CN104406605A - Aircraft-mounted multi-navigation-source comprehensive navigation simulation system - Google Patents

Abstract

The invention discloses an aircraft-mounted multi-navigation-source comprehensive navigation simulation system. In the system, each navigation simulation sub system really simulates functions, performances and interfaces which are same as those of a model according to a self-navigation principle. A scheduling system responds a display control command, a load command and load data information sent by a mission system, and drives the navigation simulation sub systems to perform simulation calculation with combination of environment simulation data. A combined navigation system, with an INS simulation sub system as a main navigation source, performs integrated navigation based on and fault detection federated Kalman filter with GNSS, TACAN and IFDL as auxiliary navigation sources. Information fusion is achieved through a generalized federated Kalman filter in a two-stage structure constructed through a federated filter algorithm, wherein a main filter performs total information fusion and performs information distribution and information reset to each sub filter. Optimal estimation is carried out to a result of each sub filter through a dynamic information distribution coefficient. Meanwhile, the INS simulation sub system is subjected to closed-loop negative feedback correction through an estimated error by the system.

Description

Airborne many navigation sources integrated navigation analogue system

Technical field

The present invention relates to a kind of integrated navigation analogue system being mainly used in airborne platform, in particular for airborne many navigation integrated navigation analogue systems.

Background technology

In the prior art, various navigation sources with different principle of work for aircraft provides various navigation information, as inertial navigation system (INS), satellite navigation system (GNSS, as GPS, the Big Dipper etc.), chain navigational system (IFDL) etc. between Tacan (TACAN) and machine, if be only used alone any one in them during navigation, then always there is such or such drawback: (1) INS system is due to the constant value error of zero of accelerometer, the Initial Alignment Error of gyrostatic constant value drift system, between and produce attitude error, whole inertial navigation system can be made to occur positioning error, (2) although GNSS has the whole world, round-the-clock, precision is high and without advantages such as accumulated errors, anti-interference is poor, be subject to terrain shading and signal interruption occurs, (3) TACAN is a kind of short-range radio navigational system, and it cannot use outside ground station overlay area, and distance station navigation error far away is larger, (4) IFDL utilizes time of arrival and arrival direction between formation member to carry out relative positioning, and its location information updating rate is low, without attitude information, without shortcomings such as Geographic Navigation abilities.Along with the development of airmanship and control theory, integrated navigation system has numerous navigation sensor modules, has constituted a Multi-sensor Navigation infosystem.Information fusion method solves the strong means of multi-sensor information overall treatment problem.By the optimum fusion of multi-source information, precision and the reliability of integrated navigation system effectively can be improved.Airborne many navigation sources integrated navigation system adopts advanced navigation information Intelligent Fusion technology various navigational system to be combined, make full use of its redundancy and complementarity, form an organic whole, annex the advantage of each system and complementary shortcoming, while raising navigation accuracy, realize carrying out fault detect to each navigation sources in real time, diagnosis, isolated fault equipment and reconfiguration system, greatly improve the reliability of navigation information, make the synthetical electronics information system can provide unified for opportunity of combat all the time in the battlefield surroundings of complexity, optimum, reliably, navigational parameter during long boat.

Domestic analogue system or be single navigational system, or be that a few single navigational system is combined, the integrated navigation system that investigation and application is maximum mainly contains GPS+INS and GPS+INS+TACAN.All air navigation aids all design for specific specific requirement, can contain at present all typical navigation members, the air navigation aid of adaptation round voyage also do not occur.For the combination of dissimilar navigational system, often adopt Kalman filtering.A shortcoming of Kalman filtering is when the state of system is undergone mutation (as course or speed of a ship or plane generation flip-flop), and the system state that filtering obtains can not the change of at once tracker time of day, thus have impact on filter effect.A large amount of test figures is needed in the process of navigation algorithm research, the exploitation of integrated navigation system wave filter, and True Data is difficult to obtain, emulation technology is utilized to produce test figure, can be the research of navigation algorithm, the exploitation of integrated navigation system wave filter provides a kind of effective means, not only convenient but also economical.

Summary of the invention

The present invention is directed to the weak point that prior art exists, one is provided to comprise navigation information source more comprehensively, comprehensive function is stronger, the scope of application is wider, compages are more reasonable, can either be used alone, also can carry out interoperability with other analogue system, there is stronger versatility, and airborne many navigation sources integrated navigation analogue system of accessible emulation platform simulated environment.

Above-mentioned purpose of the present invention can be reached by following measures, a kind of airborne many navigation sources integrated navigation analogue system, comprise the fault-tolerance combined navigation system based on INS/GNSS/TACAN/IFDL, dispatching system, radio altimeter ALT simulation subsystem, microwave landing MLS simulation subsystem, instrument landing ILS simulation subsystem, micro-distancer DME/P simulation subsystem, single navigational system such as chain IFDL between machine, it is characterized in that: adopt the stratification that line strainer model combines with the filtrator mode based on software component, modular Distributed Simulation Architecture, each single navigation simulation subsystem emulates the function identical with model realistically according to self navigation principle, performance and interface, the aobvious control instruction that dispatching system response task system issues, load instructions and loading data message, independently select single navigation sources or integrated navigation to carry out navigator fix, and combining environmental emulated data drives various navigation simulation subsystem to perform simulation calculating, integrated navigation system is using INS simulation subsystem as main navigation sources, respectively with GNSS, TACAN, IFDL is as assisting navigation source, state vector is INS navigational parameter error, the Generalized Federated Kalman filter adopting Federated Filtering to build two-layer configuration completes information fusion, senior filter carries out total information fusion, and information distribution and replacement are carried out to each subfilter, GNSS/INS subfilter 1, TACAN/INS subfilter 2 and IFDL/INS subfilter 3 complete the state vector local estimation of INS/GNSS/TACAN/IFDL integrated navigation system respectively, senior filter adopts multidate information partition factor to carry out optimal estimation to the result of subfilter.In fault-tolerance combined navigation system, adopt state residual χ ²method of inspection carries out diagnosis and detection fault in real time to each subfilter; Fuzzy logic adaptive controller FLAC constantly monitors whether residual error r is zero-mean white noise; The weighted factor (k) of the estimation error variance of each subfilter is controlled again, for adjustment and measurement noises variance matrix Q according to fuzzy inference rule _kand R _k, finally according to modern control theory, antithetical phrase navigation sources INS implements output calibration or close-loop feedback corrects.

The present invention has following beneficial effect compared to prior art:

Airborne many navigation sources integrated navigation simulation system software that the present invention proposes adopts open architecture design, according to stratification, modular principle of design, accessible Iarge-scale system platform emulation environment, can either be used alone, also can emulate with other and carry out interoperability, select different navigation subsystem according to different user's requests and different application purposes, improve adaptability and the portability of system.

