CN102175260A - Error correction method of autonomous navigation system - Google Patents

Abstract

The invention discloses an error correction method of an autonomous navigation system. In the autonomous navigation system based on an ultraviolet earth sensor and a star sensor, the ultraviolet earth sensor is used for measuring a geocentric vector in a satellite local system, and the star sensor is used for measuring an attitude transformation matrix of the satellite local system. Since the sensors inevitably have relative installation error, and a filter cannot eliminate the system error, the navigation accuracy becomes poor. In the method provided by the invention, the relative installation error of the ultraviolet earth sensor and the star sensor is modeled, and the installation error is expanded into a state variable so as to perform navigation filtering. Assuming that the satellite can obtain high-accuracy position measurement information by a GPS receiver or ground orbit determination within certain period of time, the system error can be estimated and corrected in time by filtering. The method provided by the invention has simple operation and can obviously improve the navigation accuracy.

Description

A kind of autonomous navigation system error calibration method

Technical field

The invention belongs to spacecraft independent navigation field, relate to the bearing calibration of a kind of spacecraft Navigation system error.

Background technology

Adopting imaging-type earth sensor and star sensor to carry out autonomous navigation of satellite is a kind of typical astronomical navigation method, and the independent navigation precision is subjected to influence of various factors such as sensor measuring accuracy, the model accuracy of describing the measured target correlation properties, navigation algorithm.Compare with stochastic error, the systematic error of navigation sensor (as measured deviation, alignment error, model error etc.) wants big many to the influence of navigation accuracy.This is because Navigation Filter can suppress the measurement noise, but some often is worth or has the measuring error of certain rule powerless.Therefore be to improve one of gordian technique of autonomous navigation of satellite system navigation accuracy to demarcating with compensation of systematic errors such as normal value measured deviation, sensor alignment error at rail.

Zhang Chunqing, Li Yong etc. in the paper of delivering on volume the 2nd phase aerospace journal March in 2006 the 27th " the self calibration filtering that the satellite autonomous Orbit is determined " based on the observability theoretical proof of linear time varying system system deviation during for unknown normal value, the state of autonomous navigation system and normal value deviation all are considerable, further with system deviation as state variable, come the corrective system deviation by self-calibrating filter, how the alignment error of sensor is not proofreaied and correct but wherein relate to.

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 autonomous navigation system error calibration method has been proposed, can solve the excessive problem of navigation error that the sensor alignment error causes, improve the navigation accuracy of system by the On-line Estimation installation deviation.

Technical solution of the present invention is: the autonomous navigation system error calibration method, and step is as follows:

(1) in the autonomous navigation system based on ultraviolet earth sensor and star sensor, the ultraviolet earth sensor is observed the earth, obtains the earth's core vector value under the satellite body system

Star sensor by importance in star map recognition, obtains the attitude transition matrix that inertia is tied to satellite body system to star observation

Wherein the inertial system initial point is in ground ball center, and the X-axis positive dirction is pointed to the first point of Aries along the intersection of earth equatorial plane and ecliptic plane, and the Z axle is the normal of equatorial plane, positive dirction directed north direction, and Y-axis becomes right-handed system with X, Z axle; The initial point of satellite body system points to the earth's core at the satellite barycenter for absolute orientation three axis stabilized satellite Z axle, and X-axis is vertical with the Z axle and point to the satellite velocities direction, and Y-axis becomes right-handed system with X, Z axle;

(2) the alignment error angle of establishing earth sensor and star sensor is Φ=[φ _xφ _yφ _z], then inertia is tied to the attitude transition matrix measured value of satellite body system

For

$C_i^{b^{\′}}=\{I_{3\×3}-[\mathrm{\Φ}\×\left]\right\}{C}_i^b,$

In the formula

For inertia is tied to the attitude transition matrix theoretical value that satellite body is, [Φ *] is the antisymmetric matrix that vectorial Φ constitutes,

(3) according to the result of step (1) and step (2), obtain the earth's core direction vector under the inertial system

The measurement equation be

${\hat{u}}^I={\left(C_i^{b^{\′}}\right)}^T{\hat{u}}^B={\left(C_i^b\right)}^T\{I_{3\×3}+[\mathrm{\Φ}\×\left]\right\}{\hat{u}}^B$

(4) obtain the high precision position metrical information by GPS receiver or ground orbit determination, with three alignment error angles with satellite position vector, satellite velocity vector as state variable, the filtering of navigating is resolved, and alignment error is estimated in real time and proofreaied and correct.

