CN105652297A - Method and system for realizing real-time orbit determination for single satellite navigation positioning system - Google Patents
Method and system for realizing real-time orbit determination for single satellite navigation positioning system Download PDFInfo
- Publication number
- CN105652297A CN105652297A CN201410648215.4A CN201410648215A CN105652297A CN 105652297 A CN105652297 A CN 105652297A CN 201410648215 A CN201410648215 A CN 201410648215A CN 105652297 A CN105652297 A CN 105652297A
- Authority
- CN
- China
- Prior art keywords
- orbit determination
- kalman filter
- satellite navigation
- time
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Position Fixing By Use Of Radio Waves (AREA)
- Navigation (AREA)
Abstract
The invention provides a method and a system for realizing real-time orbit determination for a single satellite navigation positioning system. The method comprises steps that a should-be-switched-to orbit determination work mode is determined according to a present orbit determination work mode and the available star number of the satellite navigation positioning system, and switching operation is carried out; the observation data of the satellite navigation positioning system is transferred from an earth-fixed coordinate system to an orbit coordinate system through coordinate transformation; the observation data after coordinate transformation is inputted to a Kalman filter, time update for the state quantity of the Kalman filter is carried out, and orbital element one-step forecast is acquired; measurement update for the state quantity of the Kalman filter is carried out by utilizing the observation data of the satellite navigation positioning system. The Kalman filtering method on the basis of a dynamic orbit model and the system are employed, being closer to practical modeling, switching among different orbit determination modes can be realized, demands of high precision real-time orbit determination and autonomous satellite navigation can be satisfied, and the method and the system can be applied to multiple types of satellite navigation positioning systems such as GPS, GLONASS, second-generation Beidou and GALILEO.
Description
Technical field
The invention belongs to satellite-navigation positioning field, especially relate to a kind of real-time orbit determination implementation method of monophyly based on satellite navigation and location system measurement and system.
Background technology
GPS/GLONASS be a kind of can the satellite navigation system of spatial intersection location of timing and range finding, it is possible to provide round-the-clock, continuous, real-time, high accuracy three-dimensional position, three-dimensional speed and time information to Global Subscriber. The observation mode of satellite independent navigation is mainly divided into star-navigation and radist to navigate two kinds of modes. Star-navigation utilizes the measurement of optomagnetic characteristic to nature celestial body to realize satellite independent navigation, and radist navigation utilizes satellite-carried wireless electric measuring instrument can obtain the observed value of velocity of variation of Distance geometry distance of the satellite beacon known relative to position. Above-mentioned two class autonomous navigation methods have his own strong points, and wherein radist method has higher navigation accuracy, but needs the support of artificial beacon, and celestial body measuring method has autonomous ability completely. Research satellite autonomous navigation technology mainly considers following four aspects: the measurement type of navigation, the foundation of kinetic model, navigation algorithm and onboard computer. Navigation locating method utilizes observation model and navigation principle that observed quantity converts to required position and the method for speed.
According to whether considering the impact of the disturbing force suffered by low orbit satellite and the relation with perturbation mechanical model, the navigation algorithm of satellite independent navigation can be divided into pure geometric method, KINETIC METHOD and comprehensive dynamic method three kinds: the ultimate principle of pure geometric method is linear intersection in space, and it refers to and does not rely on any mechanical model, completely finds range data to the method for Orbit determination for LEOs by space radio, namely power method is the orbit determination method in traditional sense, utilizes track kinetics and statistics or filtering theoretical, obtains accurate and stable track solution from usually more sparse with having the measurement of noise, and the precision of Dynamic orbit determination depends primarily on mechanical model, in order to solve in Dynamic orbit determination Problems existing in Dynamic model error and geometric method, the scientists such as Yunck propose comprehensive dynamic orbit determination method KINETIC METHOD and geometric method joined together, use Kalman filtering form, using the kinetics track obtained after filtering in batches as with reference to track, perturbation factors that process noise parameter comes in absorption dynamics model not consider is addition of again and not by the Dynamic model error of accurate model in follow-up sequential filtering process, by the most preferably power between kinetic model and the geological information of satellite navigation and location system observed value offer, make the impact of pure kinetic model weakened.
That the shortcoming of above-mentioned comprehensive dynamic orbit determination method is to be applied in engineering practice modeling is simple, can not fully close actual model. Meanwhile, existing spaceborne satellite navigation and location system receiving apparatus adopts simple geometry method to carry out One-Point Location usually, and its real-time accuracy is relatively low.
Summary of the invention
It is an object of the invention to: overcome comprehensive dynamic orbit determination method modeling in prior art simple, fully close actual model and existing employing simple geometry method can not carry out the relatively low defect of the spaceborne satellite navigation and location system real-time accuracy of One-Point Location, it is provided that the satellite-navigation orbit determination method of a kind of real-time high-precision.
Given this, the technical solution of the present invention is: a kind of single real-time orbit determination implementation method of satellite navigation and location system, comprises the following steps:
According to the available star number of current orbit determination operating mode and satellite navigation and location system, it is determined that should be switched to orbit determination operating mode and perform switching;
Satellite navigation and location system observed data is consolidated ordinate transform to rail coordinate road system by coordinate conversion from ground;
The observed data input card Thalmann filter of coordinate conversion will be passed through, and the quantity of state of Kalman filter is carried out time renewal, obtain track radical a step of forecasting;
After the described time has upgraded, by satellite navigation and location system observed data, the quantity of state of Kalman filter is carried out measurement updaue.
Further, the quantity of state of Kalman filter is carried out measurement updaue and comprises by described satellite navigation and location system observed data:
By pseudo range observed quantity, the pseudorange filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue;
With Doppler measurements, doppler's filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue.
Further, the quantity of state of Kalman filter is carried out time renewal and comprises by described foundation track kinetic model:
With the track radical quantity of state of previous epoch, calculate the disturbing force that navigation system with time and rangine is subject to;
Synthesize acceleration perturbation item according to described disturbing force, and calculate the velocity of variation of each quantity of state;
The velocity of variation of each quantity of state described is added on previous epoch, by perturbed motion equation, calculates the predicted value of the quantity of state of current epoch.