The navigation information source comprised more comprehensively.Airborne many navigation sources integrated navigation system in the present invention substantially covers the navigate mode of current all main flows, specifically covers the analogue system of INS simulation subsystem, GNSS simulation subsystem, TACAN simulation subsystem, ALT system, MLS simulation subsystem, ILS system, DME/P system, IFDL relative navigation system and integrated navigation (INS+GNSS+TACAN+IFDL integrated navigation).Can the intactly single navigation feature such as Simulation with I NS, GNSS, TACAN, ALT, MLS, ILS, DME/P, IFDL, the integrated navigation function of normal-sub navigation sources signal reconstruct system globe area algorithm simulation GNSS+INS, TACAN+INS, IFDL+INS, GNSS+TACAN+INS+IFDL can also be utilized, can according to aircraft the different task stage work select different navigation system, reproduce the function of each single navigation sources and all kinds of integrated navigation system, error characteristics, interface features and data transmission relations as far as possible realistically.

Comprehensive function is stronger, and the scope of application is wider.The machine navigation information and IFDL chain navigation information comprehensively complete consonance function, complementary function and remaining function.This invention is not only applicable to the accurate geographic position navigation of carrier aircraft self, is applicable to accurate relative position between formation opportunity of combat yet and determines.Function: the MLS+DME/P that such as coordinates consonance can provide aircraft relative to the positional information of landing point, and TACAN+ALT consonance can provide aircraft relative to the positional information etc. of control tower.Complementation and remaining function: INS+GNSS+TACAN+IFDL is combined, mutually learn from other's strong points to offset one's weaknesses, remaining each other, to strengthen, sophisticated systems function and performance, obtain the multidimensional precise navigation information of Gao Gengxin rate, improve system accuracy, reliability, strengthen antijamming capability, realize high-precision Relative Navigation and Geographic Navigation.

Compages are more reasonable.The stratification that the present invention adopts line strainer model to combine with the filtrator mode based on software component, modular Distributed Simulation Architecture, can increase sub-navigation sources and sub-navigation sources subsystem very easily, software has extraordinary extendability and opening; Simulation software adopts VC++ multitasking form, each sub-navigation sources subsystem software carries out the process of parametric degree functional software by mission thread mode, realized the exchanges data of each several part between data task by the global sharing memory of message queue or internal system.

There is comparatively strong fault tolerance performance.Integrated navigation technology of the present invention employs two-stage Federated Kalman Filter, wherein GNSS/INS subfilter 1, TACAN/INS subfilter 2 and IFDL/INS subfilter 3 complete the state vector local estimation of INS/GNSS/TACAN/IFDL integrated navigation system respectively, senior filter carries out total information fusion, and information distribution and replacement are carried out to each subfilter, overcome and be used alone INS, GNSS, the cumulative errors produced respectively during TACAN or IFDL is large, thrashing, become large apart from navigation error during station distance, the problems such as navigation information is not comprehensive, and by the sub-navigation sources that multistage fault detection technique automatism isolation breaks down, make full use of the sub-navigation sources output information of normal work, obtain reliable, navigational parameter is estimated accurately, the close-loop feedback simultaneously utilizing this estimated result to realize antithetical phrase navigation sources INS corrects.

The present invention can be used for test and the Performance Analysis of aerocraft system, also can be used for various navigation model system error analysis and airborne single navigation sources navigational parameter is determined, also be applicable to the engineer applied algorithm research of INS/GNSS/TACAN/IFDL integrated navigation system, help to break through integrated navigation gordian technique.

Accompanying drawing explanation

Below in conjunction with drawings and Examples, the present invention is further described.

Fig. 1 is the present invention's airborne many navigation sources integrated navigation emulate system architecture figure.

Fig. 2 is the true software composition diagram of the present invention's airborne many navigation sources integrated navigation analogue system.

Fig. 3 is INS/GNSS/TACAN/IFDL integrated navigation process flow diagram of the present invention.

Fig. 4 is FLAG structured flowchart of the present invention.

Fig. 5 is fuzzy logic controller FLAG principle schematic of the present invention.

Fig. 6 is INS/GNSS/TACAN/IFDL integrated navigation fault detection and diagnosis FDIR theory diagram of the present invention.

Fig. 7 is Fuzzy ArtMap structural drawing of the present invention.

Fig. 8 is INS simulation subsystem simulation calculation flow process figure of the present invention.

Fig. 9 is GNSS simulation subsystem calculation flow chart of the present invention.

Figure 10 is that TACAN simulation subsystem information of the present invention resolves workflow diagram.

Figure 11 is MLS+DME/P simulation subsystem functional software process flow diagram of the present invention.

Figure 12 is that ALT simulation subsystem height measurements of the present invention resolves workflow diagram.

Figure 13 is that IFDL simulation subsystem measured value of the present invention resolves workflow diagram.

Embodiment

Consult Fig. 1, Fig. 2.In embodiment described below, airborne many navigation sources integrated navigation analogue system comprises airborne many navigation sources emulation part, environmental simulation simulator and external tasks components of system as directed.Specifically comprise inertial navigation system (INS), satellite navigation system (GNSS), Tacan (TACAN), radio altimeter (ALT), microwave land (MLS), instrument landing (ILS), micro-distancer (DME/P), single navigational system, integrated navigation system, environmental simulation simulator and the external tasks system such as chain (IFDL) between machine.Single navigation sources subsystem reflects the error characteristics of existing type navigator, interface relationship and Performance Characteristics truly; Algorithm of combined navigation subsystem selects INS simulation subsystem as main navigation sources, carry out the integrated navigation process based on Federated Kalman Filtering and fault detect process respectively using GNSS, TACAN, IFDL as assisting navigation source, simultaneity factor utilizes estimation error amount to implement close loop negative feedback to INS navigation sources and corrects.Wherein airborne many navigation sources simulation part is divided and is comprised navigation simulation part and scheduling portion.Navigation simulation part is divided into again single navigation simulation system and combined navigation simulation system.Navigation simulation part comprises: chain IFDL Relative Navigation simulation subsystem between inertial navigation INS simulation subsystem, satellite navigation GNSS simulation subsystem, Tacan TACAN simulation subsystem, radio altimeter ALT simulation subsystem, microwave landing MLS simulation subsystem, instrument landing ILS simulation subsystem, micro-distancer DME/P simulation subsystem, machine.The exemplary functions that they realize is as follows respectively:

1. INS simulation subsystem has considered inertia type instrument error characteristics, establishes the random walk model of constant error, first-order Markov process, simulates the measurement data of gyroscope and accelerometer more realistically.

