CN101788296A

CN101788296A - SINS/CNS deep integrated navigation system and realization method thereof

Info

Publication number: CN101788296A
Application number: CN201010101429A
Authority: CN
Inventors: 王新龙; 吴小娟
Original assignee: Beihang University
Current assignee: Beihang University
Priority date: 2010-01-26
Filing date: 2010-01-26
Publication date: 2010-07-28
Anticipated expiration: 2030-01-26
Also published as: CN101788296B

Abstract

The invention discloses a SINS/CNS deep integrated navigation system and its implementation method. The navigation system includes a strapdown inertial navigation system, an astronomical navigation system, an integrated navigation filter, and an inertial navigation attitude measurement information construction unit; the implementation method includes step 1 , large field of view star sensor assisted strapdown inertial navigation system to obtain a high-precision mathematical level benchmark; step two, CNS positioning based on the mathematical level benchmark; step three, establish the state model and measurement model of the SINS/CNS deep combined system; step Four: integrated navigation system information fusion; step five, SINS and CNS assist each other to realize high-precision positioning; the present invention utilizes star sensor high-precision attitude information to assist SINS, and obtains high-precision SINS strapdown matrix as a mathematical level reference for CNS positioning, On this basis, the SINS is fully corrected by using the position and attitude of the CNS, realizing the deep combination of SINS/CNS, and finally achieving high-precision positioning and navigation.

Description

A kind of SINS/CNS deep integrated navigation system and its implementation

Technical field

The present invention relates to a kind of SINS/CNS deep integrated navigation system and its implementation, belong to the integrated navigation technical field.

Background technology

Celestial navigation system (CNS) good concealment, independence be strong, carrier positions and high-accuracy posture information can be provided, but output information is discontinuous, and can be subjected to the influence of external environment in some cases, when using separately, be difficult to satisfy high precision, high performance navigator fix requirement.

The bearing accuracy of tradition celestial navigation system depends primarily on the measuring accuracy of horizontal reference precision and heavenly body sensor.Because heavenly body sensor can more easily reach 1 rad observation precision at present, the precision of Horizon becomes the main factor that influences the celestial navigation precision.Inertial platform horizontal reference error is subjected to the error effect of its core component gyro at present, and list is extremely difficult from the precision that the improvement instrument design improves the inertia horizontal reference, and cost performance reduces; Direct responsive Horizon precision such as employing infrared horizon are lower; Employing based on starlight refraction the quadratic method precision is higher sensitively indirectly, but the uncertainty of atmospheric refraction model has directly influenced the precision and the reliability of horizontal reference.High level of accuracy benchmark technology has become the celestial navigation of realization high precision and has needed one of gordian technique of capturing badly.

And strapdown inertial navigation system (SINS) have navigational parameter comprehensively, output in time continuously, outstanding advantage such as good concealment, antijamming capability be strong, but because the existence of initial alignment error and inertia device error, the SINS error increases in time, be difficult to work alone for a long time, need to increase other metric informations, utilize the method for combination to improve the integrated navigation performance.Both have the characteristics of mutual supplement with each other's advantages SINS, CNS, and both combinations can be given full play to separately advantage, and along with the intensification of combined level, the overall performance of SINS/CNS combined system will be far superior to each autonomous system.

20th century, because detecting, celestial body is only limited to the detection of under the difference moment, different attitude, finishing different celestial bodies, astronomy/inertia combination can only be adopted system-level simple combination navigation mode, celestial navigation system receives the position and the attitude information of inertial navigation system output, regularly the drift of inertial navigation system is proofreaied and correct.This simple combination pattern can damping inertial navigation site error disperse, but its navigation accuracy is subjected to the restriction of the horizontal attitude precision of inertial navigation, and the horizontal attitude precision of inertial navigation depends on the precision of inertial sensor error, therefore this combined method corrective action and little.

To the middle and later periods nineties 20th century, because the development of big visual field celestial body Fast Detection Technique, celestial navigation system can be finished many stars synchronous detection in a certain moment, and the attitude information of under the outside initial information prerequisite of (comprising horizontal reference), determining carrier coordinate relative inertness coordinate (identical with the attitude information of gyro output).Therefore, be that the astronomy/inertia optimum combination pattern of core all can realize on theoretical and engineering fully with compensation inertial navigation gyroscopic drift, but this integrated mode is merely able to compensate the navigation error that causes because of gyroscopic drift in the inertial navigation.

At present, celestial navigation system adopts starlight to reflect the flat method of indirect geodetic and obtains the high-precision independent horizontal reference, can carry out high-precision independent and determine position (three-dimensional coordinate), course and attitude, thereby inertial navigation system is realized thoroughly optimum comprehensively the correction.This astronomy/inertia optimum combination pattern not only can compensate the random drift of gyro, and can the compensated acceleration meter etc. the error that causes of other factors, but the method can only be used on the high-altitude sail body more than the 30km at present, can not realize FR application.

In sum, utilize many stars synchro measure and instantaneous definite carrier inertia attitude principle, solve this gordian technique of celestial navigation high level of accuracy benchmark, and develop high precision astronomy applied widely/inertia deep integrated navigation pattern to replace traditional system-level, rough astronomy/inertia integrated mode, thereby realizing the hi-Fix navigation, is the main direction of studying of astronomy/inertia best of breed navigational system.

Summary of the invention

The objective of the invention is to determine the deficiency of scheme in order to overcome the existing level benchmark, utilize the characteristics of SINS/CNS combination sensor, a kind of SINS/CNS deep integrated navigation system and its implementation are proposed, this method has made full use of the navigation information of each subsystem, has improved the precision of SINS/CNS integrated navigation system.

The dark combination implementing method of a kind of SINS/CNS specifically may further comprise the steps:

Step 1: the auxiliary SINS of star sensor obtains high-precision mathematics horizontal reference;

Step 2: carry out the CNS location based on the mathematics horizontal reference;

Step 3: set up dark combined system state equation of SINS/CNS and measurement equation;

A. make up the dark combined system state model of SINS/CNS;

B. make up the dark combined system of SINS/CNS and measure model;

Step 4: integrated navigation system information fusion;

The auxiliary mutually hi-Fix of realizing of step 5: SINS and CNS.

