _HX(n)-θ _HX(n-1)+θ _CX(n)-θ _CX(n-1)+C _A33(N-1))

When attitude angle is a small amount of, calculate attitude angle as follows:

${\mathrm{\θ}}_X=\frac{1}{2}(C_{A23}-C_{A32})$

${\mathrm{\θ}}_Y=\frac{1}{2}(C_{A31}-C_{A13})$

${\mathrm{\θ}}_Z=\frac{1}{2}(C_{A12}-C_{A21})$

10. based on the satellite high-precision attitude and heading reference system that wideband is measured, for completing arbitrary described satellite high-precision attitude determination method measured based on wideband in claim 1 to 9.