2. GNSS simulation subsystem can drive the deduction computing of satellite constellation according to satellite ephemeris, the reference informations such as standard satellite position, speed are provided for receiver navigator fix, support select four satellites of best geometric dilution of precision and select all visible stars two kinds of satellite selection methods simultaneously.

3. the TACAN simulation subsystem navigation mode that can arrange according to emulation, completes aircraft relative to TACAN surface beacon platform orientation, oblique distance, speed, emulate to the parametric degree of platform time to be flown.

4. the high function of survey of ALT simulation subsystem simulation pressure altimeter, provides the barometer altitude of aircraft.

5. MLS/ILS simulation subsystem and DEM P with the use of, main analog aircraft relative to the azimuth information of orientation, ground control desk, machine relative to the pitch information of ground pitch control subsystem platform and the aircraft range information relative to DME/P platform; Then on this basis, rotational coordinates Iterative algorithm (RGS algorithm) is adopted to obtain the real-time position information of aircraft relative to landing point.

6. IFDL Relative Navigation simulation subsystem completes two or more formation member composition relative navigation system function in IFDL net, on the time division multiple access communication function i ntegration of IFDL system, time of arrival TOA and arrival direction DOA is utilized to carry out precise distance measurement and interaction locations data, the source utilizing active Time transfer receiver RTT to obtain and the clock correction of user clock, then carry out certain data processing, make member obtain its regulation relative coordinate system in coordinate position.

The input of various navigation simulation system is nominal trajectory data and the weather environment information of earth station information, the in the air formation aircraft that environmental simulation provides, and output is the navigation information that often kind of navigation simulation system is supplied to external tasks system.Dispatching system completes the control selected different navigation subsystem, and namely aircraft independently selects single navigation sources or integrated navigation to carry out navigator fix according to scheduler subsystem in full airline operation.The aobvious control instruction that dispatching system response task system issues, load instructions and loading data message, system combining environmental emulated data drives various navigation simulation subsystem to perform simulation calculating.

INS+GNSS+TACAN+IFDL combined navigation simulation system, it is using INS simulation subsystem as main navigation sources, with GNSS, TACAN, IFDL is as assisting navigation source, specifically include three subfilter (GNSS/INS subfilters 1, TACAN/INS subfilter 2, IFDL/INS subfilter 3) and a senior filter (also claiming global filtering device), they adopt indirect filtering mode work: each subfilter all adopts the state vector local estimation of fuzzy logic adaptive Kalman filter completion system, its main flow is by fuzzy logic adaptive controller FLAC, each subfilter constantly monitors whether the residual error r of each subfilter is zero-mean white noise, the weighted factor (k) of the estimation error variance of each subfilter is controlled again, for adjustment process and measurement noises variance matrix Q according to fuzzy inference rule _kand R _k, thus constantly adjust subfilter gain K, make it reach global optimum or suboptimal estimation, then senior filter carries out the common condition Vector Fusion synthesis of subfilter and the time upgrades, output reliably, global optimum's estimator of navigational parameter error accurately, last according to modern control theory, antithetical phrase navigation sources INS implements output calibration or close-loop feedback corrects.

In the design of fault-tolerance combined navigation system, integrated navigation fault detection and diagnosis technology FDIR is adopted to detect in real time and fault diagnosis integrated navigation system.Here utilization state residual error χ ²method of inspection carries out data validity detection to each subfilter, when subfilter non-fault, the filter result of subfilter is sent into Federated Filters; After mutation failure being detected, subfilter by automatism isolation, will utilize other normal-sub navigation sources signal reconstruct system globe area algorithm.In order to identify fault and reconstruct, needing to position fault, Isolate Problem source, failure judgement amplitude and time of origin can be come by the recognition capability of Fuzzy ArtMap fuzzy neural network.

The stratification that airborne many navigation sources integrated navigation simulation system software adopts line strainer model to combine with the filtrator mode based on software component, modular Distributed Simulation Architecture, adopt pipes from single navigation sources subsystem to the message of algorithm of combined navigation subsystem, navigation subsystem realizes adopting the filtrator mode based on software component.When integrated navigation increases sub-navigation sources, increase pipeline; When whole integrated navigation analogue system increases sub-navigation sources subsystem, increase filtrator or/and corresponding software component, thus ensure that extendability and the opening of airborne many navigation sources integrated navigation simulation software.

Consult Fig. 3.Implementation process about INS+GNSS+TACAN+IFDL combined navigation simulation system is expressed as follows: INS/GNSS/TACAN/IFDL combined navigation simulation system, Generalized Federated filtering technique is adopted to complete information fusion, it is completed by a senior filter and three subfilters, state vector is INS navigational parameter error, each subfilter (GNSS/INS subfilter 1, TACAN/INS subfilter 2, IFDL/INS subfilter 3) adopt indirect filtering mode work, the estimation of fuzzy logic adaptive controller FLAG completion status vector local, senior filter (also claiming global filtering device) carries out common condition Vector Fusion and the time renewal of subfilter, export reliable, global optimum's estimator of navigational parameter error accurately.Last according to modern control theory, antithetical phrase navigation sources INS implements output calibration or close-loop feedback corrects.

1. the foundation of Airborne Inertial navigational system ins error state equation

First carrier aircraft position, speed and acceleration choose the solid rectangular coordinate system of ground heart, and attitude angle is the deviation angle of carrier aircraft rectangular coordinate system relative to the geographical rectangular coordinate system in sky, carrier aircraft northeast, and wave filter inertia system state error vector is set to:

X _SINS＝[δX _SINSδY _SINSδZ _SINSδV _XSINSδV _YSINSδV _ZSINSφ _xφ _yφ _zδf _axδf _ayδf _azδV _θxδV _θyδV _θz] ^T

(1)

Wherein δ X _sINS, δ Y _sINS, δ Z _sINSfor the site error of carrier aircraft; δ V _xSINS, δ V _ySINS, δ V _zSINSfor the velocity error of carrier aircraft; φ _x, φ _y, φ _zfor the attitude error of carrier aircraft; δ f _ax, δ f _ay, δ f _azfor the acceleration error of carrier aircraft; δ V _{θ x}, δ V _{θ y}, δ V _{θ z}for the attitude angular velocity error of carrier aircraft.

Then set up carrier aircraft inertial navigation system ins error state equation, as sub-senior filter, the state equation of GNSS/INS subfilter 1 and TACAN/INS subfilter 2 is as follows:

X _i(k)＝F _i(k，k-1)X _i(k-1)+G _i(k，k-1)W _i(k) (2)

Wherein i=1,2 represent subfilter 1 and subfilter 2, F respectively _i(k, k-1) is t _k-1moment is to t _ktime etching system state-transition matrix, W _ik () is t _ktime etching system noise vector, G _i(k, k-1) is t _k-1moment is to t _ktime etching system noise drive matrix; X _i(k-1) be t _k-1moment carrier aircraft INS errors quantity of state, X _ik () is t _kmoment carrier aircraft INS errors quantity of state.