A kind of SINS/CNS deep integrated navigation system comprises strapdown inertial navitation system (SINS), celestial navigation system, integrated navigation wave filter, inertial navigation attitude measurement information tectonic element;

Strapdown inertial navigation system comprises inertial measurement cluster and navigation calculation unit; Inertial measurement cluster records angular velocity and the specific force of carrier with respect to inertial space, send the navigation calculation unit to the carrier angular velocity that obtains with than force information, the navigation calculation unit calculates the positional information latitude L of carrier in real time by mechanics layout algorithm according to the inertial measurement cluster information transmitted _IAnd longitude λ _I, speed and attitude; The navigation calculation unit is with the positional information L of carrier simultaneously _I, λ _IBe input in the integrated navigation wave filter, the SINS navigational parameter is input in the inertial navigation attitude measurement information tectonic element, described navigational parameter is the position L of current navigation time t, carrier _I, λ _IAnd attitude, the navigation calculation unit also inputs to celestial navigation system with SINS strapdown matrix as the mathematics horizontal reference and measures the elevation angle computing unit, with the positional information L of carrier _I, λ _IBe input to celestial navigation system analytic Height difference locating module;

Celestial navigation system comprises that big visual field star sensor, many vectors decide the appearance unit, measure the elevation angle computing unit and resolve the difference in height locating module;

Big visual field star sensor synchronization can be observed the starlight Vector Message that obtains three and three above fixed stars, and the observation information that obtains is offered many vectors respectively decide appearance unit and measurement elevation angle computing unit; Many vectors are decided the appearance unit starlight Vector Message that receives are handled, and obtain the attitude information in carrier relative inertness space

And with the carrier inertia attitude information of determining

Be input in the integrated navigation wave filter; Measure the elevation angle computing unit and utilize the starlight Vector Message of star sensor transmission and the mathematics horizontal reference that the navigation calculation unit provides, obtain the fixed star measurement elevation angle H on plane relatively ₀, and the elevation angle measurement information inputed to analytic Height difference locating module; The carrier positions information L that analytic Height difference locating module provides according to the fixed star elevation angle information that measures the transmission of elevation angle computing unit and navigation calculation unit _I, λ _I, obtain the carrier latitude L of astronomical fixation _C, longitude λ _C, and with astronomical fixation L as a result _C, λ _CBe input to the integrated navigation wave filter as measurement information;

Inertial navigation attitude measurement information tectonic element is according to the navigational parameter of navigation calculation unit transmission, obtain inertial navigation attitude measurement information, described inertial navigation attitude measurement information be strapdown inertial navigation system determine be transformed into the direction cosine matrix of carrier coordinate system from equator, the earth's core inertial coordinates system

Inertial navigation attitude measurement information tectonic element is with the inertial navigation attitude measurement information that obtains

Offer the integrated navigation wave filter;

The integrated navigation wave filter is decided the SINS positional information L that appearance unit, analytic Height difference locating module, inertial navigation attitude measurement information tectonic element provide respectively according to navigation calculation unit, many vectors _I, λ _I, CNS attitude measurement information

, CNS positioning result L _C, λ _CWith inertial navigation attitude measurement information

Handle by Kalman filtering, the navigational parameter of strapdown inertial navigation system and the error of inertial measurement cluster are estimated, and it is fed back in the SINS navigation calculation unit, corresponding error is proofreaied and correct and compensated.

The invention has the advantages that:

(1) the present invention utilizes star sensor high-accuracy posture information to assist SINS, and real-time monitored also compensates mathematical platform misalignment and the gyroscopic drift of SINS, from obtaining high-precision SINS strapdown matrix as the mathematics horizontal reference, is used for the CNS location;

(2) the analytic Height difference method based on the mathematics horizontal reference is positioned error modeling, and survey the probabilistic influence of deduction mathematics horizontal reference in the equation, thereby eliminated the correlativity between CNS positioning error and the SINS attitude error in the position margin of error;

(3) utilize big visual field many stars of star sensor synchro measure and instantaneous definite carrier inertia attitude principle, auxiliary SINS obtains high-precision mathematics horizontal reference by star sensor, CNS can provide high precision position and attitude information simultaneously, SINS is realized comprehensively optimum the correction, further improved the precision of mathematics horizontal reference and astronomical fixation precision on this basis;

(4) SINS/CNS deep integrated navigation pattern of the present invention is auxiliary mutually by SINS, CNS, has given full play to the advantage of each subsystem, finally can realize high-precision location navigation.

Description of drawings

Fig. 1 is the structural representation of a kind of SINS/CNS deep integrated navigation system of the present invention;

Fig. 2 is a kind of process flow diagram of SINS/CNS deep integrated navigation system implementation method;

Fig. 3 is the synoptic diagram that the auxiliary SINS of star sensor of the present invention obtains high precision mathematics horizontal reference;

Among the figure:

1-strapdown inertial navigation system 2-celestial navigation system 3-integrated navigation wave filter 4-inertial navigation attitude measurement information

Tectonic element

The many vectors of 101-inertial measurement cluster 102-navigation calculation unit 201-big visual field star sensor 202-are decided the appearance unit

It is fixed that 203-measures elevation angle calculating 204-analytic Height difference

Position, unit module

Embodiment

The present invention is described in further detail below in conjunction with drawings and Examples.

A kind of SINS/CNS deep integrated navigation system of the present invention as shown in Figure 1, comprises strapdown inertial navitation system (SINS) 1, celestial navigation system 2, integrated navigation wave filter 3, inertial navigation attitude measurement information tectonic element 4;

Strapdown inertial navigation system (SINS) 1 comprises inertial measurement cluster (IMU) 101 and navigation calculation unit 102.Inertial measurement cluster 101 (IMU) records angular velocity and the specific force of carrier with respect to inertial space, send navigation calculation unit 102 to the carrier angular velocity that obtains with than force information, navigation calculation unit 102 calculates the positional information latitude L of carrier in real time by mechanics layout algorithm according to inertial measurement cluster 101 information transmitted _IAnd longitude λ _I, speed and attitude.Navigation calculation unit 102 is with the positional information L of carrier _I, λ _IBe input in the integrated navigation wave filter 3, the SINS navigational parameter is input in the inertial navigation attitude measurement information tectonic element 4, described navigational parameter is the position L of current navigation time t, carrier _I, λ _IAnd attitude, navigation calculation unit 102 also inputs to measurement elevation angle computing unit 203 in the celestial navigation system 2 with SINS strapdown matrix as the mathematics horizontal reference, with the positional information L of carrier _I, λ _IBe input to the analytic Height difference locating module 204 in the celestial navigation system;

Celestial navigation system (CNS) 2 comprises that big visual field star sensor 201, many vectors decide appearance unit 202, measure elevation angle computing unit 203 and resolve difference in height locating module 204;

Big visual field star sensor 201 synchronizations can be observed the starlight Vector Message that obtains three and three above fixed stars, and the observation information that obtains is offered many vectors respectively decide appearance unit 202 and measurement elevation angle computing unit 203.Many vectors are decided the starlight Vector Message that 202 pairs of appearance unit receive and are handled, and obtain the attitude information in carrier relative inertness space And with the carrier inertia attitude information of determining

Be input in the integrated navigation wave filter 3.Measure elevation angle computing unit 203 and utilize the starlight Vector Message of star sensor 201 transmission and the mathematics horizontal reference that navigation calculation unit 102 provides, obtain the fixed star measurement elevation angle H on plane relatively ₀, and the elevation angle measurement information inputed to analytic Height difference locating module 204.The carrier positions information L that analytic Height difference locating module 204 provides according to the fixed star elevation angle information that measures 203 transmission of elevation angle computing unit and navigation calculation unit 102 _I, λ _I, obtain the carrier latitude L of astronomical fixation _C, longitude λ _C, and with astronomical fixation L as a result _C, λ _CBe input to integrated navigation wave filter 3 as measurement information;