Further, the real-time orbit determination implementation method of described single satellite navigation and location system, also comprises:
By judging the value of the convergency of clock correction, available star number and geometric dilution of precision, it is determined that the condition that Kalman filter starts.
Further, the real-time orbit determination implementation method of described single satellite navigation and location system, also comprises:
Judge whether the sub-minimum of the observed quantity of pseudorange and the difference of calculated amount transfinites, if it does, judge that Kalman filter is dispersed;
Whether position and the mould of the difference of the position of channel plate One-Point Location after judging track plate filtering transfinite, if it does, judge that Kalman filter is dispersed;
When monitoring Kalman filter and disperse, heavily open described Kalman filter.
Further, the real-time orbit determination implementation method of described single satellite navigation and location system, also comprises:
On spaceborne receiving apparatus, regularly note Kalman filter conversion parameter go forward side by side row data backup;
Described data backup comprises hot standby part and cold standby part, and the device that corresponding Backup Data is accessed comprises fast device and device at a slow speed.
Further, the real-time orbit determination implementation method of described single satellite navigation and location system, also comprises:
It is updated to benchmark with the time of quantity of state, rejects overproof observed quantity by setting thresholding.
Further, the real-time orbit determination implementation method of described single satellite navigation and location system, also comprises:
Arranging two buffer area, the main circulating program of interrupt service routine and Kalman filter is alternate access Liang Ge buffer memory district chronologically.
Present invention also offers a kind of single real-time orbit determination Project Realization system of satellite navigation and location system, comprising:
Orbit determination operating mode handover module: for the available star number according to current orbit determination operating mode and satellite navigation and location system, it is determined that the orbit determination operating mode that should adopt;
Coordinate transferring, for consolidating ordinate transform to track system by coordinate conversion from ground by satellite navigation and location system observed data;
Track radical a step of forecasting module, for the quantity of state of Kalman filter being carried out time renewal according to track kinetic model, obtains track radical a step of forecasting;
Kalman filtering module, carries out measurement updaue by satellite navigation and location system observed data to the quantity of state of Kalman filter.
Further, the real-time orbit determination Project Realization system of described single satellite navigation and location system, also comprises:
Injection molding block in data, notes Kalman filter conversion parameter for regularly on spaceborne receiving apparatus.
The present invention's advantage compared with prior art is: the present invention adopts the kalman filter method based on track kinetic model and system, has combined KINETIC METHOD and Kalman Filter Estimation method. In KINETIC METHOD, it is to increase the model accuracy of orbit determination; In Kalman Filter Estimation method, by the linearizing of perturbed motion equation in kinetic model and numerical integration, Kalman filter being generalized to extended BHF approach method. The method and system are closer to actual modeling, and real-time accuracy height. While improving positioning precision, also possess orbit determination function, by judging the available star number of current satellite navigation and location system, automatically switch between different orbit determination patterns, the needs of orbit determination and satellite independent navigation when meeting high-precision real. And can be applied in the multiple satellite navigation and location systems such as GPS, GLONASS, Beidou II and GALILEO.
Accompanying drawing explanation
A kind of single real-time orbit determination engineering implementation method schema of satellite navigation and location system that Fig. 1 provides for the embodiment of the present application;
A kind of orbit determination operating mode switching machine drawing that Fig. 2 provides for the embodiment of the present application;
A kind of coordinate transformation relation figure that Fig. 3 provides for the embodiment of the present application;
A kind of track radical a step of forecasting model diagram that Fig. 4 provides for the embodiment of the present application;
The graph of a relation of a kind of guarantee condition that Fig. 5 provides for the embodiment of the present application and state;
A kind of Kalman filtering schema that Fig. 6 provides for the embodiment of the present application;
Injection molding block and backup machine drawing in a kind of data that Fig. 7 provides for the embodiment of the present application;
The internal memory structure iron of a kind of pair buffers that Fig. 8 provides for the embodiment of the present application;
A kind of single real-time orbit determination Project Realization system architecture figure of satellite navigation and location system that Fig. 9 provides for the embodiment of the present application;
Figure 10 is the mathematics emulation structure iron of the present invention;
The physics emulation structure iron that Figure 11 provides for the embodiment of the present application;
The System's composition block diagram of the signal simulator that Figure 12 provides for this embodiment of the present application;
The graphicerrors that the spaceborne one-point positioning method that Figure 13 provides for the embodiment of the present application positions;
Figure 14 stablizes the graphicerrors of filtering for spaceborne real-time orbit determination method that the embodiment of the present application provides;
The graphicerrors of the spaceborne real-time orbit determination method short-time forecast that Figure 15 provides for the embodiment of the present application.
Embodiment
The embodiment of the present invention mainly provides a kind of single real-time orbit determination implementation method of satellite navigation and location system, simple for solving comprehensive dynamic orbit determination method modeling comparison in engineering practice in prior art, can not fully close to the problem of actual model, the scheme that existing spaceborne satellite navigation and location system receiving apparatus adopts simple geometry method to carry out One-Point Location is improved to spaceborne real-time orbit determination method, it is to increase its real-time accuracy.
Schema shown in Figure 1, the real-time orbit determination implementation method of described single satellite navigation and location system at least comprises the following steps:
Step S101, according to the available star number of current orbit determination operating mode and satellite navigation and location system, it is determined that the orbit determination operating mode that should be switched to also performs switching;
For GPS/GLONASS, according to the available star number of current orbit determination operating mode and GPS/GLONASS, it is determined that the orbit determination operating mode that this second should adopt, it is achieved orbit determination work is from the mutual switching between single GPS orbit determination operating mode and single GLONASS orbit determination pattern.
Step S102, consolidates ordinate transform to track system of coordinates by coordinate conversion from ground by satellite navigation and location system observed data;
By GPS/GLONASS observed data by coordinate conversion from WGS84 solid system of coordinates or PZ90.2 ordinate transform to track system.
Step S103, will pass through the observed data input card Thalmann filter of coordinate conversion, and according to track kinetic model, the quantity of state of Kalman filter is carried out time renewal, obtain track radical a step of forecasting;
Send by the GPS/GLONASS observed data of coordinate conversion into Kalman filter, by Runge-Kutta numerical integration, the quantity of state of Kalman filter being carried out track radical a step of forecasting in the state equation of Kalman filter, the time namely performing Kalman filter upgrades.