2. the foundation of the observation equation of GNSS/INS subfilter 1

The linearization measurement equation of deviation between the observed quantity of GNSS position and speed and INS simulation subsystem state estimator is set up under the solid rectangular coordinate system of ground heart, as the observation equation of subfilter 1, specific as follows:

Z ₁(k)＝H ₁(k)X ₁(k)+V ₁(k) (3)

Wherein H ₁=[I _{6 × 6}0 _{6 × 9}] be t _kmoment position and speed measurement matrix, V ₁k () is t _kthe position and speed measurement noise vector of moment GNSS receiver.As GPS/SINS subfilter adopts position and speed to carry out array mode, the input observed quantity of Kalman filter is that the output information of inertial navigation SINS and the output information of GPS navigation are subtracted each other.

3. the foundation of the observation equation of TACAN/INS subfilter 2

The linearization measurement equation of the deviation between TACAN observed quantity (containing oblique distance, reverse aximuth and air pressure height value) and INS simulation subsystem state estimator (solving oblique distance, reverse aximuth and height value by location estimation information is counter) is set up under Tacan platform polar coordinate system, as the observation equation of TACAN/INS subfilter 2, specific as follows:

Z ₂(k)＝H ₂(k)X ₂(k)+V ₂(k) (4)

Wherein H ₂=[F _t0 _{3 × 12}] be t _kmoment measurement matrix,

$F_T=\left[\begin{array}{ccc}-\cos \mathrm{\α}& -\sin \mathrm{\α}& 0\\ \frac{1}{m}\sin \mathrm{\α}& \frac{1}{m}\cos \mathrm{\α}& 0\\ 0& 0& 1\end{array}\right]\×\left[\begin{array}{ccc}-\sin B\cos L& -\sin B\sin L& \cos B\\ -\sin L& \cos L& 0\\ -\cos B\cos L& -\cos B\sin L& -\sin B\end{array}\right];$

Wherein α, B and L are respectively the geodesic line reverse angle, the latitude of aircraft and the longitude that are provided by inertial navigation; V ₂k () is t _kthe height error that the oblique distance that moment TACAN measures, reverse aximuth measurement noise vector sum pressure altimeter are measured.

4. the foundation of the state equation of IFDL/INS subfilter 3

IFDL/INS subfilter adopts the adaptive filter method based on Kalman filtering, first the kinematics model of network members is set up, system equation using clocking error model as IFDL/INS subfilter, then set up based on TOA time of arrival, arrival direction DOA, RTT etc. measure the observation model of parameter as observation equation, last according to the crosslinked relational design Relative Navigation filter construction between observed quantity and quantity of state, complete network members position, the real-time estimation of the quantity of states such as time, achieve the high precision Relative Navigation of user member relative to source simultaneously.

Choose sky, northeast geographic coordinate system, definition IFDL/INS integrated navigation error state amount X ₃:

X ₃＝[φ _e,φ _n,φ _u,δλ,δL,δh,δV _e,δV _n,δV _u,cδt,cδf] ^T

Wherein φ _e, φ _n, φ _ufor INS east orientation, north orientation, sky are to platform error angle, δ λ, δ L, δ h are INS longitude, latitude and height error, δ V _e, δ V _n, δ V _ufor INS east orientation, north orientation and sky are to velocity error, c δ t, c δ f is range difference and range difference rate of change.

Set up X ₃error state equation, as the state equation of subfilter 3, specific as follows:

X ₃(k)＝F ₃(k，k-1)X ₃(k-1)+G ₃(k，k-1)W ₃(k) (5)

Wherein F ₃(k, k-1) is t _k-1moment is to t _ktime etching system state-transition matrix, W ₃k () is t _ktime etching system noise vector, G ₃(k, k-1) is t _k-1moment is to t _ktime etching system noise drive matrix; X ₃(k-1) be t _k-1moment carrier aircraft INS errors quantity of state, X ₃k () is t _kmoment carrier aircraft INS errors quantity of state.

5. the foundation of the observation equation of IFDL/INS subfilter 3

The observed quantity of IFDL/INS subfilter 3 is that BA exports height h _bheight h is exported with INS _ithe TOA pseudo range observed quantity that provides of difference, IFDL Relative Navigation user terminal with calculate the difference of distance, difference, the RTT measured value of the DOA angle of pitch that DOA position angle and difference and user's terminal of computer azimuth angle that user's terminal provides provide and the calculating angle of pitch.Concrete observation equation is as follows:

Z ₃(k)＝H ₃(k)X ₃(k)+V ₃(k) (6)

Wherein observed quantity

$Z_3=\left[\begin{array}{c}\mathrm{\Δh}\\ \mathrm{\Δ\ρ}\\ \mathrm{\Δ\β}\\ \mathrm{\Δ\ϵ}\\ \mathrm{c\δt}-\mathrm{c\δ}{t}_s+{\mathrm{\υ}}_{\mathrm{rtt}}\end{array}\right]=\left[\begin{array}{ccccc}\left[\begin{array}{ccc}0& 0& 0\end{array}\right]& \left[\begin{array}{ccc}0& 0& -1\end{array}\right]& \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& 0& 0\\ \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& \left[\begin{array}{ccc}\frac{x_s-x_u}{{\mathrm{\ρ}}_c}& \frac{y_s-y_u}{{\mathrm{\ρ}}_c}& \frac{z_s-z_u}{{\mathrm{\ρ}}_c}\end{array}\right]T_1& \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& 0& 0\\ \left[\begin{array}{ccc}-\frac{Y_s}{r_c^2}& \frac{X_s}{r_c^2}& 0\end{array}\right]T_3& \left[\begin{array}{ccc}-\frac{Y_s}{r_c^2}& \frac{X_s}{r_c^s}& 0\end{array}\right]R_L^b(T_2-A^T{T}_1)& \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& 0& 0\\ \left[\begin{array}{ccc}\frac{X_s{Z}_s}{{\mathrm{\ρ}}_c^2{r}_c}& \frac{Y_s{Z}_s}{{\mathrm{\ρ}}_c^2{r}_c}& -\frac{r_c}{{\mathrm{\ρ}}_c^2}\end{array}\right]T_3& \left[\begin{array}{ccc}\frac{X_s{Z}_s}{{\mathrm{\ρ}}_c^2{r}_c}& \frac{Y_s{Z}_s}{{\mathrm{\ρ}}_c^2{r}_c}& -\frac{r_c}{{\mathrm{\ρ}}_c^2}\end{array}\right]R_L^b(T_2-A^T{T}_1)& \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& 0& 0\\ \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& \left[\begin{array}{ccc}0& 0& 0\end{array}\right]& 1& 0\end{array}\right]X_3+V_3---\left(7\right)$

