CN104991266A - Train satellite positioning method and train satellite positioning system based on collaborative integrity monitoring - Google Patents

Abstract

The invention discloses a train satellite positioning method and a train satellite positioning system based on collaborative integrity monitoring. The method comprises the steps of acquiring offline basic data, predicting feature point positioning error offline protection limits distributed in a time-space range to be positioned to form a time-space distribution feature library, extracting observation information of a navigation satellite to a target train in real time and calculating the target train positioning error online protection limit in real time according to the observation information, calculating the feature point positioning error offline protection limit corresponding to the running time and mileage value of the target train positioning error online protection limit in the time-space distribution feature library, respectively comparing the target train positioning error online protection limit and the feature point positioning error offline protection limit corresponding to the running time and mileage value of the target train positioning error online protection limit with an alarm threshold, and adjusting the target train positioning decision scheme in real time according to the result of comparison. By adopting the technical scheme of the invention, the function of train positioning integrity monitoring can be improved effectively.

Description

A kind of train satellite positioning method based on collaborative integrity monitoring and system

Technical field

The present invention relates to the train Positioning Technology field based on satellite navigation system.More specifically, a kind of train satellite positioning method based on collaborative integrity monitoring and system is related to.

Background technology

Train, as the main carriers of rail transportation system, adopts train operation control system (abbreviation train control system), train to follow the trail of early warning system (abbreviation early warning system) and has vital effect to the security of its operation in transit.In order to realize effective Train Control, and effectively protection is implemented when there is the security risks such as train from overtaking collision, accurately, reliable train operation state information, particularly train position in orbit, speed, directions etc. determine that corresponding safety precautions can play the key condition of useful effect.Conventional technological means, as taken turns shaft speed sensor, radar Doppler, trackside transponder etc., be limited by the contradiction of performance, Cost-benefit Equilibrium, be unfavorable for the technical need meeting autonomy-oriented, cost degradation, high efficiency, facilitate satellite navigation and locate this technology runs the safety-relevant applications fields such as control application & development at railroad train.

With GPS (Global Positioning System, GPS) be GLONASS (Global Navigation Satellite System) (the Global Navigation Satellite System of representative, GNSS) application in rail transportation system progressively worldwide receives publicity in recent years, adopt train-installed GNSS receiver and aiding sensors, track map data base carries out integrated, form Facility location unit (Location DeterminationUnit, LDU), can with the construction maintenance cost of relative moderate, higher performance level, significant vehicle-mounted independence is characterized as train control system, the location-based railway service system such as early warning system provides train position, speed, the status informations such as direction, assist and guarantee train operating safety.But, GNSS is adopted to realize train location, except blocking because of signal under the complicated terrain environment such as mountain valley, tunnel, station, multipath effect, the factor such as interference cause except penalty, also may there is relatively large deviation because Navsat exists the status information that fault causes LDU to export compared with true value, and then affect train operating safety.Based on this, the error level of uncertainty of effective integrity monitoring means to current train satnav is adopted to assess, alarm is provided under the condition exceeding pre-set limit, effectively can isolate the security risk that the factors such as satellite failure cause, make related application meet the principle of railway system's fault-safety principle.Receiver autonomous integrity monitoring (Receiver Autonomous IntegrityMonitoring is extensively adopted in conventional scheme, RAIM) technology is estimated available navigation number of satellite, the distribution of satellite constellation geometry, satellite failure state etc., thus the adverse effect of isolated fault where necessary, exception condition.But, RAIM may because the factors such as institute's receiving satellite quantity not sufficient cause its availability cannot carry out complete covering to the train journey of plan, and the integrity monitoring carried out in real time in Train Dynamic operational process may exist certain erroneous judgement and missed detection risk, and then affect it and play locational decision supporting function accurately and timely and assist train controls, follows the trail of effective enforcement close to functions such as early warning.

Therefore, need to provide a kind of train satellite positioning method based on collaborative integrity monitoring and system.

Summary of the invention

The object of the present invention is to provide a kind of train satellite positioning method based on collaborative integrity monitoring and system, for the railway one dimension orbital coordinate system binding characteristic that train satnav faces, consider that the availability that conventional RAIM technology may exist covers not enough and missed detection risk, utilize GNSS Navsat multidate information, satellite health status information, dedicated track map datum, the prior imformations such as train operation plan data, off-line prediction is carried out to the integrity monitoring result of train operational process in transit according to plan, and then Coordination Decision is carried out with real-time LDU result in train operation in transit, determine final train locating information process structure and the result of decision, effectively guarantee the security of train satnav.