Step S104, after the described time has upgraded, carries out measurement updaue by satellite navigation and location system observed data to the quantity of state of Kalman filter.
After time upgraded, the quantity of state of Kalman filter is carried out measurement updaue by the GPS/GLONASS observed data being input to Kalman filter.
The present embodiment is for GPS/GLONASS, essence is the observation model based on space radio range finding and frequency measurement, therefore described real-time orbit determination method can be applied in the observation model based on the satellite navigation and location system such as Beidou II, GALILEO further, repeats no more in the present embodiment.
Fig. 2 is a kind of switching machine drawing of the orbit determination operating mode described in embodiment of the present invention step S101, sketches as follows:
If under being currently in single GPS orbit determination operating mode, then:
(1) as long as channel plate location and GPS can be no less than 4 by star number, current orbit determination pattern still selects single GPS orbit determination pattern;
(2) if channel plate is located, GPS can be less than 4 by star number and GLONASS can be not less than 4 by star number, then current orbit determination pattern switches to single GLONASS orbit determination pattern;
(3) if channel plate location and GPS and GLONASS can all be less than 4 by star number, or during channel plate non-locating, rail joints software is forced to put non-locating mark, entering extrapolation (namely upgraded by the Kalman filter time and obtain track radical), current orbit determination operating mode remains unchanged.
If under being currently in single GLONASS orbit determination operating mode, then:
(1) as long as channel plate location and GPS can be no less than 6 by star number, current orbit determination pattern switches to single GPS orbit determination pattern;
(2) if channel plate is located, GPS can be less than 6 by star number and GLONASS can be not less than 4 by star number, then current orbit determination pattern remains single GLONASS orbit determination pattern;
(3) if channel plate is located, GPS can be less than 6 by star number, being more than or equal to 4, and GLONASS can be less than 4 by star number, current orbit determination pattern switches to single GPS orbit determination pattern;
(4) if channel plate location and GPS and GLONASS can all be less than 4 by star number, or during channel plate non-locating, rail joints software is forced to put non-locating mark, enters extrapolation, and orbit determination operating mode remains unchanged.
Fig. 3 is a kind of graph of a relation of coordinate conversion in embodiment of the present invention step S102. It is described that by satellite navigation and location system observed data, by coordinate conversion, from ground, solid ordinate transform, to track system, specifically can realize in the following ways:
Being transformed in perturbation model by observed data, mainly comprise and calculate life position, the flat radical of close radicalization, track radical and position and speed, ground is and track system admittedly, and track is tied to the conversion such as MJD0 and J2000. Wherein, unify when the current data of wave filter is UTC the quantity of state under track system of coordinates; Data of its input are the user's position and speed under gps time, WGS84 and the pseudorange of visual navigation star and the measuring vol of doppler; Its data exported are the instantaneous radical under track system of coordinates, by the conversion in the path shown in Fig. 3, it is possible to obtain the position and speed under WGS84, position and speed under J2000, mean of date equatorial coordinate system and the flat root of track, wink root.
With reference to the track radical a step of forecasting model diagram shown in figure 4, according to track kinetic model, the quantity of state of Kalman filter is carried out time renewal described in embodiment of the present invention step S103 and at least can comprise:
(1) with the track radical quantity of state of previous epoch, the corresponding disturbing force that navigation system with time and rangine is subject to is calculated;
(2) synthesize acceleration perturbation item according to described disturbing force, and calculate the velocity of variation of each quantity of state;
(3) velocity of variation of each quantity of state described is added on previous epoch, by perturbed motion equation, calculates the predicted value of the quantity of state of current epoch.
As the state equation of Kalman filter, its cardinal principle is by setting up track kinetic model, the quantity of state of Kalman filter is carried out a step of forecasting by track radical a step of forecasting, and its track kinetic model is synthesized by the superposition theorem of power primarily of models such as earth gravitational field, atmospherical drag, optical pressure.
When carrying out a step of forecasting, with quantity of states such as the track radicals of previous epoch, corresponding disturbing force is calculated by models such as the earth gravitational field in track kinetics, atmospherical drag, optical pressure, by the superposition theorem of power synthesis acceleration perturbation item, substitute into the velocity of variation calculating each quantity of state in perturbed motion equation, adopt numerical integration to be added on previous epoch, form the predicted value of the quantity of state of current epoch.
With reference to the Kalman filtering schema shown in figure 6, by satellite navigation and location system observed data, the quantity of state of Kalman filter is carried out measurement updaue described in embodiment of the present invention step S104 and at least can comprise:
(1) by pseudo range observed quantity, the pseudorange filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue;
(2) with Doppler measurements, doppler's filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue.
The Kalman filtering schema of Fig. 6, describes the algorithm structure of the Kalman filter of belt track kinetic model. In Fig. 6, the left side is the renewal process of the quantity of state of Kalman filter, and the right is the renewal process of the varivance matrix of Kalman filter. Kalman filter state amount is being carried out on the basis of a step of forecasting by track radical a step of forecasting, the quantity of state of Kalman filter carries out filtering in two steps: the first step, by pseudo range observed quantity, the pseudorange filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue, for ensureing the accuracy of the coordinate information of forecast;2nd step, carries out measurement updaue with Doppler measurements to doppler's filtering error variance matrix of the quantity of state of Kalman filter, for ensureing the accuracy of the speed information of forecast. The precedence of above-mentioned two steps can change.