The present invention's advantage compared with prior art is: the inventive method is at the characteristics of " ultraviolet earth sensor+star sensor " autonomous navigation system, relative alignment error to ultraviolet earth sensor and star sensor has been carried out modeling, alignment error is extended for state variable, carries out UKF (Unscented Kalman Filter) navigation filtering with satellite position vector, satellite velocity vector and resolve.Supposing in satellite a period of time can be by GPS receiver or ground orbit determination acquisition high precision position metrical information, and then wave filter can be estimated in real time and proofreaies and correct alignment error, thereby can improve navigation accuracy.

Description of drawings

Fig. 1 is the process flow diagram of the inventive method;

Fig. 2 is the navigation results synoptic diagram of the sensor alignment error not being proofreaied and correct in the embodiment of the invention;

Fig. 3 is the navigation results synoptic diagram that adopts in the embodiment of the invention after the inventive method is proofreaied and correct the sensor alignment error.

Embodiment

As shown in Figure 1, be the FB(flow block) of the inventive method, key step is as follows:

(1) definition coordinate system, inertial system initial point are in ground ball center, and the X-axis positive dirction is pointed to the first point of Aries along the intersection of earth equatorial plane and ecliptic plane, and the Z axle is the normal of equatorial plane, positive dirction directed north direction, and Y-axis becomes right-handed system with X, Z axle.Satellite body be initial point at the satellite barycenter, point to the earth's core for absolute orientation three axis stabilized satellite Z axle, X-axis is vertical with the Z axle and point to the satellite velocities direction, Y-axis becomes right-handed system with X, Z axle;

(2) in " ultraviolet earth sensor+star sensor " autonomous navigation system, the ultraviolet earth sensor is observed the earth, obtains the earth's core vector value under the satellite body coordinate system

Star sensor obtains inertia attitude quaternion q=[q to star observation by importance in star map recognition ₁q ₂q ₃q ₄], calculate the attitude transition matrix that inertia is tied to satellite body system

$C_i^{b^{\′}}=\left[\begin{array}{ccc}{q}_1^2-q_2^2-q_3^2+q_4^2& 2(q_1{q}_2+q_3{q}_4)& 2(q_1{q}_3-q_2{q}_4)\\ 2(q_1{q}_2-q_3{q}_4)& -q_1^2+q_2^2-q_3^2+q_4^2& 2(q_2{q}_3+q_1{q}_4)\\ 2(q_1{q}_3+q_2{q}_4)& 2(q_2{q}_3-q_1{q}_4)& -q_1^2-q_2^2+q_3^2+q_4^2\end{array}\right]$

(3) consider that unavoidably there are installation deviation in earth sensor and star sensor, establishing the alignment error angle is Φ=[φ _xφ _yφ _z], φ _x, φ _y, φ _zBe low-angle, then have

$C_i^{b^{\′}}=C_b^{b^{\′}}{C}_i^b=\{I_{3\×3}-[\mathrm{\Φ}\×\left]\right\}{C}_i^b$

In the formula

For inertia is tied to the real attitude transition matrix that satellite body is, and

Be the attitude transition matrix that measures.The antisymmetric matrix that [Φ *] expression is made of vectorial Φ,

$[\mathrm{\Φ}\×]=\left[\begin{array}{ccc}0& -{\mathrm{\φ}}_z& {\mathrm{\φ}}_y\\ {\mathrm{\φ}}_z& 0& -{\mathrm{\φ}}_x\\ -{\mathrm{\φ}}_y& {\mathrm{\φ}}_x& 0\end{array}\right]$

(4) three components with alignment error angle Φ are extended for the filter state variable, promptly

R=[x y z wherein] be the position vector of satellite, Velocity for satellite.

