CN104316947B - GNSS/INS ultra-tight combination navigation apparatus and relative navigation system thereof - Google Patents

Abstract

The invention discloses a global navigation satellite system (GNSS)/inertial/navigation system (INS) ultra-tight combination navigation apparatus and a relative navigation system thereof. On the basis of the traditional GNSS/INS ultra-tight combination navigation apparatus, the provided GNSS/INS ultra-tight combination navigation apparatus uses a distributed filtering structure. Slave filters are used for carrying out pretreatment on I,Q signals of all satellite channels; and the processed information is sent to a combined filter. Meanwhile, the GNSS/INS ultra-tight combination navigation apparatus is also applied to a relative navigation system; while the navigation state and the inertia measurement of a carrier itself are received, navigation information and a carrier phase of another carrier are also received; with a dual-different carrier phase and inertia measurement, recursion filtering resolving is carried out by using an optimal kalman filter and an optimal relative position and a relative speed between each two carriers among multiple carriers are estimated.

Description

GNSS/INS hypercompact combination navigations device and relative navigation system

Technical field

The invention belongs to navigation field, more says it is GNSS/INS hypercompact combination navigations device and relative navigation system.

Background technology

Global navigation satellite system GNSS (Global Navigation Satellite System), inertial navigation system INS(InertialNavigationSystem)；GNSS/INS hypercompact combination navigations are the study hotspots of integrated navigation in recent years. Satellite navigation and inertial navigation have complementary advantage, and both compound modes can efficiently solve satellite navigation signals by geographical position Put and environmental disturbances, and inertial navigation system measurement error with time cumulation the problems such as.GNSS/INS compound modes have loose group Conjunction, tight integration and hypercompact compound mode.Wherein, hypercompact compound mode is the study hotspot in the field.This mode make use of inertia Auxiliary of the navigation information to GNSS signal acquisition and tracking, realize GNSS information tracking parameter and inertial navigation Position And Velocity it Between information fusion.Under complex environment, the hypercompact combined systems of GNSS/INS have stronger capacity of resisting disturbance, can be complexity Motion carrier under situation, there is provided the critical function such as accurate and stable navigator fix, attitude determination and inertia alignment.

At present GNSS/INS hypercompact combination navigation filter structure majorities belong to centralized data processing.Generally I, Q signal Frequency is in MHz magnitudes, and inertia is exported in 100Hz magnitudes, and centralized processing directly focuses on two kinds of information, intractability Greatly, it is difficult to realize.

Distributed navigation system is a kind of new Navigation System Design theory.Distributed navigation is to see in multiple navigation carriers Into a navigation system, by navigation system between multiple carriers navigation information be exchanged with each other with information sharing it is whole to improve The navigation accuracy of individual system, while the failure tolerant level of the system of raising.Distributed navigation system can be divided into centralized configuration And distributed frame.Lead-Follower structures are a kind of common centralized configurations, and its thought is to make in certain navigation carrier For navigation terminal of leading a group, other conducts follow navigation carrier, the information between each navigation terminal be it is visible upwards, downwards not It is visible.Distributed frame each navigation carrier exists together a rank, and the navigation information of each carrier is sent to Data processing It is uniformly processed.Distributed navigation is in a group of planes is coordinated to form into columns using widely.

Relative navigation system is mainly used in the relative of satellite formation flying, Autonomous rendezvous and docking task and group of planes collaboration formation The acquisition of navigation information.At present, the relative navigation system of view-based access control model is usually used in satellite Autonomous spacecrafts rendezvous task.

Relative navigation system is the key technology that a group of planes cooperates with formation flight.Relative Navigation information between flight carrier Acquisition is to ensure the essential condition that Formation Flying keeps and converts.Its precision also can produce very big to formation flight effect Affect.Rely on satellite navigation system more traditional Relative Navigation and autonomy is poor.Satellite navigation signals easily receive external interference, Reliability cannot be guaranteed, once satellite-signal is interfered, Relative Navigation precise decreasing is quickly.How Relative Navigation is improved Precision strengthen flight autonomy simultaneously and stability be current Relative Navigation research difficult point.