In orbit determination model, based on GPS/GLONASS star between the mode of radist, the observed quantity used is for user is to the pseudorange �� of each navigation system with time and rangine and pseudorange velocity of variation(pseudorange velocity of variation and doppler); For meeting the needs of Kalman filter, it is assumed that the noise being loaded on pseudorange and pseudorange velocity of variation is white Gaussian noise, need the variance to noise to carry out prior estimate simultaneously;
In the calculation, Kalman filter is divided into ten steps, and the first two step mainly adopts a step of forecasting of track kinetic model that the quantity of state of Kalman filter is carried out time renewal, upgrades the prediction error variance matrix of Kalman filter simultaneously. Eight steps below, to being divided into two parts, adopt pseudorange and pseudorange velocity of variation, as observed quantity, the quantity of state of Kalman filter is carried out measurement updaue respectively. In measurement updaue, in order to avoid overproof observed quantity to affect the stability of Kalman filter, introduce unruly-value rejecting mechanism, it is updated to benchmark with the time of quantity of state, by setting the mode of thresholding, overproof observed quantity is rejected. Filtering error variance matrix in Kalman filter is symmetric matrix, in long-time iterative process, owing to the round-off error calculated may cause it to lose symmetry, in the calculation, adopts the mode of simple corresponding arithmetical mean to carry out pressure symmetrical. Step summary is as follows:
The first step, by tkThe filter value in moment utilizes a step of forecasting (numerical integration) to calculate tk+1The predicted value in moment, initial t0The filter value in moment, the value of its six radicals is calculated by " initial estimation ", andInitial estimation be taken as zero, b,Initial estimation by t0The One-Point Location in moment provides;
2nd step: by tkThe filtering error variance matrix in moment calculates tk+1The prediction error variance matrix in moment;
3rd step: judge pseudorange outlier, compute pseudo-ranges observing matrix;
4th step: compute pseudo-ranges gain matrix;
5th step: compute pseudo-ranges filter value;
6th step: compute pseudo-ranges filtering error variance matrix;
7th step: differentiate doppler's outlier, calculates doppler's observing matrix;
8th step: calculate doppler's gain matrix;
9th step: calculate doppler's filter value;
Tenth step: calculate doppler's filtering error variance matrix;
In the above-mentioned processing mode the quantity of state of Kalman filter being carried out measurement updaue:
To the numerical stability issues of the varivance matrix P of Kalman filter, propose to force symmetrical method, two elements pressing diagonal lines symmetry in square formation are taken out by the method, after alignment carries out arithmetical mean, replace original two element, thus realize symmetry, with UD decomposition method comparatively speaking, the algorithm of the method is simple, feasibility height in actual applications.
In Kalman filter in the method for estimation of the noise of observed data, adopt the prior estimate Confirming model error matrix Q of measurement noises, the method is relatively more effective for the observation model that noisiness is stable, can effectively suppress the kick in observed data simultaneously.
The graph of a relation of a kind of guarantee condition that Fig. 5 provides for the embodiment of the present application and state, in the technical scheme that the present embodiment provides, in order to ensure the stability that described Kalman filter works for a long time, it is also possible to comprise two kinds of main guarantee conditions:
One is that Kalman filter starts condition, namely by judging the value of the convergency of clock correction, available star number and geometric dilution of precision, it is determined that whether Kalman filter can start.
Clock correction stablizes (convergence) condition: the pseudorange biases b continous-stable caused by clock correction, namely | and bk-bk-1| < blimAnd | bk-b'k-1| < blim. Wherein b'0=b0, andN=30sec, blim=30m, k are current moment epoch, k �� [12 ... n].
Two is the condition judging that Kalman filter is dispersed. Mainly comprise:
(1) judge whether the sub-minimum of the observed quantity of pseudorange and the difference of calculated amount transfinites, if it does, judge that Kalman filter is dispersed;
Filter divergence condition 1: do not transfinite according to the sub-minimum of the observed quantity of pseudorange and the difference of calculated amount, the super b of difference both namely in the time of continuous n secondli, it is determined that for Kalman filter is dispersed. | min'{ ��o-��c}k|n< blim, wherein n=60sec, blim=120m.
(2) whether position and the mould of the difference of the position of channel plate One-Point Location after judging track plate filtering transfinite, if it does, judge that Kalman filter is dispersed;
Kalman filter divergence case 2: do not transfinite according to the mould of the position after track plate filtering and the difference of the position of channel plate One-Point Location, the super b of difference both namely in the time of continuous n secondlim, it is determined that it is filter divergence, wherein n=60sec, blim=120m.
(3) when monitoring Kalman filter and disperse, described Kalman filter is heavily opened.
In such scheme, introduce the monitoring condition that Kalman filter is dispersed. When monitoring Kalman filter and disperse, recover by heavily opening the mode of Kalman filter. Advantage, in the process of transformation classic card Thalmann filter, simplifies algorithm; To in the monitoring of Kalman filter, have employed the result directly utilizing observed data to carry out One-Point Location, the result precision of One-Point Location is decided by the precision of measuring vol and the star condition of navigation system with time and rangine completely, therefore its stability separated can ensure, the divergent state carrying out monitoring card Thalmann filter as benchmark is feasible.
It is the data of the present invention are noted and backup machine drawing with reference to figure 6. In the embodiment of the present invention, model error matrix in Kalman filter needs the prior estimate to kalman filter state amount error, and the characteristic such as this estimation and satellite orbit height is relevant, in order to reach best coupling to improve orbit determination precision, when orbit altitude changes, it is necessary to adjust in time. Simultaneously, based on the orbit determination software of GPS, its navigation system with time and rangine position is in (WGS84) with being defined in admittedly, the track kinetic model of orbit determination algorithm is defined in inertial system (track system of coordinates), need to carry out coordinate conversion when calculating, accurate coordinate conversion needs to revise the astronomical constant such as earth rotation correction and pole shifting correction, and the exact value of these data derives from ground official website. Consequently, it is desirable to regular upper note data backup.
In order to ensure the accurate degree of Kalman filter, the high precision conversion prior burning of parameter required for it is in spaceborne receiving apparatus storage unit, and period carries out regularly upper note by ground in-orbit, and the data of upper note back up simultaneously in described storage unit.
Data backup is divided into hot standby part (RAM) and cold standby part (FLASH or EEPROM), and usual hot standby part is stored in fast device (RAM), and cold standby part is stored at a slow speed (FLASH or EEPROM) in device. The steering routine of access fast device, is arranged in master routine, represents with solid line in fig. 8; Access the steering routine of device at a slow speed, it is arranged in timer interrupt service routine, is represented by dotted lines in fig. 8.
In engineering practice, owing to observed data is easily subject to external interference. In order to avoid the noise of the abnormal short period of time such as the kick being carried in observed quantity on the impact of Kalman filter, embodiments of the invention can also be incorporated herein the rejecting mechanism of observed quantity outlier, therefore the real-time orbit determination engineering implementation method of single satellite navigation and location system that the present embodiment provides, it is also possible to comprising:
Kalman filter is introduced outlier judgment mechanism, is updated to benchmark with the time of quantity of state, reject overproof observed quantity by setting the mode of thresholding.Described outlier is surveying in the set that forms of data to a sequential tracks of dynamic target, and substantial deviation major part data are presented the sub-fraction data point of variation tendency.