Wherein (x _uy _uz _u) and (x _sy _sz _s) be the estimation geocentric rectangular coordinate in user and source; (X _sy _sz _s) be the estimated coordinates of source in user's carrier coordinate system,

${\mathrm{\ρ}}_c=\sqrt{{(x_s-x_u)}^2+{(y_s-y_u)}^2+{(y_s-y_u)}^2}---\left(8\right)$

$A=\left[\begin{array}{ccc}-\sin {\mathrm{\λ}}_u& -\cos {\mathrm{\λ}}_u\sin L_u& \cos {\mathrm{\λ}}_u\cos L_u\\ \cos {\mathrm{\λ}}_u& -\sin {\mathrm{\λ}}_u\sin L_u& \sin {\mathrm{\λ}}_u\cos L_u\\ 0& \cos L_u& \sin L_u\end{array}\right]---\left(9\right)$

$T_1=\left[\begin{array}{ccc}-(R_e(1+f{\sin }^2{L}_u)+h_u)\cos L_u\sin {\mathrm{\λ}}_u& 2R_{e}f\sin L_u{\cos }^2{L}_u\cos {\mathrm{\λ}}_u-(R_e(1+f{\sin }^2{L}_u)+h_u)\cos {\mathrm{\λ}}_u{\sin L}_u& \cos L_u\cos {\mathrm{\λ}}_u\\ (R_e(1+f{\sin }^2{L}_u)+h_u)\cos L_u{\cos \mathrm{\λ}}_u& 2R_{e}f\sin L_u{\cos }^2{L}_u\sin {\mathrm{\λ}}_u-(R_e(1+f{\sin }^2{L}_u)+h_u)\sin {\mathrm{\λ}}_u\sin L_u& \cos L_u\sin {\mathrm{\λ}}_u\\ 0& 2R_{e}f{(1-f)}^2{\sin }^2{L}_u\cos L_u+(R_e(1+f{\sin }^2{L}_u){(1-f)}^2+h_u)\cos L_u& \sin L_u\end{array}\right]---\left(10\right)$

$T_2=\left[\begin{array}{cc}-(x_s-x_u)\cos {\mathrm{\λ}}_u-(y_s-y_u)\sin {\mathrm{\λ}}_u& 0\\ (x_s-x_u)\sin {\mathrm{\λ}}_u\sin L_u-(y_s-y_u)\cos {\mathrm{\λ}}_u\sin L_u& -(x_s-x_u)\cos {\mathrm{\λ}}_u\cos L_u-(y_s-y_u)\sin {\mathrm{\λ}}_u\cos L_u-(z_s-z_u)\sin L_u\\ -(x_s-x_u)\sin {\mathrm{\λ}}_u\cos L_u+(y_s-y_u)\cos {\mathrm{\λ}}_u\cos L_u& -(x_s-x_u)\cos {\mathrm{\λ}}_u\sin L_u-(y_s-y_u)\sin {\mathrm{\λ}}_u\sin L_u+(z_s-z_u)\cos L_u\end{array}\right]---\left(11\right)$

$\begin{array}{c}{T}_3=\left[\begin{array}{c}{E}_s\sin r\sin y\cos p+N_s\sin r\cos y\cos p+U_s\sin p\sin r\\ -E_s\sin y\sin p-N_s\cos y\sin p+U_s\cos p\\ -E_s\sin y\cos p\cos r-N_s\cos y\cos p\sin r-U_s\sin p\cos r\end{array}\right.\\ E_s(-\sin r\cos y+\cos r\sin y\sin p)+N_s(\sin r\sin y+\cos r\cos y\sin p)-U_s\cos p\cos r\\ 0\\ E_s(\cos y\cos r+\sin y\sin p\sin r)+N_s(-\sin y\cos r-\cos y\sin p\cos r)-U_s\cos p\sin r\\ \left[\begin{array}{c}{E}_s(\cos r\cos y+\sin r\cos y\sin p)+N_s(-\sin y\cos r-\cos y\sin p\cos r)-U_s\cos p\sin r\\ E_s\cos y\cos p-N_s\sin y\cos p\\ E_s(-\sin y\sin r-\cos y\sin p\cos r)+N_s(-\cos y\sin r+\sin y\sin p\sin r)\end{array}\right.\end{array}---\left(12\right)$

$R_L^b=\left[\begin{array}{ccc}\cos r\cos y+\sin r\sin y\sin p& -\cos r\sin y+\sin r\cos y\sin p& -\cos p\sin r\\ \sin y\cos p& \cos y\cos p& \sin p\\ \cos y\sin r-\sin y\sin p\cos r& -\sin y\sin r-\cos y\sin p\sin r& \cos p\cos r\end{array}\right]---\left(13\right)$

6. the fusion treatment of each subfilter Output rusults

Subfilter common condition Vector Fusion and time renewal is carried out in senior filter, complete global optimum or the suboptimal estimation of each subfilter (GNSS/INS subfilter 1, TACAN/INS subfilter 2 and IFDL/INS subfilter 3) Output rusults and senior filter estimated value, export global optimum's estimator that is accurate, navigational parameter error reliably.According to decentralized global filtering formula, global optimum's estimated result is as follows:

$P^{-1}\left(k\right)=P^{-1}\left(k\right|k-1)+{\mathrm{\Σ}}_{i=1}^N{M}_i^T(P_i^{-1}\left(k\right)-P_i^{-1}\left(k\right|k-1))M_i---\left(14\right)$

Wherein N=1, the number of 2,3 subfilters worked for this moment.

7. the correction process of final fusion results

The evaluated error quantity of state that 6. walk the INS/GNSS/TACAN/IFDL integrated navigation of output is fed back in INS simulation subsystem, and corresponding error correction is fallen.Here adopt remapping method, senior filter finally exported each subfilter is estimated that mean squared error matrix and subfilter quantity of state reset as follows:

Wherein β _i(i=1,2,3) are called " the information distribution factor ".