Inertial navigation attitude measurement information tectonic element 4 is according to the navigational parameter of navigation calculation unit 102 transmission, obtain inertial navigation attitude measurement information, described inertial navigation attitude measurement information be strapdown inertial navigation system 1 determine be transformed into the direction cosine matrix of carrier coordinate system from equator, the earth's core inertial coordinates system

, inertial navigation attitude measurement information tectonic element 4 is with the inertial navigation attitude measurement information that obtains

Offer integrated navigation wave filter 3;

Integrated navigation wave filter 3 is decided the SINS positional information L that appearance unit 202, analytic Height difference locating module 204, inertial navigation attitude measurement information tectonic element 4 provide respectively according to navigation calculation unit 102, many vectors _I, λ _I, CNS attitude measurement information

CNS positioning result L _C, λ _CWith inertial navigation attitude measurement information

Handle by Kalman filtering, the navigational parameter of strapdown inertial navigation system 1 and the error of inertial measurement cluster are estimated, and it is fed back in the SINS navigation calculation unit 102, corresponding error is proofreaied and correct and compensated;

In whole SINS/CNS deep integrated navigation system,, finally realize high-precision location navigation by assisting mutually between strapdown inertial navigation system 1, the celestial navigation system 2.

A kind of SINS/CNS deep integrated navigation system implementation method of the present invention, flow process specifically may further comprise the steps as shown in Figure 2:

Step 1: star sensor 201 auxiliary strapdown inertial navigation systems 1 in big visual field obtain high-precision mathematics horizontal reference;

Owing to utilize the strapdown matrix of SINS can realize that carrier coordinate system arrives the coordinate conversion of platform coordinate system, obtain the expression of measurement vector at platform coordinate system, therefore the strapdown matrix that resolves is equivalent to set up mathematical platform, and the angle of rotation speed of geographic coordinate system is input in the calculation procedure of " mathematical platform ", but the surface level of platform real-time follow-up carrier loca.But the horizontal attitude information of SINS is drifted about in time, directly utilizes the strapdown matrix as horizontal reference, can cause the astronomical fixation error to be dispersed.And star sensor is equivalent to not have the gyro of drift, therefore utilize star sensor observation celestial body orientation to proofread and correct the drift of mathematical platform, the pure SINS mathematics horizontal reference that overcomes long time continuous working improves the precision of mathematics horizontal reference owing to the error that gyroscopic drift etc. causes.

The synoptic diagram of big visual field star sensor 201 auxiliary SINS acquisition high precision mathematics horizontal references as shown in Figure 3, big visual field star sensor 201 obtains the multidimensional starlight Vector Message of three or three above fixed stars in synchronization observation, many then vectors are decided appearance unit 202 pairs of multidimensional starlight Vector Message and are handled, and obtain the direction cosine matrix of the relative the earth's core of measurement coordinate system s equator inertial system I of big visual field star sensor 201

In conjunction with the installation Matrix C of big visual field star sensor 201 on carrier _b ^s, obtaining carrier is the direction cosine matrix of b with respect to equator, the earth's core inertial system I

It is the direction cosine matrix of the relative the earth's core of b equator inertial system I that the navigation information that inertial navigation attitude measurement information tectonic element 4 is exported by navigation calculation unit 102 constructs the definite carrier of SINS

Attitude information with 201 outputs of big visual field star sensor

Be complementary, described navigation information is the locating information L of current navigation time t, SINS _I, λ _IAnd mathematics horizontal reference

Described direction cosine matrix

Concrete computing method be:

Utilize the locating information L of SINS _I, λ _IConstruct the location matrix of SINS

Limit obtains being transformed into the be connected direction cosine matrix C of coordinate system e of the earth from equator, the earth's core inertial system I according to current navigation time t _I ^eStrapdown matrix in conjunction with SINS

And location matrix Obtain:

{\hat{C}}_{I}^{b} = {\hat{C}}_{n}^{b} {\hat{C}}_{e}^{n} C_{I}^{e} = {({\hat{C}}_{b}^{n})}^{T} {\hat{C}}_{e}^{n} C_{I}^{e} - - - (1)

Consider the influence of factors such as alignment error and gyroscopic drift, the SINS mathematical platform is that n ' and navigation coordinate are to have mathematical platform misalignment vector between the n

φ _E, φ _N, φ _UFor east, north, day to misalignment, thereby SINS strapdown matrix

Error relevant with misalignment; And because the latitude error δ L of SINS location _I, longitude error δ λ _IExistence, to be nc do not overlap with the actual n of Department of Geography in the calculating of SINS, and position error vector δ P=[-δ L is arranged _Iδ λ _ICos L _Iδ λ _ISin L _I] ^T, the navigation that must cause SINS to determine is n with respect to the be connected direction cosine matrix of coordinate system e of the earth Position deviation δ L with SINS _I, δ λ _IClose.Then there is following relation:

{\hat{C}}_{e}^{n} = (I - [δP \times]) C_{e}^{n}

Wherein, C _n ^b, C _e ⁿNavigation system, navigation are the be connected direction cosine matrix of coordinate system of the relative earth relatively to be respectively real carrier system.

Ignore the above error term of secondary and secondary, then the definite carrier of SINS is the direction cosine matrix of equator, relative the earth's core inertial system

Can be expressed as

{\hat{C}}_{I}^{b} = C_{I}^{b} + C_{n}^{b} [φ \times] C_{e}^{n} C_{I}^{e} - C_{n}^{b} [δP \times] C_{e}^{n} C_{I}^{e}

Suppose that desirable free from error carrier is that the direction cosine battle array of the relative the earth's core of b equator inertial system I is C _I ^bBecause the measuring accuracy of star sensor is very high, can think that the carrier system of star sensor output is with respect to the direction cosine matrix of equator, the earth's core inertial system Be real direction cosine matrix C _I ^bMeasure white noise acoustic matrix V with star sensor _sStack, that is:

{\tilde{C}}_{I}^{b} = C_{I}^{b} + V_{s} - - - (5)

With the carrier of being determined by SINS, star sensor respectively is the direction cosine matrix of equator, relative the earth's core inertial system

Between difference note make Z _s, then have

Z_{s} = {\hat{C}}_{I}^{b} - {\tilde{C}}_{I}^{b} = C_{n}^{b} [φ \times] C_{e}^{n} C_{I}^{e} - C_{n}^{b} [δP \times] C_{e}^{n} C_{I}^{e} - V_{s} - - - (6)

Because platform misalignment and gyroscope constant value drift have coupled relation, be the direction cosine battle array of equator, relative the earth's core inertial system with star sensor with the carrier that SINS determines respectively by integrated navigation wave filter 3 Carry out information fusion, can estimate SINS mathematical platform misalignment and gyroscopic drift in real time, then SINS mathematical platform misalignment and gyroscopic drift are revised, to improve the mathematics horizontal reference