The content of the invention

Belong to centralized data processing to solve current GNSS/INS hypercompact combination navigations filter structure majority, to data Processing speed requirement is very high, and system complex realizes the big problem of difficulty, the invention provides a kind of hypercompact combinations of GNSS/INS are led Boat device, this is installed on Distributed filtering structure to realize, including GNSS radio-frequency front-ends, Inertial Measurement Unit, tracking ring Road unit and junction filter；Junction filter is connected respectively with Inertial Measurement Unit, tracking loop unit, tracking Loop unit is also connected with GNSS radio-frequency front-ends；

Wherein, tracking loop unit includes several visible satellite passages, each visible satellite passage correspondence one Integral Processing unit and from wave filter；GNSS radio-frequency front-ends are received and processed and send data to after satellite signal data tracking and catch Loop unit is obtained, the I of each visible satellite passage, Q signal are at corresponding Integral Processing unit in tracking loop unit Send after reason to corresponding carries out the state vector that pretreatment obtains from wave filter from wave filter；Junction filter receives inertia and surveys The metrical information and the state vector from wave filter of amount unit, using the information for receiving the carrier state information and load are estimated Wave phase simultaneously generates loop control instruction tracking loop unit is controlled, wherein the carrier state information includes position Put, speed and attitude.

Further preferred version, current, super from filter process in GNSS/INS hypercompact combination navigations device of the present invention Front and delayed I, Q signal；

From state vector X of wave filter_i=[δ w_i δφ_i δt_i M_i]^T, wherein δ w_iFor estimating carrier frequencies error, δ φ_i It is carrier phase estimation difference, δ t_iFor phase estimator error, M_iFor the carrier amplitude of signal, i be from wave filter is corresponding can See satellite channel number, i=1,2 ... N, N are visible satellite port number, []^TRepresenting matrix transposition；

The observed quantity Z from wave filter_iFor current I, Q signal value, advanced I in integration period T, Q signal value and delayed I, Q believe Number value, be expressed as：

Wherein, T₀For the initial time of integration period, ε is that receiver produces the advanced and hysteresis between of pseudo noise code Time migration.

Further preferred version, in GNSS/INS hypercompact combination navigations device of the present invention, the state of junction filter to Amount

Wherein, δ x are site error state, and δ v are velocity error state,For attitude error state, δ w are that gyroscope is missed Difference attitude vectors, δ a are acceleration error state vector；δt_uFor the clock jitter of receiver, δ t_rBe receiver clock it is random Drift error amount；

The measurement vector of junction filter is：

Wherein, δ ρ_iWithIt is respectively the difference and the difference of pseudorange rates of the pseudorange of GNSS and INS outputs.

Simultaneously the present invention in order to solve to rely on traditional Relative Navigation satellite navigation system and autonomy is poor asks more Topic, present invention also offers a kind of relative navigation system based on GNSS/INS hypercompact combination navigation devices, the system is used to estimate Optimum relative position and relative velocity in meter multichip carrier between each two carrier, it is characterised in that each carrier correspondence one GNSS/INS hypercompact combination navigation devices, data transmission unit and the data processing unit being sequentially connected；

Wherein, data transmission unit includes wire transmission and is wirelessly transferred；Data processing unit includes that what is be connected with each other moves States model construction unit and Relative Navigation wave filter；

The metrical information of Inertial Measurement Unit in GNSS/INS hypercompact combination navigation devices, from filter status vector Carrier state information that carrier wave metrical information and junction filter are estimated, the wire transmission of carrier phase Jing are corresponding to the carrier In dynamic model construction unit；The carrier state information and carrier phase also Jing that simultaneously junction filter is estimated is wirelessly transmitted to In the dynamic model construction unit of other carriers；

The corresponding dynamic model construction unit of each carrier is using wire transmission and is wirelessly transferred the information of acquisition determining Dynamic model is built after the quantity of state of dynamic model and observed quantity；According to dynamic model, Relative Navigation wave filter utilizes Kalman Optimal filter recurrence equation group, is calculated the optimal estimation value of error state, relative position to two carriers and relatively speed Degree compensates and then obtains the relative position and relative velocity of optimum.