Consult Fig. 4, Fig. 5.GNSS/INS subfilter 1, TACAN/INS subfilter 2 and IFDL/INS subfilter 3 adopt fuzzy logic adaptive Kalman filter, main thought constantly monitors whether residual error r is zero-mean white noise by fuzzy logic adaptive controller FLAC, then the weighted factor (k) of the estimation error variance of each subfilter is controlled according to fuzzy inference rule, for adjustment process and measurement noises variance matrix Q _kand R _k, thus reach constantly adjustment filter gain K, make it perform the object of optimal estimation always.The average being input as subfilter residual error of FLAC and variance, the output of FLAC is the weighting factor for the system and observation model noise adjusting subfilter.The design process of FLAC comprises fuzzification process, fuzzy control rule generative process and anti fuzzy method process.Fuzzification process is that the average of subfilter residual error and variance are carried out obfuscation, namely carries out domain quantification, fuzzy division, fuzzy expression; Fuzzy control rule generative process generates control law by expertise exactly, then represents in table form; Anti-fuzzy method in the present invention adopts gravity model appoach, is weighted mean thought in essence, and its weighting coefficient gets the degree of membership of corresponding element, and correlation computations formula is:

$u=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\mathrm{\μ}\left(u_i\right)\×u_i}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\mathrm{\μ}\left(u_i\right)}---\left(17\right)$

In formula, u represents the exact value calculated, μ (u _i) represent subordinate function, u _irepresent the element of fuzzy set.

After completing fuzzy rule design according to the method described above, just can be weighted Kalman filtering fusion calculation.If process and measurement noises variance matrix are exponential function:

R _k＝Rα ^-2(k+1),Q _k＝Qα ^-2(k+1)(18)

Wherein α >=1, matrix Q and R is constant simultaneously.For α > 1, along with time k increases, Q, R reduce, and represent that nearest observation information has larger weighting.If α=1, conventional Kalman filtering will be obtained.

Definition weighted filtering one-step prediction Square Error matrix is:

P ^α(k|k-1)＝P(k|k-1)α ^2k(19)

Then FUZZY WEIGHTED Kalman filtering algorithm can be described below:

Residual error wherein in measuring process is

Consult Fig. 6, Fig. 7.In the design of fault-tolerance combined navigation system, adopt integrated navigation fault detection and diagnosis technology FDIR to carry out diagnosis and detection fault in real time to integrated navigation system, when subfilter non-fault, the filter result of subfilter is sent into Federated Filters; After mutation failure being detected, subfilter will be isolated, and the filter result of subfilter can not send into Federated Filters.Here system has the function of automatic separating fault navigation sources, such as when GNSS information is blocked or disturb can not the used time, fault diagnosis detecting unit can detect automatically, and GNSS subsystem is isolated from Generalized Federated Filters, utilize normal-sub navigation sources signal reconstruct system globe area algorithm.INS/GNSS/TACAN/IFDL integrated navigation fault detection and diagnosis FDIR mainly comprises state residual χ ²inspection and Fuzzy ArtMap fuzzy neural network two parts, wherein use state residual χ ²method of inspection judges data validity, in order to identify fault and reconstruct, needing to position fault, coming Isolate Problem source, failure judgement amplitude and time of origin by the recognition capability of Fuzzy ArtMap fuzzy neural network.

1. fault detection algorithm (state residual χ ²method of inspection)

In the design of fault-tolerance combined navigation system, utilization state residual error χ ²method of inspection carries out detecting and the mutation failure of isolating subsystem in some subfilter.Calculate fault detect function lambda (k):

${\mathrm{\λ}}_i\left(k\right)=r_i^T\left(k\right)U_i^{-1}\left(k\right)r_i\left(k\right)---\left(22\right)$

Wherein r _ik () represents by observed reading at z _ik estimated value that () obtains through Kalman filtering with the predicted value obtained according to prior imformation recursion by state propagator difference.Residual error and residual variance are respectively:

Wherein p _i(k|k) the optimal State Estimation value of i-th subsystem and the error covariance matrix of its correspondence is respectively; p _sthe estimated value that k state propagator that () is respectively i-th subsystem obtains and error covariance matrix thereof.λ _ik () function obeys degree of freedom is the χ of m ²distribution, λ _i(k) ~ χ ²m (), m is the dimension of state X.The criterion of breakdown judge is for working as time, think that system works is normal; When time, then think that system has occurred fault (wherein be called detection threshold, α is called alert rate by mistake).

2. fuzzy neural network Fault Identification and separation algorithm

As state residual χ ²after method of inspection detects fault, in order to identify fault and reconstruct, need to position fault.Because the state estimation of different faults to Kalman filter of the airborne navigation sources of difference has Different Effects, detect fault moment k by analyzing _bstate χ ²the result of inspection carrys out localizing faults.At k _bmoment, when detecting fault, incites somebody to action now χ ²the result of inspection is input to Fuzzy ArtMap1 network as feature mode after obfuscation, by net mate to the corresponding source of trouble, thus determines fault category.For System recover, only identify that out of order sub-navigation sources is inadequate, preferably can to the amplitude that moment and fault occur that is out of order, this must be completed by the identification of Fuzzy ArtMap2 network failure.

For INS/GNSS, identify by the fault of Fuzzy ArtMap method to INS/GNSS integrated navigation and be separated.The state variable X of integrated navigation has 15, is the error term of inertial navigation, refers to step 1.The step making online classification by Fuzzy ArtMap1 neural network is as follows:

1. as shown in table 1 for fault type set by INS/GNSS subfilter:

Table 1 fault verification type

Abort situation	Fault size
		X-axis gyroscope .Deg/h	0.5，1.0，2.0，3.0，4.0，5.0
Y-axis gyroscope .Deg/h	0.5，1.0，2.0，3.0，4.0，5.0
		X-axis accelerometer. μ g	500，600，700，800，900，1000
Y-axis accelerometer. μ g	500，600，700，800，900，1000
		GNSS receiver saltus step	10，20，50，70，100，150

2. emulation produces the state χ of the different amplitude faults in different navigation source ²assay, is carried out Fuzzy processing, obtains the degree of membership numerical value between [0,1], and namely this numerical value can be used as the input pattern sample a of Fuzzy ArtMap1 neural network _i, obfuscation formula is

<math><math display = 'block'> <mrow> <msub> <mi>a</mi> <mi>i</mi> </msub> <mo>=</mo> <mo>&amp;lsqb;</mo> <msub> <mi>&amp;lambda;</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>&amp;minus;</mo> <munder> <mi>min</mi> <mi>i</mi> </munder> <mrow> <mo>(</mo> <msub> <mi>&amp;lambda;</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>&amp;rsqb;</mo> <mo>/</mo> <mo>&amp;lsqb;</mo> <msub> <mi>&amp;lambda;</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>&amp;minus;</mo> <munder> <mi>min</mi> <mi>i</mi> </munder> <mrow> <mo>(</mo> <msub> <mi>&amp;lambda;</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>&amp;rsqb;</mo> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1,2</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <mi>n</mi> <mo>&amp;minus;</mo> <mo>&amp;minus;</mo> <mo>&amp;minus;</mo> <mrow> <mo>(</mo> <mn>25</mn> <mo>)</mo> </mrow> </mrow></math>

Wherein with be respectively and ask λ _ik the minimum value of () and maximal value, n is the dimension of state variable.