By being rejected to the observed quantity being confirmed to be outlier, it is possible to get rid of the kick in take off data to the impact of kalman filter state amount, be conducive to improving orbit determination precision.
With reference to the internal memory structure iron of the pair buffers shown in figure 7, in the embodiment of the present invention, in order to avoid major cycle and interrupt service routine to read and write the possibility of same piece of internal memory simultaneously, the data integrity of the shared drive solving the main circulating program of interrupt service routine and Kalman filter exchanges problem, introduces pair buffers alternate access Liang Ge buffer memory district chronologically.
When each respective external is interrupted, CPU (central processing unit) arranges DMA (immediate address access unit) and prepares to receive number. After harvesting number, oneself triggers an interruption, and CPU arranges again DMA, then the result of resolving of upper one second is sent out. Twin Cache Architecture is the structure of principal and subordinate's formula. According to the basic norm exchanged, the important degree of write data is higher than the important degree reading data. DMA can be parallel with CPU, and major cycle and interrupt service routine exchange data based on pair buffers, thus ensure that the integrity of data.
It is the real-time orbit determination implementation method of single satellite navigation and location system of embodiment of the present invention offer above, the method is based on the kalman filter method of track kinetic model, while improving positioning precision, also possesses orbit determination function, especially when can the TV star can not carry out One-Point Location less than 4, can automatic forecasting, there is Orbit extrapolation ability, and can by judging the available star number of current satellite navigation and location system, automatically switch between different orbit determination patterns, the needs of orbit determination and satellite independent navigation when meeting high-precision real. And can be applied in the multiple satellite navigation and location systems such as GPS, GLONASS, Beidou II and GALILEO.
Corresponding to aforesaid method embodiment, the embodiment of the present application additionally provides a kind of single real-time orbit determination Project Realization system of satellite navigation and location system, and module diagram shown in Figure 9, comprising:
Orbit determination operating mode handover module: for the available star number according to current orbit determination operating mode and satellite navigation and location system, switching orbit determination operating mode;
Coordinate transferring, for consolidating ordinate transform to track system by coordinate conversion from ground by satellite navigation and location system observed data;
Track radical a step of forecasting module, for the quantity of state of Kalman filter is carried out time renewal, obtains track radical a step of forecasting;
Kalman filtering module, carries out measurement updaue by satellite navigation and location system observed data to the quantity of state of Kalman filter.
Corresponding with embodiment of the method part, the system described in the application can also comprise:
Injection molding block in data, notes Kalman filter conversion parameter for regularly on spaceborne receiving apparatus.
Double buffers module, for when the major cycle of Kalman filter and interrupt service routine exchange data, protecting the integrity of data frame.
The major cycle module of wave filter, interrupt service routine module, be respectively used to realize calling and starting interrupt routine of master routine.
Guarantee condition module, for ensureing the stability that described Kalman filtering module works for a long time.
The mode of operation of each module is with to realize principle consistent with embodiment of the method part above, repeats no more.
The real-time orbit determination Project Realization system of single satellite navigation and location system described in the present embodiment can be transplanted in the embedded platform based on DSP, while improving positioning precision, also possesses orbit determination function, by judging the available star number of current satellite navigation and location system, automatically switch between different orbit determination patterns, the needs of orbit determination and satellite independent navigation when meeting high-precision real.And can be applied in the multiple satellite navigation and location systems such as GPS, GLONASS, Beidou II and GALILEO.
It is as follows that the present embodiment additionally provides mathematics simulation stage and physics simulation stage to described method and system:
In mathematics simulation stage, by the time system of the theoretical investigation Confirming model of algorithm and coordinate system, choose perturbation model and the method for estimation of navigation algorithm, set up observation model and state model, PC uses the algorithm based on C language realized, use measured data the orbit determination precision of model to be verified simultaneously. In physics simulation stage, main being transplanted to by the mathematics simulation software of spaceborne orbit determination technology is on the embedded hardware platform of core taking DSP, under storage space and the restriction of operation time, it is achieved the reliability of running software in physical environment, and realize the stability of Output rusults.
Figure 10 is the mathematics emulation structure iron of the present invention. In mathematics simulation stage, main contents are divided into three parts: one is time system and coordinate system, mainly the definition of the time used in algorithm and system of coordinates is described in detail; Two is track kinetic model, and this model is mainly used in quantity of state is carried out time renewal in Kalman filter, and perturbation model wherein is mainly terrestrial gravitation field model, atmospherical drag model and optical pressure model; Three is Kalman filter, choose track radical and set up the state equation based on track kinetic model as quantity of state, Kalman filter carries out time renewal, adopts pseudorange and the two kinds of observed quantities of pseudorange velocity of variation to set up measurement equation, in Kalman filter, quantity of state is carried out measurement updaue.
The process of mathematics emulation is as follows:
1. the GPS/GLONASS signal simulator that the embodiment of the present invention adopts is Sprient8000, and emulator provides radiofrequency signal to GPS/GLONASS receiving apparatus, provides the some position file of theoretical track simultaneously.
2. examine software by ground from GPS/GLONASS receiving apparatus, gather original observed data and the position and speed of navigation system with time and rangine under corresponding epoch, be stored as a position file.
3. on PC, ground inspection software reads observed data and the some position file of theoretical track simultaneously, and observed data is carried out orbit determination calculating, obtains orbit determination result and compares with theoretical track, to analyze orbit determination precision.
Figure 11 is the physics emulation structure iron of the present invention. In physics simulation stage, orbit determination software is transplanted to from PC taking DSP be treater On board computer, it is necessary to by data interface by make in the way of reading and writing of files by synchronous serial interface on star exchange data mode. Working out Embedded orbit determination software on this basis, main purpose not only realizes algorithm on onboard computer in Real-Time Filtering and the orbit determination precision forecast in-orbit, and to be realized the stability of orbit determination result. Orbit determination software mainly solves three problems: note and safeguards in word length restriction, data.