For achieving the above object, the present invention adopts following technical proposals:

Based on a train satellite positioning method for collaborative integrity monitoring, the method comprises the steps:

S1, collection satellite almanac data, satellite health status data, dedicated track map datum and train operation plan data;

S2, each data gathered according to step S1, predict the positioning error off-line protection limit value of each unique point distributed in space-time unique to be positioned, composition spatial-temporal distribution characteristic storehouse;

S3, extract real-time Navsat to the observation information of the target train operated in space-time unique to be positioned, and calculate target train positioning error in real time according to observation information and protect limit value online;

S4, calculate in spatial-temporal distribution characteristic storehouse protect the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value;

S5, utilize target train positioning error to protect limit value online and protect limit value to compare with alarm threshold respectively with its time of running, mileage value characteristic of correspondence point location error off-line, and according to comparative result real-time adjustment aim train locational decision scheme.

Preferably, the dedicated track map datum in step S1 is obtained by map data base, and this map data base comprises positional information, attributive character information, the terrain information of each track key point.

Preferably, step S2 comprises following sub-step further:

S2.1, according to train operation plan to the mileage scope S in space-time unique to be positioned ₀according to spatial parameter σ _scarry out equal interval sampling, obtain S ₀/ σ _s+ 1 mileage value;

S2.2, be engraved in given time window T to during plan corresponding to each mileage value respectively _minterior according to time parameter σ _tcarry out equal interval sampling, obtain each mileage value distinguish 2 corresponding (T _m/ σ _t)+1 time value;

S2.3, be based upon feature point set { P}, wherein i-th unique point P (S of distribution in space-time unique to be positioned _i, T _i) by the mileage value S of this unique point _iwith time value T _iform;

S2.4, determine each unique point P (S _i, T _i) available observation satellite;

S2.5, predict each unique point P (S distributed in space-time unique to be positioned _i, T _i) positioning error off-line protection limit value E _l(S _i, T _i), formula is as follows:

R _t＝I-H _tΦ _t；

$E_L(S_i,T_i)={\mathrm{\σ}}_0\·\underset{j}{\max }\sqrt{\frac{{\mathrm{\Φ}}_{t,1j}^2+{\mathrm{\Φ}}_{t,4j}^2+{\mathrm{\Φ}}_{t,7j}^2}{R_{t,\mathrm{jj}}}};$

In formula, I is unit matrix; H _tfor equivalent measurement matrix; Φ _tfor filter gain matrix; Unique point P (S _i, T _i) time value T _icorrespond to time variable t; σ ₀for the detection coefficient selected according to false dismissal probability; Φ _{t, 1j}for filter gain matrix Φ _tthe 1st row, jth column matrix element; Φ _{t, 4j}for filter gain matrix Φ _tthe 4th row, jth column matrix element; Φ _{t, 7j}for filter gain matrix Φ _tthe 7th row, jth column matrix element; R _{t, jj}for matrix R _tjth row, jth column matrix element.

Preferably, also comprise the steps: after step S2.5

The alarm threshold AL that S2.6, basis are preset, judges the availability effective range in the spatial-temporal distribution characteristic storehouse of each positioning feature point error off-line protection limit value, and as the reference of planning, adjustment train operation plan, formula is:

If unique point P _ipositioning error off-line protection limit value E _l(S _i, T _i)>=AL, then this unique point P (S _i, T _i) mark availability be invalid;

If unique point P _ipositioning error off-line protection limit value E _l(S _i, T _i) <AL, then this unique point P (S _i, T _i) mark availability be effective.

Preferably, calculating target column car positioning error in step S3 protects the formula of limit value to be online:

R _t＝I-H _tΦ _t；

$E_Z(S_t,T_t)={\mathrm{\&sigma;}}_0\&CenterDot;\underset{j}{\max }\sqrt{\frac{{\mathrm{\&Phi;}}_{t,1j}^2+{\mathrm{\&Phi;}}_{t,4j}^2+{\mathrm{\&Phi;}}_{t,7j}^2}{R_{t,\mathrm{jj}}}};$

In formula, E _z(S _t, T _t) protect limit value online for target train positioning error; I is unit matrix; H _tfor equivalent measurement matrix; Φ _tfor filter gain matrix; The T time of running of target train _tcorrespond to time variable t; σ ₀for the detection coefficient selected according to false dismissal probability; Φ _{t, 1j}represent filter gain matrix Φ _tthe 1st row, jth column matrix element; Φ _{t, 4j}represent filter gain matrix Φ _tthe 4th row, jth column matrix element; Φ _{t, 7j}represent filter gain matrix Φ _tthe 7th row, jth column matrix element; R _{t, jj}representing matrix R _tjth row, jth column matrix element.