3.Fuzzy ArtMap1 tests input amendment ai.The preprocessing field F of Fuzzy ArtMap1 ₀ ^abe 15 dimensions, 15 dimension state variables of correspondence system; Input is than competiting place F ₁ ^ait is 15 × 2=30 dimension; Output category field F ₂ ^abe 5 dimensions, corresponding 5 class fault categories.Fuzzy ArtMap1 network parameter is arranged: Selection parameter α=0.01, Vigilance parameter ρ=0.8, learning rate is β=0.01,5 class faults correctly can be classified through Fuzzy ArtMap1.

4.Fuzzy ArtMap2 network carries out fault parameter identification.The classification results of Fuzzy ArtMap1 is sent into the estimation that Fuzzy ArtMap2 network carries out fault parameter.Method is as state residual χ ²method of inspection is at k _bmoment, when detecting sub-navigation sources fault, Fuzzy ArtMap2 network chose state residual χ corresponding with it ²it is the data window of 60 that test statistics λ (k) constructs a length, sets up to the different faults amplitude of this source of trouble the state χ that data window is 60 ²test statistics λ (k) value sequence.Using λ (k) value sequence in data window as ART in Fuzzy ArtMap2 network _athe input of module, simultaneously by the fault amplitude type coding of correspondence input ART _bmodule, ART _athe input dimension of module is 60, ART _bmodule input dimension depend on fault amplitudes type number, general direct use 5 binary codings are distinguished fault, ART _bmodule input dimension is 5.Fuzzy just can carry out fault parameter identification to this sensor fault after ArtMap2 network training.

Consult Fig. 8.INS navigation simulation subsystem software receives the ideal trajectory data (comprising position, speed, angular velocity, acceleration and attitude) of the aircraft of peripheral environment, first by above-mentioned data message, from sky, northeast, geographical ordinate transform, to body axis system, is used as the nominal value of accelerometer emulation acceleration and gyroscope emulation angular velocity under body axis system; Then the random walk model error of constant error, first-order Markov process is set up according to inertia type instrument error characteristics, for analog acceleration meter and gyrostatic measuring error; Again the error amount that above-mentioned nominal value and simulation produce is added, obtains INS in body axis system analogue measurement data, input to INS computing unit; Calculate position, the velocity information of the calculating of attitude matrix T and carrier aircraft finally by coordinate transform and INS, export navigation information.

Consult Fig. 9.GNSS simulation subsystem software function module mainly contain GNSS constellation emulation module, systematic error emulation module, visible star judge module, receiver emulation module and rear end state resolve module.Wherein constellation emulation module is the position and the speed that adopt Orbit simulation model and corresponding BD2, GPS constellation ephemeris parameter to calculate satellite; Systematic error emulation module establishes star clock error model, ephemeris error model, tropospheric error model and ionospheric error model; Visible star judge module mainly judges whether selected constellation meets the minimum elevations restriction that user points to the vector of satellite; Receiver emulation module and state computation module are the navigation algorithms with reference to actual GNSS receiver, select four satellites to carry out navigation calculation or utilize all visible stars to adopt least-squares algorithm to carry out navigation calculation.

Consult Figure 10.Resolve in workflow diagram in the information of TACAN simulation subsystem, when radio navigation mode is TACAN navigation, TACAN function on, after receiving the corresponding scheduling parameter of TACAN, carries out absolutely empty pattern and the switching of air-ground pattern according to scheduling parameter information; Absolutely empty pattern and air-ground pattern have R and R/T two spermotype, the former only realizes location process, provides between aircraft or magnetic north position angle between aircraft and surface beacon platform, and the latter realizes location and range finding process simultaneously, namely, except providing magnetic north position angle, oblique distance size is given.

Consult Figure 11.(airfield approach way selection MLS landing approach is applicable to) in MLS+DME/P functional software process flow diagram, MLS functional software receives from peripheral environment data, specifically comprises land station (containing DME/P platform, Azimuth Station and pitching platform) information, airplane information and weather information; First judge the state of undercarriage, if undercarriage is the state put down, then enters subsequent treatment module, otherwise return invalid data, show to be not suitable for carrying out landing maneuver; When undercarriage is in down state, after running parameter maps, enter antenna function processing unit, this unit mainly completes following operation:

(1) carry out radiation coverage, namely utilize aircraft and both land stations positional information calculation to go out position angle and the angle of pitch of the former relative the latter, then judge whether aircraft falls within beam area according to antenna beamwidth and azimuth pitch angle;

(2) if aircraft falls into the coverage of antenna, then read the altitude figures of land station and aircraft region, calculate the intervisibility situation between them;

(3) if can intervisibility between land station and aircraft, then according to the distance between aircraft and land station and given weather information, Free propagation loss and the aerial loss of signal be calculated;

(4) the different yield value in all directions is calculated by reading antenna pattern data;

(5) calculate the received radiation power arrived on uphole equipment direction, and calculate corresponding sensitivity according to ground parameter, contrast with the sensitivity of Current terrestrial receiver, judge whether to receive carrier aircraft radiation data;

(6) if can receive, then start to enter MLS functional software module, otherwise carry out next data receiver clapped.For MLS functional software module, it receives navigation information and land station's data message of each aircraft platform, calculates the position angle of aircraft relative to Azimuth Station respectively, and aircraft is relative to the angle of pitch of pitching platform, and aircraft is relative to the range information of DME/P platform; Then according to DME/P distance error model and MLS angular error model, above distance, position angle and angle of pitch information are added error; Again by RGS coordinate conversion, obtain the distance azimuth pitch information of the relative landing point of aircraft; The angle-data of this information and angle reference value are done difference, produces the angu-lar deviation relative to preset position angle and the preset angle of pitch; Finally the navigational parameter of conversion longitudinal separation, orientation and pitch deviation is given aobvious control unit to show.

Consult Figure 12.Resolve in workflow diagram in ALT height measurements, ALT functional software obtains carrier aircraft sea level elevation, carrier aircraft landform floor level, the carrier aircraft angle of pitch (β) and roll angle (γ) from peripheral environment data; First carrier aircraft sea level elevation and carrier aircraft landform floor level are subtracted each other true altitude value (h) obtaining ALT; When 1500≤h≤6000 (rice), during beta, gamma≤40 °, the Gauss distribution random numbers of the error amount of ALT measuring height to be standard deviation be h/300; When h≤1500 (rice), during beta, gamma≤40 °, the Gauss distribution random numbers that the error amount of ALT measuring height is mean square deviation is h/200; The error that ALT measures is carried out sliding window smothing filtering along time shaft, the error amount after smothing filtering is added in h, obtains ALT measured value.