Except that the gravitation item of center, only consider J in the satellite orbit kinetic model ₂The item perturbation, its component form is

In the formula

μ=GE is the terrestrial gravitation constant, and Re is an earth radius, J ₂Be humorous coefficient of second order band, w _x, w _y, w _zBe system noise.

(5) measuring equation is

${\hat{u}}^I={\left(C_i^{b^{\′}}\right)}^T{\hat{u}}^B={\left(C_b^{b^{\′}}{C}_i^b\right)}^T{\hat{u}}^B={\left(C_i^b\right)}^T{\left(C_b^{b^{\′}}\right)}^T{\hat{u}}^B$

$={\left(C_i^b\right)}^T\{I_{3\×3}+[\mathrm{\Φ}\×\left]\right\}{\hat{u}}^B$

Because measured value is to the relatively difficulty of calculating reciprocal partially of quantity of state, therefore the technology of the present invention adopts UKF (the Unscented Kalman Filter) filtering algorithm that need not differentiate, and specific algorithm can " utilize the spacecraft autonomous navigation method research of UKF " with reference to Liu Wei, the Yang Bo paper that the 23rd volume the 3rd phase Aerospace Control was delivered June in 2005.

(6) can be in supposition satellite a period of time by GPS receiver or ground orbit determination acquisition high precision position metrical information, then wave filter can be estimated in real time and proofreaies and correct alignment error.

Embodiment

For the clearer advantage that shows this method, carry out mathematical simulation at this, simulated conditions: 500 kilometers of satellite altitudes, inclination angle 97.4 degree, the earth's core orientation measurement error 0.02 degree supposes that ultraviolet sensors is respectively 50 ", 40 ", 30 along three coordinate axis alignment error angles ".

Fig. 1 is the simulation result of alignment error not being proofreaied and correct, and Fig. 2 proofreaies and correct post-layout simulation results exhibit figure for adopting the inventive method to alignment error.As can be seen from the figure, positional precision is not 1620.059 meters during the update the system error, and positional precision is 125.525 meters after the update the system error, therefore adopt the inventive method that the alignment error angle is estimated to proofread and correct after, can improve navigation accuracy greatly.

The content that is not described in detail in the instructions of the present invention belongs to those skilled in the art's known technology.

Claims (1)

1. autonomous navigation system error calibration method is characterized in that step is as follows:

(1) in the autonomous navigation system based on ultraviolet earth sensor and star sensor, the ultraviolet earth sensor is observed the earth, obtains the earth's core vector value under the satellite body system Star sensor by importance in star map recognition, obtains the attitude transition matrix that inertia is tied to satellite body system to star observation

Wherein the inertial system initial point is in ground ball center, and the X-axis positive dirction is pointed to the first point of Aries along the intersection of earth equatorial plane and ecliptic plane, and the Z axle is the normal of equatorial plane, positive dirction directed north direction, and Y-axis becomes right-handed system with X, Z axle; The initial point of satellite body system points to the earth's core at the satellite barycenter for absolute orientation three axis stabilized satellite Z axle, and X-axis is vertical with the Z axle and point to the satellite velocities direction, and Y-axis becomes right-handed system with X, Z axle;

(2) the alignment error angle of establishing earth sensor and star sensor is Φ=[φ _xφ _yφ _z], then inertia is tied to the attitude transition matrix measured value of satellite body system

For

$C_i^{b^{\′}}=\{I_{3\×3}-[\mathrm{\Φ}\×\left]\right\}{C}_i^b,$

In the formula

For inertia is tied to the attitude transition matrix theoretical value that satellite body is, [Φ *] is the antisymmetric matrix that vectorial Φ constitutes,

(3) according to the result of step (1) and step (2), obtain the earth's core direction vector under the inertial system

The measurement equation be

${\hat{u}}^I={\left(C_i^{b^{\′}}\right)}^T{\hat{u}}^B={\left(C_i^b\right)}^T\{I_{3\×3}+[\mathrm{\Φ}\×\left]\right\}{\hat{u}}^B$

(4) obtain the high precision position metrical information by GPS receiver or ground orbit determination, with three alignment error angles with satellite position vector, satellite velocity vector as state variable, the filtering of navigating is resolved, and alignment error is estimated in real time and proofreaied and correct.

Images