Precision, concrete grammar is as follows:

Gyroscope among the IMU101 and accelerometer are exported the specific force of carrier in real time

And angular velocity

And send navigation calculation unit 102 to.In navigation calculation unit 102, the ratio force information of carrier

After the mathematics horizontal reference is handled, be directly inputted into the northern bit platform formula inertial reference calculation process of finger and carry out navigation calculation, obtain the position L of carrier _I, λ _I, navigation information such as speed, attitude; Utilize the estimated value of misalignment

Can calculate the attitude error rectification Matrix C _{N '} ⁿ, pass through formula

To the strapdown matrix Carry out the misalignment correction, can improve the mathematics horizontal reference

Precision; And from gyrostatic output

In the error delta of deduction gyroscopic drift in real time ω _Ib ^b, carry out gyroscopic drift error compensation, can access the angular velocity vector information in carrier relative inertness space more accurately

{\tilde{ω}}_{ib}^{b} = {\hat{ω}}_{ib}^{b} - δ ω_{ib}^{b},

And then in conjunction with the mathematics horizontal reference and refer to that it is the angular velocity of equator, relative the earth's core inertial system that northern bit platform formula inertial reference calculation obtains navigating

Calculate the attitude speed of relative navigation system of high-precision carrier system

And the direction of passage cosine matrix differential equation

{\overset{\cdot}{C}}_{b}^{n} {= C}_{b}^{n} Ω_{nb}^{n}

Carry out the strapdown matrix update, can further improve the precision of strapdown matrix (mathematics horizontal reference), wherein Ω _Nb ^bBe angular velocity vector Multiplication cross matrix in carrier coordinate system.

Last strapdown inertial navigation system 1 obtains high-precision strapdown matrix

It is the mathematics horizontal reference.

Measure the high precision mathematics horizontal reference that elevation angle computing unit 203 utilizes the auxiliary SINS of star sensor to obtain

In conjunction with the multidimensional starlight Vector Message that star sensor 201 observations in big visual field obtain, can obtain the elevation angle H of many observation astrologies for ground level ₀, realize the celestial navigation location by analytic Height difference locating module 204 then.Concrete steps are as follows:

A. utilize star sensor 201 auxiliary strapdown inertial navitation system (SINS) 1 to obtain high-precision SINS strapdown matrix

It is transferred to celestial navigation system 2 as the mathematics horizontal reference is used for the CNS location; And navigation calculation unit 102 output positional information L _I, λ _IThe latitude initial value Lat of iteration is provided for analytic Height difference method location _AP, longitude initial value Lon _AP

B. big visual field star sensor 201 synchronizations can be observed the expression of starlight vector in star sensor measurement coordinate system s and equator, the earth's core inertial system I of three or three above fixed stars that obtain.Consider star sensor measurement noise V _i ^sExistence, actual measurement obtains the position vector of fixed star i in star sensor measurement coordinate system s system and is:

{\tilde{X}}_{i}^{s} = X_{i}^{s} + V_{i}^{s} - - - (7)

In the formula, X _i ^sBe fixed star i real position vector in star sensor measurement coordinate system s.

At navigation time t, the auxiliary SINS of star sensor obtains high-precision mathematics horizontal reference

Installation Matrix C in conjunction with star sensor _b ^s, the starlight vector that can obtain fixed star i is expression among the n (ENU Department of Geography) at navigation coordinate

{\tilde{X}}_{i}^{n} = {\tilde{C}}_{b}^{n} C_{s}^{b} {\tilde{X}}_{i}^{s} = {\tilde{C}}_{b}^{n} {(C_{b}^{s})}^{T} {\tilde{X}}_{i}^{s} - - - (8)

With the starlight position vector that calculates With being defined in navigation is celestial body elevation angle among the n

And position angle

Expression can be described as:

{\tilde{X}}_{i}^{n} = {[\begin{matrix} \cos {\tilde{h}}_{i}^{n} \sin {\tilde{A}}_{i}^{n} & \cos {\tilde{h}}_{i}^{n} \cos {\tilde{A}}_{i}^{n} & \sin {\tilde{h}}_{i}^{n} \end{matrix}]}^{T} = {[\begin{matrix} p_{1} & p_{2} & p_{3} \end{matrix}]}^{T} - - - (9)

The starlight vector that then can obtain fixed star i is a elevation angle among the n in navigation

And position angle

{\tilde{h}}_{i}^{n} = \arcsin (p_{3}) - - - (10)

{\tilde{A}}_{i}^{n} = \arctan (p_{1} / p_{2}) - - - (11)

Because to choose sky, northeast Department of Geography is n as navigation coordinate, the fixed star i starlight vector that obtains is a elevation angle among the n in navigation

The measurement elevation angle H of the relative surface level of fixed star just ₀, can be directly used in analytic Height difference method location.

C. analytic Height difference locating module 204 receives the carrier positions information L that SINS determines _I, λ _IAs iteration initial value Lat _AP, Lon _AP, and in conjunction with the elevation angle H that measures many relative surface levels of fixed star that elevation angle computing unit 203 provides ₀, use analytic Height difference method can directly obtain the carrier longitude and latitude that is similar to, can rapidly converge to higher precision by the iterative resolution altitude difference method, finally obtain the latitude information L of CNS location _C, longitude information λ _C

Step 3: set up the dark combined system state model of SINS/CNS and measure model;

Measure the mathematics horizontal reference that elevation angle computing unit 203 provides according to SINS navigation calculation unit 102

Determine the measurement elevation angle H of observation fixed star ₀, and offer analytic Height difference locating module 204; And analytic Height difference locating module 204 utilizes the fixed star elevation angle measurement information H that obtains ₀, and navigation resolves the carrier positions L of unit 102 transmission _I, λ _IAs iterative initial value, it is definite that utilization analytic Height difference method is carried out the CNS position, thereby the CNS positioning error is relevant with SINS horizontal attitude error.If ignored the relation between measurement information and the state variable, can influence the estimated accuracy of Kalman filter, and may cause system's instability.Therefore set up CNS Model of locating error based on the mathematics horizontal reference, and the influence of elimination of level fiducial error in the measurement equation of the position of integrated navigation wave filter 3.

The error model of SINS/CNS deep integrated navigation system is made up of SINS, CNS error model.