Further preferred version, in relative navigation system of the present invention, the quantity of state of dynamic model and observed quantity are as follows：

The quantity of state of dynamic model is x_GPSAnd x_INS

x_INS=[δ Δ v δ Δ x δ w_m δa_m]^T；

In formula, Δ x is the relative position of two carriers,For double difference integer ambiguity, Δ m_φSingle poor carrier phase is more Tracking error, δ T are tropospheric error；δ I are ionospheric error；δ Δ v are the relative velocity error state of two carriers, δ Δ x For the relative position error state of two carriers, δ w_mFor carrier m gyro error attitude vectors；δa_mFor carrier m acceleration errors State vector, subscript m is carrier label；

The observed quantity of dynamic model is z=[Δ v Δ x]^T, Δ v is the relative velocity of two carriers；Δ x is two loads The relative position of body.

The present invention has compared with prior art following significant advantage：

(1) present invention adopts Distributed filtering structure first with, to each satellite channel I, Q signal carries out pre- from wave filter Process, then will send to junction filter from the pretreated information of wave filter, it is to avoid centralized data processing to data at Reason rate request is very high, and system complex realizes the big problem of difficulty；

(2) in based on the hypercompact combination relative navigation systems of GNSS/INS, GNSS/INS hypercompact combination navigation modes are make use of, Both advantages are fully combined, information fusion mode is deep into into track loop I, the level of Q signal strengthens GNSS receiver Capture and ability of tracking to satellite-signal, and then improve the anti-dynamic and capacity of resisting disturbance of system.Inertia survey is avoided simultaneously The problems such as amount error is with time cumulation；(3) relative navigation system of GNSS/INS hypercompact combination navigation devices of the invention passes through It is wirelessly transferred and the low-rate data (carrier phase and carrier navigation information) of each carrier is transmitted to the data processing unit of carrier Shared, the enhancement information redundancy of information between carrier is completed, and then improves estimated accuracy.

Description of the drawings

Fig. 1 is the GNSS/INS hypercompact combination navigation schematic diagram of device of the present invention；

Fig. 2 is the relative navigation system schematic diagram based on GNSS/INS hypercompact combination navigation devices of the present invention；

Fig. 3 is the workflow diagram of the data processing unit of the present invention；

With reference to the accompanying drawings and detailed description the present invention is described in further detail；

Specific embodiment

As shown in figure 1, a kind of GNSS/INS hypercompact combination navigations device of the invention, the device is based on Distributed filtering structure To realize, including GPS radio frequency front end, Inertial Measurement Unit, tracking loop unit and junction filter；The tracking Loop unit includes N number of visible satellite passage, and each visible satellite passage corresponds to an Integral Processing unit and from wave filter； GPS radio frequency front end receiver satellite-signal, and power amplification, frequency conversion and digital-to-analogue conversion are carried out, by the data is activation after conversion To tracking loop unit, the I of each visible satellite passage, Q signal are at corresponding integration in tracking loop unit Send after reason cell processing to corresponding carries out the state vector that pretreatment obtains from wave filter from wave filter；Junction filter connects The metrical information and the state vector from wave filter of Inertial Measurement Unit are received, using above- mentioned information the carrier state information is estimated With carrier phase and generate loop control instruction tracking loop unit is controlled.