On PC, the variable in orbit determination software can adopt the float-point arithmetic of high precision, thus ensure that the precision of orbit determination; The DSP used in software simulating is the treater based on 32, its compiler carried only support 32 floating numbers fundamental operation (+,-, ��, �� ,=) and the fundamental operation of mathematical function storehouse and 40 floating numbers, under existing resource, the precision of orbit determination software obviously can not meet requirement, and this is exactly word length restriction. In the process of through engineering approaches, the main mode adopting introducing 40 and 64 storehouse functions, the method exchanging precision with Time and place for is to ensure the computing precision of Kalman filter.
In orbit determination software, its core navigation algorithm adopts the Kalman filter based on track kinetic model, and observed quantity is based on the satellite navigation system of GPS/GLONASS. As the outside input of navigation algorithm, time epoch in navigation message is the GPS/GLONASS time, the coordinate of employing be be WGS84 admittedly; Time epoch that the track kinetic model of navigation algorithm adopts is UTC time, and the coordinate of employing is track system of coordinates. Therefore, in the algorithm, it is necessary to input data are carried out the conversion of time and coordinate, it is the support that the coordinate conversion carrying out high precision between inertial system needs one group of parameter admittedly on ground, the astronomical constant of this group can only obtain from network, therefore, is merely able to note on regularly in the application of software. The astronomical constant of upper note carries out data backup in chip external memory, mainly in order to the data that cause after preventing orbit determination software from resetting are lost.
Kalman filter comprises coloured noise in engineering practice and likely causes the quantity of state of Kalman filter to wander off in observed quantity, cause Kalman filter to be dispersed time serious. Therefore, need in the algorithm to detect the divergent state of Kalman filter, and recover by heavily opening the mode of Kalman filter. In addition, by the Kalman filter of outside data-driven, outside data may be interrupted; When outside data outage, Kalman filter automatically switches to state of forecasting in-orbit, when not measuring moment renewal, data output frequencies according to priori arranges a step of forecasting that forecast step-length carries out under track kinetic model, when, after externally measured date restoring, the forecast maintenance time being alignd with the externally measured time; The pseudorange biases caused by clock correction now loaded in take off data after alignment is not yet stable, directly introduces the permanent stability that will affect Kalman filter, and therefore setting model waits; Until its clock correction switches back to filter state after stablizing.
By the orbit determination software of engineering chemistry database, when ensureing orbit determination precision, it is possible to meet the operation time of onboard computer and the restriction of storage space, and realize the stability of real-time orbit determination result.
The physics emulation test of the present invention is as follows:
Export the performance index of data: position precision, is better than 10m (1 ��, three axles); Velocity accuracy, is better than 0.05m/s (1 ��, three axles); Orbit prediction precision, forecasts that in 100 minutes, precision is better than 40m; Data Update frequency, 1Hz.
The examination of the real-time orbit determination software of GPS/GLONASS is divided into ground simulation checking and examination of flying in-orbit.
Ground simulation checking is realized by GPS/GLONASS signal simulator, and GPS/GLONASS signal simulator can be used for assessing orbit determination precision and the kinetic characteristic of high dynamically user's GPS/GLONASS receiving apparatus. Figure 12 is the System's composition block diagram of GPS/GLONASS signal simulator of the present invention. This example GPS/GLONASS signal simulator adopts Spirent8000.
Simulate user satellite by GPS/GLONASS signal simulator to fly in-orbit, the track real-time orbit determination of plate software of its GPS/GLONASS receiving apparatus carried, the reference data that result data and the emulator of orbit determination directly provide is compared, it is possible to the precision of the assessment real-time orbit determination of Kalman filter.
When cutting off the signal source of emulator, Kalman filter enters forecast state. The reference data that result data under forecast state in orbit determination and emulator directly provide is compared, it is possible to the precision of assessment Kalman filter independent navigation.
The location of spaceborne One-Point Location software, the steady filter function of spaceborne real-time orbit determination software and short-time forecast function having been tested respectively by emulator, result is as follows:
Figure 13 is the graphicerrors that spaceborne one-point positioning method positions. As can be seen from the figure, nearly 10 meters of the position precision of the output data of spaceborne One-Point Location software, velocity accuracy is at about 0.6 meter/second, and these data have been the higher positioning accuracy that can reach based on the spaceborne one-point positioning method of least square estimation and geometric method.
Figure 14 is the graphicerrors that spaceborne real-time orbit determination method stablizes filtering. As can be seen from the figure, the position precision of the output data of spaceborne real-time orbit determination software is at 4��5 meters, and velocity accuracy, at 0.006��0.008 meter/second, will be much better than the requirement of task. By comparing with the positioning error of spaceborne One-Point Location software it will be seen that the orbit determination precision of spaceborne real-time orbit determination software is significantly improved.
Figure 15 is the graphicerrors of spaceborne real-time orbit determination method short-time forecast. Forecasting after 100 minutes, position residual error is 22.1501m, and meets positionerror in the forecast time period and be less than the task requirement of 40m. When illustrating that this Kalman filter does not have take off data to input within the short period of time, its result exported still can maintain higher precision.
Spaceborne real-time orbit determination software is successfully applied to certain model satellite aerial mission in-orbit with orbit determination type GPS/GLONASS receiving apparatus, and the survey rail data at actual measurement orbit determination result and ground observing and controlling center compare, and its orbit determination precision reaches the performance index of expection.
Being emulated by above-mentioned mathematics policy and physics, it can be appreciated that relative to original spaceborne one-point positioning method, the method that the application provides and system except having the raising of big step in precision, and the Performance comparision of two kinds of algorithms is as shown in table 1 below:
The spaceborne one-point positioning method of table 1 and spaceborne real-time orbit determination method Performance comparision
In sum, the described method that the present embodiment provides and system, while improving positioning precision, also possesses orbit determination function, by judging the available star number of current satellite navigation and location system, automatically switch between different orbit determination patterns, the needs of orbit determination and satellite independent navigation when meeting high-precision real. And can be applied in the multiple satellite navigation and location systems such as GPS, GLONASS, Beidou II and GALILEO.