A. make up the dark combined system state model of SINS/CNS;

Dark combined system state equation is taken as the error equation of SINS, and as the error state variable, the state equation of model is with the error of zero of platform misalignment, velocity error, site error and inertia device:

\overset{\cdot}{X} = FX + GW - - - (12)

Wherein, X is the error state vector of SINS; The error state of SINS comprises that east, north, sky are to misalignment φ _E, φ _N, φ _U, velocity error δ V _E, δ V _N, δ V _U, latitude, longitude and height error δ L _I, δ λ _I, δ h, gyro zero drift ε _Bx, ε _By, ε _BzAccelerometer zero-bit biasing ▽ _Bx, ▽ _By, ▽ _BzF is the system state matrix, W=[ω _Gx, ω _Gy, ω _Gz, ω _Dx, ω _Dy, ω _Dz] ^TBe system noise sequence, ω _Gi(i=x, y, z), ω _Dt(i=x, y z) are respectively gyroscope, accelerometer random white noise, C _b ⁿBe SINS strapdown matrix, G is a noise matrix, F _NState matrix for SINS.F, F _SAnd G (t) is respectively

B. make up the dark combined system of SINS/CNS and measure model;

1) will be the direction cosine matrix of the relative the earth's core of b equator inertial system I by SINS with the definite carrier of star sensor respectively

Between the difference note measurement amount Z that gestures _s, then obtain formula (6).

With Z _{S (3 * 3)}Be launched into column vector Z _{1 (9 * 1)}, in conjunction with the state vector X of integrated navigation system, can be listed as and write out measurement equation and be:

Z ₁＝H ₁X+V ₁ (13)

Wherein, H ₁For measuring matrix, V ₁Measurement white noise sequence for star sensor.

2) the carrier latitude L that SINS is resolved _I, longitude λ _I, the latitude L that resolves of CNS _C, longitude λ _CDifference as the position detection amount, obtain:

\{\begin{matrix} δLat = L_{I} - L_{C} = {δL}_{I} - (δ L_{C} + N_{L}) \\ δLon = λ_{I} - λ_{C} = δ λ_{I} - (δ λ_{C} + N_{λ}) \end{matrix} - - - (14)

Wherein, δ Lat and δ Lon are respectively difference of latitude, the difference of longitude of SINS, CNS location, δ L _IWith δ λ _IBe the positioning error of SINS, δ L _CWith δ λ _CBe the CNS positioning error that mathematics horizontal reference error causes, N _L, N _λBe the white Gaussian noise component in the CNS positioning error.Obtain;

Z ₂＝H ₂X+V ₂ (15)

In the formula, the position detection vector

Z

_{2} = [\begin{matrix} δLat \\ δLon \end{matrix}] = [\begin{matrix} L_{I} - L_{C} \\ λ_{I} - λ_{C} \end{matrix}];

The white noise component of CNS positioning error

V

_{2} = - [\begin{matrix} N_{L} \\ N_{λ} \end{matrix}];

Measure matrix H ₂=[H _c0 _{2 * 3}I _{2 * 2}0 _{2 * 7}], H wherein _c=M (A ^TA) ^-1A ^TB,

M = [\begin{matrix} 1 & 0 \\ 0 & 1 / \cos (L_{I}) \end{matrix}],

Suppose that n (n 〉=3) the fixed star true azimuth that the star sensor observation of big visual field obtains is A _XNi(i=1,2 ... n), the position vector in star sensor measurement coordinate system s is X _i ^s=[a _1ia _2ia _3i] ^T, as the direction cosine matrix of mathematics horizontal reference be

{\tilde{C}}_{b}^{n} = [\begin{matrix} T_{11} & T_{12} & T_{13} \\ T_{21} & T_{22} & T_{23} \\ T_{31} & T_{32} & T_{33} \end{matrix}],

Then obtain easily

A = [\begin{matrix} \cos A_{zN 1} & \sin A_{zN 1} \\ \cos A_{zN 2} & \sin A_{zN} 2 \\ \cdot \cdot \cdot & \cdot \cdot \cdot \\ \cos A_{zNn} & \sin A_{zNn} \end{matrix}],

B = \begin{matrix} {[\begin{matrix} d_{E 1} & d_{N 1} & 0 \\ d_{E 2} & d_{N 2} & 0 \\ \cdot \cdot \cdot & \cdot \cdot \cdot & \cdot \cdot \cdot \\ d_{En} & d_{Nn} & 0 \end{matrix}]}_{nx 3} \end{matrix}

d_{Ei} = \frac{T_{11} a_{1 i} + T_{12} a_{2 i} + T_{13} a_{3 i}}{\sqrt{1 - (T_{31} a_{1 i} + T_{32} a_{2 i} + T_{33} a_{3 i})^{2}}},

d_{Ni} = \frac{- (T_{21} a_{1 i} + T_{22} a_{2 i} + T_{23} a_{3 i})}{\sqrt{1 - (T_{31} a_{1 i} + T_{32} a_{2 i} + T_{33} a_{3 i})^{2}}}

3) when dark combined system is started working, because SINS has certain accumulation of error, at first utilize the auxiliary SINS of star sensor to obtain high precision mathematics horizontal reference information, this moment, Z was measured in the observation of integrated navigation system _SINS/CNS=Z ₁, corresponding measurement model is formula (13).

Auxiliary SINS obtains on the basis of high level of accuracy benchmark at star sensor, can carry out the celestial navigation location, obtains astronomical fixation latitude and longitude information L _C, λ _CThis moment, the observation of SINS/CNS deep integrated navigation system was measured:

Z_{SINS / CNS} = [\begin{matrix} Z_{1} \\ Z_{2} \end{matrix}]

This moment, corresponding measurement model was

[\begin{matrix} Z_{1} \\ Z_{2} \end{matrix}] = [\begin{matrix} H_{1} \\ H_{2} \end{matrix}] X + [\begin{matrix} V_{1} \\ V_{2} \end{matrix}]

Step 4: integrated navigation system information fusion;

The carrier positions L that integrated navigation wave filter 3 utilizes strapdown inertial navitation system (SINS) 1 and celestial navigation system 2 to determine respectively _I, λ _I, L _C, λ _C, the attitude measurement information Error state to SINS is estimated, obtains the estimation of error information of combined system navigational parameter and inertia device, and these control informations are fed back in the navigation calculation unit 102, and navigational parameter and component error are proofreaied and correct.

The auxiliary mutually hi-Fix of realizing of step 5: SINS and CNS;

Navigation calculation unit 102 obtains high-precision SINS strapdown matrix according to the SINS navigational parameter after proofreading and correct

And it is offered as the mathematics horizontal reference measure elevation angle computing unit 203, obtain the more accurate fixed star elevation angle measurement information on plane relatively, be transferred to analytic Height difference locating module 204 then and be used for CNS and locate, can improve the CNS bearing accuracy.Integrated navigation wave filter 3 receives the more accurate position quantity measurement information that analytic Height difference locating module 204 provides, realization is estimated more accurately to the SINS error state, and these error estimates are fed back to navigation calculation unit 102, can further improve the navigation accuracy of SINS, finally make up deeply and realize the precise navigation location by SINS/CNS.