Continuous position that GNSS system is provided, velocity information can aid in the initial alignment of INS systems, error estimation, Compensation and correction etc..Meanwhile, GNSS system is also by the auxiliary of INS when obtaining next in combination, and then effectively improves dynamic environment Capture and ability of tracking of the lower GNSS receiver to satellite-signal, and can guarantee that poor in satellite distribution situation or visible star number amount is few In the case of keep higher positioning accuracy and reliability.GNSS radio-frequency front-ends are nursed one's health the satellite-signal for collecting, frequency conversion And digitized.Tracking loop is responsible for the capture and tracking of signal.The hypercompact combined filter structures of GNSS/INS include combination filter Ripple device and from wave filter two parts.Junction filter is based on pseudorange, pseudorange rates and carrier phase level, should from responsible from filtering Before, the I of lead and lag, Q signal.Junction filter completes the optimum fusion output navigation letter to inertial navigation information and satellite information Breath, while sending loop control instruction to acquisition and tracking loop.

From the handling process of wave filter：I after each passage, Q signal it is integrated enter from wave filter, make each passage from State vector X of wave filter_i=[δ w_i δφ_i δt_i M_i]^T, wherein δ w_iFor estimating carrier frequencies error, δ φ_iIt is carrier phase Estimation difference, δ t_iFor phase estimator error, M_iFor the carrier amplitude of signal, i is visible satellite passage corresponding from wave filter Number, i=1,2 ... N, N be visible satellite port number, []^TRepresenting matrix transposition；

The observed quantity from wave filter is current I, Q signal value, advanced I in integration period T, Q signal value and delayed I, Q signal Value, is represented by：

Wherein, T₀For the initial time of integration period, T is integration period, ε be receiver produce the advanced of pseudo noise code with Time migration between hysteresis；

Therefore, it is from the state and measurement equation of wave filter in Distributed filtering structure：

Wherein, X_iFor quantity of state, Z_iFor observed quantity, A_i(k, k-1) represent from the k-1 moment to the state at k moment transfer away from Battle array；W_i(k-1) it is the dynamic noise at k-1 moment, C_iK () represents k moment observed differential matrixes, V_iK () is made an uproar for the observation at k moment Sound, k is the time quantum after sliding-model control, W in formula_i(k-1)、A_i(k,k-1)、V_i(k)、C_iK the specific design process of () is Prior art, referring specifically to document：.GPS satellite navigations positioning principle and method [M] more than Liu Ji. Beijing：Science Press .2003.

The handling process of junction filter：The state vector of junction filter increased on the basis of inertial navigation quantity of state is obtained from The error state estimation of wave filter, i.e.,：

Wherein, δ x, δ v andRespectively position, speed and attitude error state, δ w are gyro error attitude vectors, δ a For acceleration error state vector；δt_uFor the clock jitter of receiver, δ t_rIt is the clock Random Drift Error amount of receiver.

The measurement vector being made up of pseudorange error and pseudorange rates error is：

Wherein, δ ρ_iWithIt is respectively the difference and the difference of pseudorange rates of the pseudorange of GNSS and INS outputs, subscript N is defended for visible Star port number.

The state and observational equation of junction filter is represented by：

Wherein, X is quantity of state, and Z is observed quantity, and W (k-1) is the dynamic noise at k-1 moment, and V (k) is the observation at k moment Noise, k is the time quantum after sliding-model control, and C (k) represents k moment observed differential matrixes, and A (k, k-1) was represented from the k-1 moment To the state transfer matrix at k moment, W (k-1), V (k), C (k), the specific design process of A (k, k-1) are prior art, specifically Referring to document：.GPS satellite navigations positioning principle and method [M] more than Liu Ji. Beijing：Science Press .2003.