Claims (10)
1. the real-time orbit determination implementation method of single satellite navigation and location system, it is characterised in that, comprise the following steps:
According to the available star number of current orbit determination operating mode and satellite navigation and location system, it is determined that the orbit determination operating mode that should be switched to also performs switching;
The observed data of satellite navigation and location system is consolidated ordinate transform to track system of coordinates by coordinate conversion from ground;
The observed data input card Thalmann filter of coordinate conversion will be passed through, and the quantity of state of Kalman filter is carried out time renewal, obtain track radical a step of forecasting;
After the described time has upgraded, by the observed data of satellite navigation and location system, the quantity of state of Kalman filter is carried out measurement updaue.
2. the real-time orbit determination implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, described comprise with the quantity of state of Kalman filter is carried out measurement updaue:
By pseudo range observed quantity, the pseudorange filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue;
With Doppler measurements, doppler's filtering error variance matrix of the quantity of state of Kalman filter is carried out measurement updaue.
3. the real-time orbit determination implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, the described quantity of state to Kalman filter carries out time renewal and comprises:
With the track radical quantity of state of previous epoch, calculate the disturbing force that navigation system with time and rangine is subject to;
Synthesize acceleration perturbation item according to described disturbing force, and calculate the velocity of variation of each quantity of state;
The velocity of variation of each quantity of state described was added on previous epoch, calculates the predicted value of the quantity of state of current epoch.
4. the real-time orbit determination implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, also comprise:
By judging the value of the convergency of clock correction, available star number and geometric dilution of precision, it is determined that the condition that Kalman filter starts.
5. the real-time orbit determination engineering implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, also comprise:
Judge whether the sub-minimum of the observed quantity of pseudorange and the difference of calculated amount transfinites, if it does, judge that Ka Erman Kalman filter is dispersed;
Whether position and the mould of the difference of the position of channel plate One-Point Location after judging track plate filtering transfinite, if it does, judge that Kalman filter is dispersed;
When monitoring Kalman filter and disperse, heavily open described Kalman filter.
6. the real-time orbit determination implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, also comprise:
On spaceborne receiving apparatus, regularly note Kalman filter conversion parameter go forward side by side row data backup;
Described data backup comprises hot standby part and cold standby part.
7. the real-time orbit determination implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, also comprise:
It is updated to benchmark with the time of quantity of state, rejects overproof observed quantity by setting thresholding.
8. the real-time orbit determination implementation method of single satellite navigation and location system according to claim 1, it is characterised in that, also comprise:
Arranging two buffer area, the main circulating program of interrupt service routine and Kalman filter is alternate access Liang Ge buffer memory district chronologically.
9. the real-time orbit determination Project Realization system of single satellite navigation and location system, it is characterised in that, comprising:
Orbit determination operating mode handover module: for the available star number according to current orbit determination operating mode and satellite navigation and location system, switching orbit determination operating mode;
Coordinate transferring, for consolidating ordinate transform to track system by coordinate conversion from ground by the observed data of satellite navigation and location system;
Track radical a step of forecasting module, for the quantity of state of Kalman filter is carried out time renewal, obtains track radical a step of forecasting;
Kalman filtering module, carries out measurement updaue by satellite navigation and location system observed data to the quantity of state of Kalman filter.
10. the real-time orbit determination Project Realization system of single satellite navigation and location system according to claim 9, it is characterised in that, also comprise:
Injection molding block in data, notes Kalman filter conversion parameter for regularly on spaceborne receiving apparatus;
Double buffers module, for the integrity of the protection data frame when the major cycle of Kalman filtering module and interrupt service routine exchange data;
Guarantee condition module, for ensureing the stability that described Kalman filtering module works for a long time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410648215.4A CN105652297A (en) | 2014-11-15 | 2014-11-15 | Method and system for realizing real-time orbit determination for single satellite navigation positioning system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410648215.4A CN105652297A (en) | 2014-11-15 | 2014-11-15 | Method and system for realizing real-time orbit determination for single satellite navigation positioning system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105652297A true CN105652297A (en) | 2016-06-08 |
Family
ID=56478878
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410648215.4A Pending CN105652297A (en) | 2014-11-15 | 2014-11-15 | Method and system for realizing real-time orbit determination for single satellite navigation positioning system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105652297A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107367744A (en) * | 2017-08-22 | 2017-11-21 | 温州大学 | LEO-based GPS orbit determination method based on adaptive measuring Noise Variance Estimation |
| CN109211225A (en) * | 2017-06-29 | 2019-01-15 | 中国科学院国家天文台 | Obtain method, system and the equipment of highly elliptic orbit space object remaining orbital lifetime |
| CN109991631A (en) * | 2017-12-29 | 2019-07-09 | 上海市复控华龙微系统技术有限公司 | Satellite positioning method and device, computer readable storage medium |
| CN110706265A (en) * | 2019-11-05 | 2020-01-17 | 中国人民解放军国防科技大学 | Maneuvering target tracking method for improving SRCKF strong tracking filtering |
| CN111487659A (en) * | 2019-01-28 | 2020-08-04 | 广州市中海达测绘仪器有限公司 | State recognition method and device, computer equipment and storage medium |