Claims

1. A kind of SINS/CNS deep integrated navigation system, is characterized in that, comprises strapdown inertial navigation system, celestial navigation system, integrated navigation filter, inertial navigation attitude measurement information construction unit;

The strapdown inertial navigation system includes an inertial measurement component and a navigation calculation unit; the inertial measurement component measures the angular velocity and specific force of the carrier relative to the inertial space, and transmits the obtained angular velocity and specific force information of the carrier to the navigation calculation unit. According to the information transmitted by the inertial measurement component, the unit calculates the carrier's position information latitude LI and longitude λ _I , speed and attitude in real time through the mechanical arrangement algorithm; the navigation calculation unit inputs the carrier's position information L _I and λ _I to the integrated navigation filter In , the current navigation time t, the carrier's position L _I , λ _I and attitude are input into the inertial navigation attitude measurement information construction unit, and the navigation calculation unit also inputs the SINS strapdown matrix as a mathematical horizontal reference into the celestial navigation system The measuring height angle calculation unit of the carrier inputs the location information L _I and λ _I of the carrier into the analytic height difference positioning module in the celestial navigation system;

The celestial navigation system includes a star sensor with a large field of view, a multi-vector attitude determination unit, a measurement altitude angle calculation unit, and an analytical altitude difference positioning module;

The large field of view star sensor can observe the starlight vector information of three or more stars at the same time, and provide the obtained observation information to the multi-vector attitude determination unit and the measurement altitude calculation unit; the multi-vector attitude determination unit Process the received starlight vector information to obtain the attitude information of the carrier relative to the inertial space

and determine the inertial attitude information of the carrier

It is input into the integrated navigation filter; the measurement altitude angle calculation unit uses the starlight vector information transmitted by the star sensor and the mathematical horizontal reference provided by the navigation calculation unit to obtain the measurement altitude angle H ₀ of the star relative to the ground plane, and calculate the altitude The angular measurement information is input to the analytical height difference positioning module; the analytical height difference positioning module obtains the astronomical positioning according to the stellar altitude information transmitted by the measurement altitude angle calculation unit and the carrier position information L _I and λ _I provided by the navigation calculation unit Carrier latitude L _C , longitude λ _C , and astronomical positioning results L _C , λ _C as measurement information input to the integrated navigation filter;

The inertial navigation attitude measurement information construction unit obtains the inertial navigation attitude measurement information according to the navigation parameters transmitted by the navigation calculation unit, and the inertial navigation attitude measurement information is determined from the geocentric equator inertial coordinate system by the strapdown inertial navigation system Direction cosine matrix transformed to vector coordinate system

The inertial navigation attitude measurement information to be obtained by the inertial navigation attitude measurement information construction unit

Provided to the integrated navigation filter;

The integrated navigation filter is based on the SINS position information L _I , λ _I , and CNS attitude measurement information respectively provided by the navigation calculation unit, the multi-vector attitude determination unit, the analytical height difference positioning module, and the inertial navigation attitude measurement information construction unit

CNS positioning results L _C , λ _C and inertial navigation attitude measurement information

Through the Kalman filtering process, the navigation parameters of the strapdown inertial navigation system and the errors of the inertial measurement components are estimated, and they are fed back to the SINS navigation calculation unit to correct and compensate the corresponding errors.

2. a kind of realization method of SINS/CNS deep integrated navigation system is characterized in that, comprises the following steps:

Step 1: The large-field star sensor assists the strapdown inertial navigation system to obtain a high-precision mathematical level benchmark;

The large field of view star sensor obtains the multidimensional starlight vector information of three or more stars at the same time, and then the multi-vector attitude determination unit processes the multidimensional starlight vector information to obtain the measurement coordinate system s of the large field of view star sensor Direction cosine matrix relative to the geocentric equator inertial frame I

Combined with the installation matrix C _b ^s of the large-field star sensor on the carrier, the direction cosine matrix of the carrier system b relative to the inertial system I at the equator of the earth is obtained

The inertial navigation attitude measurement information construction unit constructs the direction cosine matrix of the carrier system b relative to the earth-centered equator inertial system I determined by SINS through the navigation information output by the navigation calculation unit

Attitude information output by star sensor with large field of view

Matching, the navigation information is the current navigation time t, the positioning information L _I of SINS, λ _I and the mathematical level reference

The direction cosine matrix of

The specific calculation method is:

Use the location information L _I and λ _I of SINS to construct the location matrix of SINS According to the current navigation time t, the direction cosine matrix C I _e converted from the geocentric equatorial inertial system I to the earth fixed coordinate system ^e is obtained, combined with the SINS strapdown matrix and the position matrix get:

{\overset{^^}{c c}}_{I I}^{b b} = = {\overset{^^}{C C}}_{n no}^{b b} {\overset{^^}{C C}}_{e e}^{n no} {C C}_{I I}^{e e} = = {(({\overset{^^}{C C}}_{b b}^{n no}))}^{T T} {\overset{^^}{C C}}_{e e}^{n no} {C C}_{I I}^{e e} - - - - - - ((11))

There is a mathematical platform misalignment angle vector between the SINS mathematical platform system n′ and the navigation coordinate system n

φ _E , φ _N , φ _U are misalignment angles in east, north and sky directions, SINS strapdown matrix

The error is related to the misalignment angle; due to the existence of latitude error δL _I and longitude error δλ _I of SINS positioning, the calculation system nc of SINS does not coincide with the actual geographic system n, and there is a position error vector δP=[-δL _I δλ _I cosL _I δλ _I sinL _I ] ^T , resulting in the direction cosine matrix of the navigation system n relative to the earth-fixed coordinate system e determined by SINS

It is related to the position deviation δL _I and δλ _I of SINS, then there is the following relationship:

{\overset{^^}{C C}}_{e e}^{n no} = = ((I I - - [[δP δP \times \times]])) {C C}_{e e}^{n no} - - - - - - ((33))

Among them, C _n ^b and C _e ⁿ are the direction cosine matrices of the real carrier system relative to the navigation system and the navigation system relative to the earth fixed coordinate system, respectively;

Neglecting the quadratic and more than quadratic error terms, the cosine matrix of the direction of the carrier body relative to the earth-centered equatorial inertial system determined by SINS

Expressed as

{\overset{^^}{C C}}_{I I}^{b b} = = {C C}_{I I}^{b b} + + {C C}_{n no}^{b b} [[φ φ \times \times]] {C C}_{e e}^{n no} {C C}_{I I}^{e e} - - {C C}_{n no}^{b b} [[δP δP \times \times]] {C C}_{e e}^{n no} {C C}_{I I}^{e e} - - - - - - ((44))

It is assumed that the direction cosine matrix of the ideal error-free carrier system b relative to the earth-centered equatorial inertial system I is C _I ^b ; it is considered that the direction cosine matrix of the carrier system output by the star sensor relative to the earth-centered equatorial inertial system

is the superposition of the real direction cosine matrix C _I ^b and the white noise array V _s measured by the star sensor, that is:

{\overset{~ ~}{C C}}_{I I}^{b b} = = {C C}_{I I}^{b b} + + {V V}_{s the s} - - - - - - ((55))

The direction cosine matrix of the carrier body relative to the earth-centered equatorial inertial system determined by the SINS and the star sensor respectively

The difference between is denoted as Z _s , then there is

{Z Z}_{s the s} = = {\overset{^^}{C C}}_{I I}^{b b} - - {\overset{~ ~}{C C}}_{I I}^{b b} = = {C C}_{n no}^{b b} [[φ φ \times \times]] {C C}_{e e}^{n no} {C C}_{I I}^{e e} - - {C C}_{n no}^{b b} [[δP δP \times \times]] {C C}_{e e}^{n no} {C C}_{I I}^{e e} - - {V V}_{s the s} - - - - - - ((66))

The misalignment angle of the platform has a coupling relationship with the constant value drift of the gyroscope, and the cosine array of the direction of the carrier body relative to the earth-centered equatorial inertial system determined by the star sensor and the SINS is determined by the integrated navigation filter.