As shown in Fig. 2 the relative navigation system based on GNSS/INS hypercompact combination navigation devices, including carrier 1 and carrier 2, the corresponding GNSS radio-frequency front-ends 11 of carrier 1, Inertial Measurement Unit 12, tracking loop unit 13, junction filter 14, data Transmission unit 15 and data processing unit 16；Wherein, junction filter 14 respectively with tracking loop unit 13, inertia measurement Unit 12, data transmission unit 15 are connected, and tracking loop unit 13 is also connected with GNSS radio-frequency front-ends 11, data transfer list Unit 15 is also connected with data processing unit 16；Data transmission unit 15 includes wire transmission 151 and is wirelessly transferred 152；At data Reason unit 16 includes connected model construction unit 161 and Relative Navigation wave filter 162；

The corresponding GNSS radio-frequency front-ends 21 of carrier 2, Inertial Measurement Unit 22, tracking loop unit 23, junction filter 24th, data transmission unit 25 and data processing unit 26；Wherein, junction filter 24 respectively with tracking loop unit 23, Inertial Measurement Unit 22, data transmission unit 25 are connected, and tracking loop unit 23 is also connected with GNSS radio-frequency front-ends 21, number Also it is connected with data processing unit 26 according to transmission unit 25；Data transmission unit 25 includes wire transmission 251 and is wirelessly transferred 252；Data processing unit 26 includes connected model construction unit 261 and Relative Navigation wave filter 262；

GNSS radio-frequency front-ends 11 are received and processed and send data to after satellite signal data tracking loop unit, with The I of each visible satellite passage, Q signal send to right after corresponding Integral Processing cell processing in track Capture Circle unit 13 That what is answered carries out the state vector that pretreatment obtains from wave filter from wave filter；

Junction filter 14 receives the metrical information of Inertial Measurement Unit：Acceleration a₁, angular velocity omega₁With from wave filter State vector, using above- mentioned information the position x of carrier 1 is estimated₁, speed v₁, attitude and carrier phase phi₁And generate loop control System instruction is controlled to tracking loop unit 13；

GNSS radio-frequency front-ends 21 are received and processed and send data to after satellite signal data tracking loop unit, with The I of each visible satellite passage, Q signal send to right after corresponding Integral Processing cell processing in track Capture Circle unit 23 That what is answered carries out the state vector that pretreatment obtains from wave filter from wave filter；

Junction filter 24 receives the metrical information of Inertial Measurement Unit：Acceleration a₂, angular velocity omega₂With from wave filter State vector, using above- mentioned information the position x of carrier 2 is estimated₂, speed v₂, attitude and carrier phase phi₂And generate loop control System instruction is controlled to tracking loop unit 23；

The metrical information of Inertial Measurement Unit 11, the carrier wave metrical information from the state vector of wave filter and combined filter The status information of carrier 1 and carrier phase phi that device 14 is estimated₁Jing wire transmission 151 is to the measurement module construction unit 161 of carrier 1 In；Simultaneously junction filter 14 will also estimate the position x of carrier 1₁, speed v₁And carrier phase phi₁Jing be wirelessly transferred 152 to In the measurement module construction unit 262 of carrier 2；The metrical information of Inertial Measurement Unit 12, the load from the state vector of wave filter The status information of carrier 2 and carrier phase phi that wave measurement information and junction filter are estimated₂Jing wire transmission 251 is to carrier 2 In measurement module construction unit 261；Simultaneously junction filter 24 will also estimate the position x of carrier 2₂, speed v₂, carrier phase φ₂Jing is wirelessly transferred in the measurement module construction unit 162 of 252 to carrier 1,

Measurement module construction unit 161 is according to by metrical information, the carrier wave measurement letter from the state vector of wave filter Breath, the status information of carrier 1, carrier phase phi₁, the status information of carrier 2 and carrier phase phi₂To determine the quantity of state of dynamic model With dynamic model is built after observed quantity,

Relative Navigation wave filter 162 utilizes Kalman's optimal filter recurrence equation group, is calculated the optimum of error state Estimated value, compensates and then obtains the position of the opposite carrier 2 of carrier 1 of optimum to the relative position and relative velocity of two carriers Put and speed；

Measurement module construction unit 261 is according to by metrical information, the carrier wave measurement letter from the state vector of wave filter Breath, the status information of carrier 2, carrier phase phi₂, the status information of carrier 1 and carrier phase phi₁To determine the quantity of state of dynamic model With dynamic model is built after observed quantity,

Relative Navigation wave filter 262 utilizes Kalman's optimal filter recurrence equation group, is calculated the optimum of error state Estimated value, compensates and then obtains the position of the opposite carrier 1 of carrier 2 of optimum to the relative position and relative velocity of two carriers Put and speed.