| CN111912295A (en) * | 2020-06-22 | 2020-11-10 | 中国人民解放军63850部队 | Trajectory drop point prediction system |
| CN111953401A (en) * | 2020-07-28 | 2020-11-17 | 中国西安卫星测控中心 | Autonomous request type orbit service system for microsatellite |
| CN111965685A (en) * | 2020-07-07 | 2020-11-20 | 北京自动化控制设备研究所 | Low-orbit satellite/inertia combined navigation positioning method based on Doppler information |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6133872A (en) * | 1997-10-17 | 2000-10-17 | Ball Aerospace & Technologies Corp. | Real time precision orbit determination system |
| JP3550520B2 (en) * | 1999-11-26 | 2004-08-04 | 富士通株式会社 | Trajectory calculation device and trajectory calculation method |
| CN102305630A (en) * | 2011-05-17 | 2012-01-04 | 哈尔滨工业大学 | Autonomous synthetic aperture radar (SAR) satellite orbit determination method based on extended kalman filter |
| CN103542854A (en) * | 2013-11-02 | 2014-01-29 | 中国人民解放军国防科学技术大学 | Autonomous orbit determination method based on satellite borne processor |
| CN103645489A (en) * | 2013-11-22 | 2014-03-19 | 航天恒星科技有限公司 | A spacecraft GNSS single antenna attitude determination method |
| CN103675861A (en) * | 2013-11-18 | 2014-03-26 | 航天恒星科技有限公司 | Satellite autonomous orbit determination method based on satellite-borne GNSS multiple antennas |
-
2014
- 2014-11-15 CN CN201410648215.4A patent/CN105652297A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6133872A (en) * | 1997-10-17 | 2000-10-17 | Ball Aerospace & Technologies Corp. | Real time precision orbit determination system |
| JP3550520B2 (en) * | 1999-11-26 | 2004-08-04 | 富士通株式会社 | Trajectory calculation device and trajectory calculation method |
| CN102305630A (en) * | 2011-05-17 | 2012-01-04 | 哈尔滨工业大学 | Autonomous synthetic aperture radar (SAR) satellite orbit determination method based on extended kalman filter |
| CN103542854A (en) * | 2013-11-02 | 2014-01-29 | 中国人民解放军国防科学技术大学 | Autonomous orbit determination method based on satellite borne processor |
| CN103675861A (en) * | 2013-11-18 | 2014-03-26 | 航天恒星科技有限公司 | Satellite autonomous orbit determination method based on satellite-borne GNSS multiple antennas |
| CN103645489A (en) * | 2013-11-22 | 2014-03-19 | 航天恒星科技有限公司 | A spacecraft GNSS single antenna attitude determination method |
Non-Patent Citations (4)
| Title |
|---|
| 刘峰等: "单点定位与实时定轨的星载算法的比较研究", 《飞行器测控学报》 * |
| 李敏等: "多模全球导航卫星系统融合精密定轨", 《测绘学报》 * |
| 贾沛璋等: "星载GPS卡尔曼滤波定轨算法", 《天文学报》 * |
| 魏东岩等: "针对高轨平台的GNSS卫星可视性分析研究", 《第四届中国卫星导航学术年会电子文集》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109211225A (en) * | 2017-06-29 | 2019-01-15 | 中国科学院国家天文台 | Obtain method, system and the equipment of highly elliptic orbit space object remaining orbital lifetime |
| CN107367744A (en) * | 2017-08-22 | 2017-11-21 | 温州大学 | LEO-based GPS orbit determination method based on adaptive measuring Noise Variance Estimation |
| CN107367744B (en) * | 2017-08-22 | 2018-05-01 | 温州大学 | LEO-based GPS orbit determination method based on adaptive measuring Noise Variance Estimation |
| CN109991631A (en) * | 2017-12-29 | 2019-07-09 | 上海市复控华龙微系统技术有限公司 | Satellite positioning method and device, computer readable storage medium |
| CN111487659A (en) * | 2019-01-28 | 2020-08-04 | 广州市中海达测绘仪器有限公司 | State recognition method and device, computer equipment and storage medium |
| CN110706265A (en) * | 2019-11-05 | 2020-01-17 | 中国人民解放军国防科技大学 | Maneuvering target tracking method for improving SRCKF strong tracking filtering |
| CN110706265B (en) * | 2019-11-05 | 2022-02-01 | 中国人民解放军国防科技大学 | Maneuvering target tracking method for improving SRCKF strong tracking filtering |
| CN111912295A (en) * | 2020-06-22 | 2020-11-10 | 中国人民解放军63850部队 | Trajectory drop point prediction system |
| CN111965685A (en) * | 2020-07-07 | 2020-11-20 | 北京自动化控制设备研究所 | Low-orbit satellite/inertia combined navigation positioning method based on Doppler information |
| CN111965685B (en) * | 2020-07-07 | 2023-01-13 | 北京自动化控制设备研究所 | Low-orbit satellite/inertia combined navigation positioning method based on Doppler information |
| CN111953401A (en) * | 2020-07-28 | 2020-11-17 | 中国西安卫星测控中心 | Autonomous request type orbit service system for microsatellite |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105652297A (en) | Method and system for realizing real-time orbit determination for single satellite navigation positioning system | |
| Mao et al. | New time-differenced carrier phase approach to GNSS/INS integration | |
| CN102620605B (en) | Global positioning system (GPS) and inertial navigation system (INS) combination guidance system for semi-physical simulation | |
| Meng et al. | Integrity monitoring for all-source navigation enhanced by Kalman filter-based solution separation | |
| CN107368659B (en) | The reliability verification method of satellite navigation system high-fidelity simulation model | |
| CN106773783B (en) | Semi-physical simulation test system and method for continuously operating reference station system | |
| CN112902989B (en) | Data error calibration method and device, electronic equipment and storage medium | |
| CN109612472A (en) | Construction method and device for autonomous navigation system of deep space probe | |
| Fan et al. | Integrated navigation fusion strategy of INS/UWB for indoor carrier attitude angle and position synchronous tracking | |
| Rehman et al. | Accurate and direct GNSS/PDR integration using extended Kalman filter for pedestrian smartphone navigation | |
| CN109752744A (en) | Multi-satellite combined orbit determination method based on model error compensation | |
| Li et al. | Novel hybrid algorithm of improved CKF and GRU for GPS/INS | |
| CN100507596C (en) | Ground GPS signal transmitter | |
| Shahrooei et al. | Falsification of learning-based controllers through multi-fidelity Bayesian optimization | |
| Wang et al. | Robust filter method for SINS/DVL/USBL tight integrated navigation system | |
| Tsog et al. | Intelligent data processing using in-orbit advanced algorithms on heterogeneous system architecture | |
| Pail et al. | GOCE gravity field analysis in the framework of HPF: operational software system and simulation results | |
| Guo et al. | Design and implementation of electric power patrol monitoring system based on Beidou and mobile network | |
| Bai et al. | Continuous positioning with recurrent auto-regressive neural network for unmanned surface vehicles in GPS outages | |
| Mao et al. | Accuracy analysis of BDS-3 experiment satellite broadcast ephemeris | |
| Chen et al. | Estimation strategy and accuracy analysis of GNSS real-time precise satellite clock error | |
| CN108427127A (en) | The test method and device of baseband chip performance | |
| Ren et al. | The Preliminary Realization and Evaluation of CTRF2020 Based on New BDS3 Technology | |
| Gutierrez et al. | A High-Performance Computing GNSS-aware Path Planning Algorithm for Safe Urban Flight Operations | |
| Gordon et al. | GSFC VLBI analysis center annual report |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160608 |