Consult Figure 13.Resolve in workflow diagram at IFDL measured value, first system carries out optimum configurations and all initialization of variable; Then travel through member's aircrafts all in formation networking, judge whether all member's flight paths are formed, if do not formed, then produce the real trace data of member's aircraft; After the track of all member's aircrafts is all formed, the highest member of geographical positional precision is selected to be Navigation Control person (source namely in process flow diagram), also as time reference member; Following process is done to other member's aircrafts all except source:

(1) distance of source in each member's aircraft carrier ball coordinate system, position angle and the angle of pitch is calculated;

(2) according to each member in advance Schedule calculate its current time time unit (epoch), time frame (frame), time slot (slot);

(3) pseudo range measurement TOA value method is utilized to obtain TOA time of arrival that P message is broadcast in source in Relative Navigation;

(4) the DOA value of P message is broadcast in the source in Relative Navigation unit that calculates;

(5) according to time unit, time frame, slot requirements, carry out reception and the transmission of TOA and DOA information.

Claims (10)

1. airborne many navigation sources integrated navigation analogue system, comprise the fault-tolerance combined navigation system based on INS/GNSS/TACAN/IFDL, dispatching system, radio altimeter ALT simulation subsystem, microwave landing MLS simulation subsystem, instrument landing ILS simulation subsystem, micro-distancer DME/P simulation subsystem, single navigational system such as chain IFDL between machine, it is characterized in that: adopt the stratification that line strainer model combines with the filtrator mode based on software component, modular Distributed Simulation Architecture, each single navigation simulation subsystem emulates the function identical with model realistically according to self navigation principle, performance and interface, the aobvious control instruction that dispatching system response task system issues, load instructions and loading data message, independently select single navigation sources or integrated navigation to carry out navigator fix, and combining environmental emulated data drives various navigation simulation subsystem to perform simulation calculating, integrated navigation system is using INS simulation subsystem as main navigation sources, respectively with GNSS, TACAN, IFDL is as assisting navigation source, state vector is INS navigational parameter error, the Generalized Federated Kalman filter adopting Federated Filtering to build two-layer configuration completes information fusion, senior filter carries out total information fusion, and information distribution and replacement are carried out to each subfilter, GNSS/INS subfilter 1, TACAN/INS subfilter 2 and IFDL/INS subfilter 3 complete the state vector local estimation of INS/GNSS/TACAN/IFDL integrated navigation system respectively, senior filter adopts multidate information partition factor to carry out optimal estimation to the result of subfilter, in fault-tolerance combined navigation system, adopt state residual χ ²method of inspection carries out diagnosis and detection fault in real time to each subfilter, fuzzy logic adaptive controller FLAC constantly monitors whether residual error r is zero-mean white noise, then controls the weighted factor (k) of estimation error variance of each subfilter according to fuzzy inference rule, for adjustment and measurement noises variance matrix Q _kand R _k, finally according to modern control theory, antithetical phrase navigation sources INS implements output calibration or close-loop feedback corrects.

2. airborne many navigation sources integrated navigation system as claimed in claim 1, it is characterized in that, INS simulation subsystem combined inertia site error characteristic, establishes the random walk model of constant error, first-order Markov process, the measurement data of realistic simulation gyroscope and accelerometer.

3. airborne many navigation sources integrated navigation system as claimed in claim 1, it is characterized in that, the navigation mode that TACAN simulation subsystem can be arranged according to emulation, completes aircraft relative to TACAN surface beacon platform orientation, oblique distance, speed, emulate to the parametric degree of platform time to be flown.

4. airborne many navigation sources integrated navigation system as claimed in claim 3, it is characterized in that, MLS/ILS simulation subsystem and DEM P with the use of, the azimuth information of simulated aircraft relative to orientation, ground control desk, the pitch information relative to ground pitch control subsystem platform and aircraft are relative to the range information of DME/P platform; Then on this basis, rotational coordinates Iterative algorithm RGS is adopted to obtain the real-time position information of aircraft relative to landing point.

5. airborne many navigation sources integrated navigation system as claimed in claim 1, it is characterized in that, IFDL Relative Navigation simulation subsystem completes two or more formation member composition relative navigation system function in IFDL net, on the time division multiple access communication function i ntegration of IFDL system, time of arrival TOA and arrival direction DOA is utilized to carry out range finding and interaction locations data, the source utilizing active Time transfer receiver RTT to obtain and the clock correction of user clock, then carry out certain data processing, make member obtain the coordinate position in the relative coordinate system of regulation.

6. airborne many navigation sources integrated navigation system as claimed in claim 1, is characterized in that, each subfilter all adopts the state vector local estimation of fuzzy logic adaptive Kalman filter completion system, constantly adjustment subfilter gain K; Senior filter carries out the synthesis of common condition Vector Fusion and the time renewal of subfilter.

7. airborne many navigation sources integrated navigation system as claimed in claim 1, is characterized in that, when subfilter non-fault, the filter result of subfilter sends into Federated Filters; After mutation failure being detected, subfilter by automatism isolation, will utilize other normal-sub navigation sources signal reconstruct system globe area algorithm.

8. airborne many navigation sources integrated navigation system as claimed in claim 1, it is characterized in that, IFDL/INS subfilter adopts the adaptive filter method based on Kalman filtering, first the kinematics model of network members is set up, system equation using clocking error model as IFDL/INS subfilter, then set up based on TOA time of arrival, arrival direction DOA, RTT etc. measure the observation model of parameter as observation equation, last according to the crosslinked relational design Relative Navigation filter construction between observed quantity and quantity of state, complete network members position, the real-time estimation of the quantity of states such as time, achieve the high precision Relative Navigation of user member relative to source simultaneously.

9. airborne many navigation sources integrated navigation system as claimed in claim 1, is characterized in that, the observed quantity of IFDL/INS subfilter 3 is that BA exports height h _bheight h is exported with INS _ithe TOA pseudo range observed quantity that provides of difference, IFDL Relative Navigation user terminal with calculate the difference of distance, difference, the RTT measured value of the DOA angle of pitch that DOA position angle and difference and user's terminal of computer azimuth angle that user's terminal provides provide and the calculating angle of pitch.

10. airborne many navigation sources integrated navigation system as claimed in claim 1, is characterized in that, as state residual χ ²method of inspection is at k _bmoment, when detecting sub-navigation sources fault, Fuzzy ArtMap2 network chose state residual χ corresponding with it ²it is the data window of 60 that test statistics λ (k) constructs a length, sets up to the different faults amplitude of this source of trouble the state χ that data window is 60 ²test statistics λ (k) value sequence; Using λ (k) value sequence in data window as ART in Fuzzy ArtMap2 network _athe input of module, simultaneously by the fault amplitude type coding of correspondence input ART _bmodule, ART _athe input dimension of module is 60, ART _bmodule input dimension depend on fault amplitudes type number, general direct use 5 binary codings are distinguished fault, ART _bmodule input dimension is 5.