As shown in figure 3, the purpose of data processing centre is to carry out error estimation to Relative Navigation and complete Relative Navigation most Excellent output.Position on carrier and speed state are estimated to be sent to the data processing list of each carrier by low-rate data chain Unit, using the Relative Navigation information after each carrier positions and speed state work difference as nominal value.The data processing of destination carrier Unit utilizes GNSS double difference carrier phasesWith the Inertia information on self-vector, optimum is carried out to Relative Navigation error state Estimate, and complete the optimal estimation of Relative Navigation state；

Specifically complete to estimate relative position using double difference carrier phase, completed using acceleration analysis and relative position Estimation to relative velocity.

Below by taking carrier 1 and carrier 2 as an example：Double difference carrier phase is obtained by difference twice：First by carrier 1 and carrier 2 Carrier wave measurement carry out first difference, this process eliminates Satellite clock errors and ephemeris error；Again by first time difference component phase Difference again is carried out to the first difference component of main satellite, receiver clock error, double difference carrier phase is again, eliminatingCan It is expressed as：

Wherein, Δ x=x₁-x₂, x₁And x₂The position of carrier 1 and carrier 2 under ECEF coordinate system (ECEF) is represented respectively Put vector；λ is carrier wavelength；For the coefficient matrix of satellite i,For the coefficient matrix of satellite j；δT^i,jWith δ I^i,jRespectively defend The troposphere residual sum Ionosphere Residual Error of star i and satellite j；It is the double difference integer ambiguity of satellite i and satellite j；For The poor carrier phase Multipath Errors of list of satellite i；For the poor carrier phase Multipath Errors of list of satellite j；For satellite i With the double difference carrier phase measurement noise of satellite j, satellite i and satellite j refers to the corresponding satellite i of carrier 1, carrier 2 is corresponding defends Star j；Here satellite i respective channel number i, satellite j respective channel number j；

Therefore, GNSS estimates that the state vector of relative position is represented by：

In formula,For double difference integer ambiguity, Δ m_φSingle poor carrier phase Multipath Errors, δ T are tropospheric error；δ I is ionospheric error.

Dynamic model after linearisation is represented by general type：

δx_GPSFor the relative position error quantity of state of GNSS, w_GFor state error, Φ_GFor coefficient matrix.

Make v₁And v₂The respectively velocity vector of the relative earth of carrier 1 and carrier 2, then

v₁=v₂+Δv+w_en×Δx (9)

It can be seen from specific force equation：

a₁=a₂+Δa+2w_en×Δv+w_en×Δx+w_en×(w_en×Δx) (10)

a₁=f₁-2w_ie×v₁+g₁ (11)

a₂=f₂-2w_ie×v₂+g₂ (12)

Wherein, subscript 1,2 represent respectively carrier 1 and carrier 2, and Δ v is the relative velocity of two carriers, and Δ x is two loads The relative position of body, a₁For the vector acceleration of the relative earth of carrier 1, a₂For the vector acceleration of the relative earth of carrier 2；f₁For The specific force output under navigational coordinate system of carrier 1；f₂The specific force output under navigation system for carrier 2；g₁Gravity suffered by carrier 1 to Amount, g₂Gravity vector suffered by carrier 2, w_enFor the angular velocity of the relative Department of Geography of navigational coordinate system；w_ieIt is used to relatively for Department of Geography The angular velocity of property system,

Formula (9), (11) and (12) are substituted into (10) and can be obtained,

Formula (12) is carried out into the state that Taylor series expansion is obtained relative position and relative velocity, its dynamic model Quantity of state x_INSFor

x_INS=[δ Δ v δ Δ x δ w₁δa₁]^T (13)

δ Δ v are the relative velocity error state of two carriers, and δ Δ x are the relative position error state of two carriers, δ w₁ For the gyro error attitude vectors of carrier 1；δa₁For the acceleration error state vector of carrier 1；

Dynamic model after linearisation is represented by

Wherein, δ x_INSRelative velocity and site error quantity of state for INS, w_INSFor state error, Φ_IFor corresponding system Matrix number.