Preferably, step S4 comprises following sub-step further:

S4.1, just calculation target train operationally carve T _tposition this position is projected on track map, tries to achieve projected position and then extrapolate projection mileage value corresponding to this projection coordinate position

S4.2, with the mileage value that projects and the time of running T _tfor index, in spatial-temporal distribution characteristic storehouse, search for the unique point of mating with it:

If existing characteristics point P is (S _k, T _k) meet and T _t=T _k, then by this unique point P (S _k, T _k) positioning error protection limit value E _l(S _k, T _k) as protecting the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value E _l(S _t, T _t);

If not existing characteristics point P (S _k, T _k) meet and T _t=T _k, then mean value method is adopted to calculate off-line predicted value as protecting the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value E _l(S _t, T _t), prediction equation is as follows:

${\hat{E}}_L(S_t^{*},T_t)=\frac{1}{4}[E_L(S_{j^{\&prime;}},T_{j^{\&prime;}})+E_L(S_{j^{\&prime;}}+{\mathrm{\&sigma;}}_S,T_{j^{\&prime;}})+E_L(S_{j^{\&prime;}},T_{j^{\&prime;}}\&PlusMinus;{\mathrm{\&sigma;}}_T)+E_L(S_{j^{\&prime;}}+{\mathrm{\&sigma;}}_S,T_{j^{\&prime;}}\&PlusMinus;{\mathrm{\&sigma;}}_T)];$

$S_{j^{\&prime;}}\&le;S_t^{*}\&le;S_{j^{\&prime;}}+{\mathrm{\&sigma;}}_S;$

T _j'≤ T _t≤ T _j'+ σ _tor T _j'-σ _t≤ T _t≤ T _j';

In formula, σ _tfor time parameter, σ _sfor spatial parameter.

Preferably, following sub-step is comprised further according to comparative result real-time adjustment aim train locational decision scheme in step S5:

If target train positioning error protects limit value online and protect arbitrary value in limit value to be more than or equal to default alarm threshold with its time of running, mileage value characteristic of correspondence point location error off-line, then target train locational decision scheme is the observation data of isolation satnav, directly adopts auxiliary positioning sensing data to calculate target column truck position;

If target train positioning error protects limit value online and protect limit value to be all less than default alarm threshold with its time of running, mileage value characteristic of correspondence point location error off-line; then target train locational decision scheme is comprehensive employing satellite positioning receiver and auxiliary positioning sensors observe data, merges estimation target column truck position by tight coupling.

Perform the train global position system based on collaborative integrity monitoring of method as described above, this system comprises:

Satellite data obtains server, for gathering satellite almanac data and satellite health status data;

Integrity status predication server, for receiving the train operation plan data that TDCS sends, according to train operation plan data, satellite health status data, dedicated track map datum, each positioning feature point error off-line protection limit value that prediction distributes in space-time unique to be positioned, composition spatial-temporal distribution characteristic storehouse;

Be placed in the level diagnosis processing terminal on target train, for according to the observation information of satellite to the target train operated in space-time unique to be positioned, real-time calculating target train positioning error protects limit value online, the time of running protecting limit value with target train positioning error is online calculated in spatial-temporal distribution characteristic storehouse, mileage value characteristic of correspondence point location error off-line protection limit value, and utilize target train positioning error protect online limit value and with its time of running, mileage value characteristic of correspondence point location error off-line protection limit value compares with alarm threshold respectively, according to comparative result real-time adjustment aim train locational decision scheme,

Be placed in the train position decision package on target train, for gathering the metrical information of satellite navigation receiver and aiding sensors, according to the target train locational decision scheme implementation train locational decision after adjustment in real time, export current location and the speed of target train in real time.

Preferably, this system also comprises: information of forecasting distributing terminals, for spatial-temporal distribution characteristic storehouse to be sent to the level diagnosis processing terminal of target train by cordless communication network.

Preferably, level diagnosis processing terminal comprises:

Moonscope messaging interface module, for receiving the observation information of satellite to the target train operated in space-time unique to be positioned;

Information of forecasting receiver module, for receiving spatial-temporal distribution characteristic storehouse;

Integrity monitoring computing module, for according to the observation information of satellite to the target train operated in space-time unique to be positioned, calculates target train positioning error in real time and protects limit value online;

Information calculates module, protects the time of running of limit value online, mileage value characteristic of correspondence point location error off-line protects limit value for calculating in spatial-temporal distribution characteristic storehouse with target train positioning error;

Diagnosis decision-making module, protect limit value online for utilizing target train positioning error and protect limit value to compare with alarm threshold respectively, according to comparative result real-time adjustment aim train locational decision scheme with its time of running, mileage value characteristic of correspondence point location error off-line;

Information injection module, for being sent to train position decision package by the target train locational decision scheme after adjustment in real time.