Carry out information fusion, estimate the misalignment angle and gyro drift of the SINS mathematical platform in real time, and then correct the misalignment angle and gyro drift of the SINS mathematical platform to improve the mathematical level benchmark

The specific method is as follows:

The gyroscope and accelerometer in the inertial measurement module output the specific force of the carrier in real time

and angular velocity

Send to the navigation calculation unit; in the navigation calculation unit, the specific force information of the carrier

After being processed by the mathematical level reference, it is directly input to the north-pointing platform inertial navigation calculation process for navigation calculation, and the carrier's position L _I , λ _I , speed, and attitude navigation information are obtained; the estimated value of the misalignment angle is used to calculate the attitude Error correction matrix C _n′ ⁿ , via the formula

strapdown matrix

Correct the misalignment angle to improve the mathematical level benchmark accuracy; the output from the gyroscope

The gyro drift error δω _ib ^b is deducted in real time, and the gyro drift error is compensated to obtain accurate angular velocity vector information of the carrier relative to the inertial space

{\tilde{ω}}_{ib}^{b} = {\hat{ω}}_{ib}^{b} - δ ω_{ib}^{b},

The angular velocity of the navigation system relative to the earth-centered equator inertial system is obtained by combining the mathematical horizontal datum and the north-pointing platform inertial navigation solution

Calculate the high-precision attitude rate of the carrier system relative to the navigation system

And through the direction cosine matrix differential equation

{\overset{&Center Dot;}{C}}_{b}^{no} = C_{b}^{no} Ω_{nb}^{b}

Update the strapdown matrix, where Ω _nb ^b is the angular velocity vector

The cross-product matrix in the carrier coordinate system;

Finally, the strapdown inertial navigation system obtains a high-precision strapdown matrix

i.e. Mathematics Proficiency Benchmark;

Step 2: Carry out CNS positioning based on the mathematical level benchmark;

a. Use star sensor to assist strapdown inertial navigation system to obtain high-precision SINS strapdown matrix

It is transmitted to the astronomical navigation system as a mathematical horizontal reference; and the position information L _I and λ _I output by the navigation calculation unit provide the iterative latitude initial value Lat _AP and longitude initial value Lon _AP for the positioning of the analytical height difference method;

b. The representation of the starlight vectors of three or more stars observed at the same time by the large-field star sensor in the star sensor measurement coordinate system s and the earth-centered equatorial inertial system I; due to the star sensor measurement noise V The existence of _i ^s , the actual measured position vector of star i in the star sensor measurement coordinate system s is:

{\overset{~ ~}{X x}}_{i i}^{s the s} = = {X x}_{i i}^{s the s} + + {V V}_{i i}^{s the s} - - - - - - ((77))

In the formula, X _i ^s is the real position vector of star i in the star sensor measurement coordinate system s;

At navigation time t, the star sensor assists SINS to obtain a high-precision mathematical horizontal reference Combined with the installation matrix C _b ^s of the star sensor, the representation of the starlight vector of the star i in the navigation coordinate system n is obtained

{\overset{~ ~}{X x}}_{i i}^{n no} = = {\overset{~ ~}{C C}}_{b b}^{n no} {C C}_{s the s}^{b b} {\overset{~ ~}{X x}}_{i i}^{s the s} = = {\overset{~ ~}{C C}}_{b b}^{n no} {(({C C}_{b b}^{s the s}))}^{T T} {\overset{~ ~}{X x}}_{i i}^{s the s} - - - - - - ((88))

The calculated starlight position vector

Use the altitude angle of the star defined in the navigation system n

and azimuth express:

{\overset{~ ~}{X x}}_{i i}^{n no} = = {[\begin{matrix} cos cos {\overset{~ ~}{h h}}_{i i}^{n no} sin sin {\overset{~ ~}{A A}}_{i i}^{n no} & cos cos {\overset{~ ~}{h h}}_{i i}^{n no} cos cos {\overset{~ ~}{A A}}_{i i}^{n no} & sin sin {\overset{~ ~}{h h}}_{i i}^{n no} \end{matrix}]}^{T T} = = {[\begin{matrix} {p p}_{11} & {p p}_{22} & {p p}_{33} \end{matrix}]}^{T T} - - - - - - ((99))

Then get the altitude angle of the starlight vector of star i in the navigation system n

and azimuth

{\overset{~ ~}{h h}}_{i i}^{n no} = = arcsin arcsin (({p p}_{33})) - - - - - - ((1010))

{\overset{~ ~}{A A}}_{i i}^{n no} = = arctan arctan (({p p}_{11} / / {p p}_{22})) - - - - - - ((1111))

Since the northeast astronomical system is selected as the navigation coordinate system n, the altitude angle of the star i starlight vector in the navigation system n is obtained

is the measured altitude angle H ₀ of the star relative to the horizontal plane;

c. The analysis height difference positioning module receives the carrier position information L _I and λ _I determined by SINS as the iterative initial values Lat _AP and Lon _AP , and combines the altitude angle H ₀ of multiple stars relative to the horizontal plane provided by the measurement altitude angle calculation unit, Use the analytical height difference method to directly obtain the approximate carrier latitude and longitude, and quickly converge to a higher accuracy through the iterative analytical height difference method to obtain the latitude information L _C and longitude information λ _C of the CNS positioning;

Step 3: Establish the state model and measurement model of the SINS/CNS deep combined system;

The error model of SINS/CNS deep integrated navigation system includes SINS and CNS error models;

a. Construct the state model of SINS/CNS deep combined system;

The state equation of the deep combined system is taken as the error equation of the SINS, and the platform misalignment angle, velocity error, position error and zero position error of the inertial device are used as the error state variables. The state equation of the model is:

\overset{\cdot &Center Dot;}{X x} = = FX FX + + GW GW - - - - - - ((1212))

Among them, X is the error state vector of SINS; the error state of SINS includes misalignment angles φ _E , φ _N , φ _U , velocity errors δV _E , δV _N , δV _U , latitude, longitude and height errors δL _I , δλ _I , δh, gyro zero drift ε _bx , ε _by , ε _bz ; accelerometer zero bias ▽ _bx , ▽ _by , ▽ _bz ; F is the system state matrix, W=[ω _gx , ω _gy , ω _gz , ω _dx , ε _dy , ω _dz ] ^T is the system noise sequence, ω _gi (i=x, y, z), ω _di (i=x, y, z) are the random White noise, C _b ⁿ is the SINS strapdown matrix, G is the noise matrix, F _N is the state matrix of SINS; F, F _S and G(t) are respectively

f = {[\begin{matrix} f_{N} & f_{S} \\ 0_{6 \times 9} & 0_{6 \times 6} \end{matrix}]}_{15 \times 15}

b. Construct the measurement model of SINS/CNS deep combined system;