Formula (8) and (14) constitute the state equation of Relative Navigation wave filter,

Observation vector z=[Δ v Δ x]^T, the dynamic model of Relative Navigation wave filter is represented by：

Wherein M (k) is k moment nominal trajectorys, and Φ (k, k-1) is represented from the k-1 moment to the state transfer matrix at k moment；C K () represents the observed differential matrix at k moment；V (k) is observing matrix, and w (k-1) is dynamic noise, and x represents quantity of state, and z is represented Observed quantity.Here Φ (k, k-1), C (k), v (k), the specific design process of w (k-1) are determined with reference to .GPS satellite navigations more than Liu Ji Position principle and method [M]. Beijing：Science Press .2003.

According to formula (15) using Kalman's optimal filter recurrence equation group, the error of relative position and relative velocity is carried out Estimate, and then draw the Relative Navigation information of optimum；

P_k,k-1=Φ_k,k-1P_k-1Φ_k,k-1+Q_k-1

By above-mentioned recurrence equation group, the optimal estimation of error state is calculatedAfterwards, to the relative position of two carriers The relative position and relative velocity of optimum are compensated and then obtained with relative velocity, with regard to Kalman's optimal filter recurrence equation Group is prior art, referring to Yu Jixiang, chief editor《Kalman filtering and its application in inertial navigation》, aviation specialized teaching material volume Careful group, 1984.

Claims (4)

1. a kind of GNSS/INS hypercompact combination navigations device, it is characterised in that the device realized based on Distributed filtering structure, Including GNSS radio-frequency front-ends, Inertial Measurement Unit, tracking loop unit and junction filter；Junction filter respectively with it is used Property measuring unit, tracking loop unit be connected, tracking loop unit is also connected with GNSS radio-frequency front-ends；

Wherein, tracking loop unit includes several visible satellite passages, each one integration of visible satellite passage correspondence Processing unit and from wave filter；GNSS radio-frequency front-ends are received and processed and send data to after satellite signal data tracking ring Road unit, the I of each visible satellite passage, Q signal are after corresponding Integral Processing cell processing in tracking loop unit Send to corresponding carries out the state vector that pretreatment obtains from wave filter from wave filter；Junction filter receives inertia measurement list The metrical information and the state vector from wave filter of unit, using the information for receiving carrier state information and carrier phase are estimated And generate loop control instruction tracking loop unit is controlled, wherein the carrier state information includes position, speed Degree and attitude, from filter process is current, lead and lag I, Q signal；

From state vector X of wave filter_i=[δ w_i δφ_i δt_i M_i]^T, wherein δ w_iFor estimating carrier frequencies error, δ φ_iIt is to carry Wave phase estimation difference, δ t_iFor phase estimator error, M_iFor the carrier amplitude of signal, i is visible to defend from wave filter is corresponding Star port number, i=1,2 ... N, N be visible satellite port number, []^TRepresenting matrix transposition；

The observed quantity Z from wave filter_iFor current I, Q signal value, advanced I in integration period T, Q signal value and delayed I, Q signal value, It is expressed as：

$Z_i=\left[\begin{array}{c}\underset{k=T_0}{\overset{T_0+T}{\Σ}}I\\ \underset{k=T_0}{\overset{T_0+T}{\Σ}}Q\\ \underset{k=T_0}{\overset{T_0+T}{\Σ}}\[I\left(\frac{\mathrm{\ϵ}}{2}\right)-I(-\frac{\mathrm{\ϵ}}{2})\]\\ \underset{k=T_0}{\overset{T_0+T}{\Σ}}\[Q\left(\frac{\mathrm{\ϵ}}{2}\right)-Q(-\frac{\mathrm{\ϵ}}{2})\]\end{array}\right]$

Wherein, T₀For the initial time of integration period, ε is the advanced time and hysteresis between that receiver produces pseudo noise code Skew.

2. GNSS/INS hypercompact combination navigations device according to claim 1, it is characterised in that the state of junction filter to Amount

Wherein, δ x are site error state, and δ v are velocity error state,For attitude error state, δ w are gyro error appearance State vector, δ a are acceleration error state vector；δt_uFor the clock jitter of receiver, δ t_rIt is the clock random drift of receiver The margin of error；

The measurement vector Z of junction filter is：

$Z={\left[\begin{array}{ccccccccc}{\mathrm{\δ\ρ}}_1& \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}_1& {\mathrm{\δ\ρ}}_2& \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}_2& {\mathrm{\δ\ρ}}_i& \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}_i& \mathrm{...}& {\mathrm{\δ\ρ}}_N& \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}_N\end{array}\right]}^T$

Wherein, δ ρ_iThe difference of the pseudorange of GNSS and INS outputs is represented,Represent that GNSS and INS exports the difference of pseudorange rates.

3. a kind of relative navigation system based on GNSS/INS hypercompact combination navigations device described in claim 2, the system is used for Estimate optimum relative position and relative velocity between each two carrier in multichip carrier, it is characterised in that each carrier correspondence one Individual GNSS/INS hypercompact combination navigation devices, data transmission unit and the data processing unit being sequentially connected；

Wherein, data transmission unit includes wire transmission and is wirelessly transferred；Data processing unit includes the dynamic analog being connected with each other Type construction unit and Relative Navigation wave filter；

The metrical information of Inertial Measurement Unit, the carrier wave from filter status vector in GNSS/INS hypercompact combination navigation devices Carrier state information, the wire transmission of carrier phase Jing that metrical information and junction filter are estimated is to the corresponding dynamic of the carrier In model construction unit；The carrier state information and carrier phase also Jing that simultaneously junction filter is estimated is wirelessly transmitted to other In the dynamic model construction unit of carrier；

The corresponding dynamic model construction unit of each carrier is using wire transmission and is wirelessly transferred the information of acquisition determining dynamic Dynamic model is built after the quantity of state of model and observed quantity；According to dynamic model, Relative Navigation wave filter is optimum using Kalman Filtering recurrence equation group, is calculated the optimal estimation value of error state, and the relative position and relative velocity of two carriers are entered Row compensation further obtains the relative position and relative velocity of optimum.

4. relative navigation system according to claim 3, it is characterised in that the quantity of state of dynamic model and observed quantity are such as Under：

The quantity of state of dynamic model is x_GPSAnd x_INS

$x_{GPS}=\[\begin{array}{ccccc}\mathrm{\Δ}x& \▿\mathrm{\Δ}N& {\mathrm{\Δm}}_{\mathrm{\φ}}& \mathrm{\δ}T& \mathrm{\δ}I\end{array}\];$

x_INS=[δ Δ v δ Δ x δ w_m δa_m]^T；

In formula, Δ x is the relative position of two carriers,For double difference integer ambiguity, Δ m_φSingle poor carrier phase multipath Error, δ T are tropospheric error；δ I are ionospheric error；δ Δ v are the relative velocity error state of two carriers, and δ Δs x is two The relative position error state of individual carrier, δ w_mFor carrier m gyro error attitude vectors；δa_mFor carrier m acceleration error states Vector, subscript m is carrier label；

The observed quantity of dynamic model is z=[Δ v Δ x]^T, Δ v is the relative velocity of two carriers.