1) The direction cosine matrix of the carrier system b relative to the geocentric equatorial inertial system I determined by the SINS and the star sensor respectively

The difference between is recorded as the attitude measurement Z _s , and the formula (6) is obtained;

Expand Z _s(3×3) into a column vector Z _1(9×1) , combined with the state vector X of the integrated navigation system, write out the measurement equation as follows:

Z ₁ =H ₁ X+V ₁ (13)

Among them, H ₁ is the measurement matrix, V ₁ is the measurement white noise sequence of the star sensor;

2) Taking the carrier latitude _L _I and longitude λ I calculated by SINS, and the difference between latitude L _C and longitude λ _C calculated by CNS as position observations, we can get:

\{\begin{matrix} δLat δLat = = {L L}_{I I} - - {L L}_{C C} = = δ δ {L L}_{I I} - - (({δL δ L}_{C C} + + {N N}_{L L})) \\ δLon δLon = = {λ λ}_{I I} - - {λ λ}_{C C} = = δ δ {λ λ}_{I I} - - (({δλ δλ}_{C C} + + {N N}_{λ λ})) \end{matrix} - - - - - - ((1414))

Among them, δLat and δLon are the latitude difference and longitude difference between SINS and CNS positioning respectively, δL _I and δλ _I are the positioning errors of SINS, δL _C and δλ _C are the CNS positioning errors caused by the mathematical horizontal reference error, N _L , N _λ is the Gaussian white noise component in the CNS positioning error; get;

Z ₂ =H ₂ X+V ₂ (15)

In the formula, the position observation vector

Z_{2} = [\begin{matrix} δLat \\ δLon \end{matrix}] = [\begin{matrix} L_{I} - L_{C} \\ λ_{I} - λ_{C} \end{matrix}];

White noise component of CNS positioning error

V_{2} = - [\begin{matrix} N_{L} \\ N_{λ} \end{matrix}];

Measurement matrix H ₂ =[H _c 0 _2×3 I _2×2 0 _2×7 ], where H _c =M·(A ^T A) ^-1 A ^T ·B,

m = [\begin{matrix} 1 & 0 \\ 0 & 1 / \cos (L_{I}) \end{matrix}],

Assuming that the true azimuth angles of n (n≥3) stars observed by the large field of view star sensor are _AzNi (i=1, 2...n), the position vector in the star sensor measurement coordinate system s is

x_{i}^{the s} = {[\begin{matrix} a_{1 i} & a_{2 i} & a_{3 i} \end{matrix}]}^{T},

The direction cosine matrix used as a math level datum is

{\tilde{C}}_{b}^{no} = [\begin{matrix} T_{11} & T_{12} & T_{13} \\ T_{twenty one} & T_{twenty two} & T_{twenty three} \\ T_{31} & T_{32} & T_{33} \end{matrix}],

then get:

A A = = [\begin{matrix} cos cos {A A}_{zN Z 11} & sin sin {A A}_{zN Z 11} \\ cos cos {A A}_{zN Z 22} & sin sin {A A}_{zN Z 22} \\ . . . . . . & . . . . . . \\ cos cos {A A}_{zNn n} & sin sin {A A}_{zNn n} \end{matrix}],,

B B = = {[\begin{matrix} {d d}_{E E. 11} & {d d}_{N N 11} & 00 \\ {d d}_{E E. 22} & {d d}_{N N 22} & 00 \\ . . . . . . & . . . . . . & . . . . . . \\ {d d}_{En En} & {d d}_{Nn n} & 00 \end{matrix}]}_{n no \times \times 33}

{d d}_{Ei Ei} = = \frac{{T T}_{1111} {a a}_{11 i i} + + {T T}_{1212} {a a}_{22 i i} + + {T T}_{1313} {a a}_{33 i i}}{\sqrt{11 - - {(({T T}_{3131} {a a}_{11 i i} + + {T T}_{3232} {a a}_{22 i i} + + {T T}_{3333} {a a}_{33 i i}))}^{22}}},,

{d d}_{Ni Ni} = = \frac{- - (({T T}_{21 twenty one} {a a}_{11 i i} + + {T T}_{22 twenty two} {a a}_{22 i i} + + {T T}_{23 twenty three} {a a}_{33 i i}))}{\sqrt{11 - - {(({T T}_{3131} {a a}_{11 i i} + + {T T}_{3232} {a a}_{22 i i} + + {T T}_{3333} {a a}_{33 i i}))}^{22}}}

3) When the deep integrated system starts to work, first use the star sensor to assist SINS to obtain high-precision mathematical horizontal reference information. At this time, the observation quantity of the integrated navigation system is Z _SINS/CNS = Z ₁ , and the corresponding measurement model is the formula ( 13);

On the basis of the high-precision horizontal reference obtained by the star sensor assisted SINS, astronomical navigation and positioning are performed to obtain astronomical positioning latitude and longitude information L _C , λ _C ; at this time, the observations of the SINS/CNS deep integrated navigation system are:

{Z Z}_{SINS SINS / / CNS CNS} = = [\begin{matrix} {Z Z}_{11} \\ {Z Z}_{22} \end{matrix}]

The corresponding measurement model at this time is

[\begin{matrix} {Z Z}_{11} \\ {Z Z}_{22} \end{matrix}] = = [\begin{matrix} {H h}_{11} \\ {H h}_{22} \end{matrix}] X x + + [\begin{matrix} {V V}_{11} \\ {V V}_{22} \end{matrix}]

Step 4: Integrated navigation system information fusion;

The integrated navigation filter uses the carrier position L _I , λ _I , L _C , λ _C , and attitude measurement information determined by the strapdown inertial navigation system and the celestial navigation system respectively.

Estimate the error state of SINS, obtain the error estimation information of the navigation parameters of the combined system and the inertial device, and feed these error information back to the navigation calculation unit to correct the navigation parameters and component errors;

Step 5: SINS and CNS assist each other to achieve high-precision positioning;

The navigation calculation unit obtains a high-precision SINS strapdown matrix according to the corrected SINS navigation parameters

Provide it as a mathematical level reference to the measurement altitude angle calculation unit to obtain more accurate measurement information of the altitude angle of the star relative to the ground plane, and transmit it to the analytical altitude difference positioning module for CNS positioning; the integrated navigation filter receives the analytical altitude The more accurate position measurement information provided by the differential positioning module realizes a more accurate estimation of the SINS error state, feeds the error estimation value back to the navigation calculation unit, further improves the navigation accuracy of the SINS, and finally through the deep combination of SINS/CNS Realize high-precision navigation and positioning.