Beneficial effect of the present invention is as follows:

1, the performance of train location integrity monitoring can effectively be promoted, particularly in train operation in transit, locating information acquisition, process may run into uncertainty, condition time variation, when Facility location unit LDU calculating may be caused to exist abnormal, undetected, utilize the off-line prediction implemented in advance to reduce the adverse effect of these factors, the decision-making that train position decision package is provided exports possesses higher safety benefit.

2, only adopt with routine compared with the LDU locational decision of RAIM technical support and there is stronger application power, the security risk that the decision process that effectively dips exists, guarantee the service level that the safety guarantee such as train control system, early warning system is applied.

3, possess significant expansibility, the application of multiple satellite navigation system pattern can be adapted to, as single mode GNSS, multimode GNSS, and different satellite navigations/aiding sensors integrated positioning system structure.

Accompanying drawing explanation

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in further detail.

Fig. 1 illustrates the train satellite positioning method process flow diagram based on collaborative integrity monitoring.

Fig. 2 illustrates the mapping relations schematic diagram of train operation plan and track map.

Fig. 3 illustrates positioning error protection limit value spatial-temporal distribution characteristic storehouse schematic diagram.

Fig. 4 illustrates target train locational decision switching principle figure.

Fig. 5 illustrates the train global position system structural drawing based on collaborative integrity monitoring.

Fig. 6 illustrates level diagnosis processing terminal structural drawing.

Embodiment

In order to be illustrated more clearly in the present invention, below in conjunction with preferred embodiments and drawings, the present invention is described further.Parts similar in accompanying drawing represent with identical Reference numeral.It will be appreciated by those skilled in the art that specifically described content is illustrative and nonrestrictive, should not limit the scope of the invention with this below.

The application scenarios of the train satellite positioning method based on collaborative integrity monitoring that the present embodiment provides is: known target train is from moment T _ato T _band AT STATION between A to B section space-time unique to be positioned in the operational plan that runs, correspondingly, by the off-line data in dedicated track map data base, satellite almanac data source as starting condition, receiver in Facility location unit adopts train to carry Beidou satellite navigation system (BeiDou Navigation Satellite System, BDS) receiver, aiding sensors in Facility location unit adopts odometer and gyro jointly to form, and combination is used for location Calculation and output.

As shown in Figure 1, the train satellite positioning method based on collaborative integrity monitoring that the present embodiment provides specifically comprises the following steps:

The basic data of S1, collection off-line, this off-line basic data comprises satellite almanac data, satellite health status data, dedicated track map datum, train operation plan data;

S2, basic data according to this off-line, predict the positioning error off-line protection limit value of each unique point distributed in space-time unique to be positioned, composition spatial-temporal distribution characteristic storehouse;

S3, by the navigational satellite receiver extract real-time Navsat settled on target train to the observation information of the target train operated in space-time unique to be positioned, and calculate target train positioning error in real time according to this observation information and protect limit value online;

S4, calculate in spatial-temporal distribution characteristic storehouse protect the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value;

S5, utilize target train positioning error to protect limit value online and protect limit value to compare, and according to comparative result real-time adjustment aim train locational decision scheme with the alarm threshold preset respectively with its time of running, mileage value characteristic of correspondence point location error off-line.

Wherein:

The detailed process of step S1 is:

Obtained by associated databases and cover T _ato T _bthe Navsat almanac of period and monitoring management information gained Navsat health status.Wherein, satellite almanac data is consistent with the effect that receiver on-board resolves gained almanac data, is and calculates all satellites three-dimensional coordinate position at a time by corresponding satellite orbit parameter.

Further, for realizing the integrity prediction of off-line, except satellite position information, also need corresponding train position as survey station target, this information is determined jointly by specific map database, train operation plan data.Wherein, map data base is obtained by the key point measurement carried out in advance, data processing, compilation process off-line, utilizes the crucial point set of track circuit to describe positional information, attributive character information, the terrain information of track, for i-th ' individual key point Q _i', corresponding numbering N _i', longitude λ _i', latitude φ _i', elevation H _i', circuit number Tr _i', mileage value S _i', terrain parameter collection Z _i', trackside facility attribute Y _i'deng; Train operation plan in basic data is by TDCS (Train Operation DispatchingCommand System, or railway despatch concentration (Centralized Train Control TDCS), CTC) system provides, and represents that train B station of dispatching a car from A station arrives the pilot process experienced with the broken line under " time m-mileage " coordinate system.For obtaining run location in transit (from track a permanent datum along orbital distance) basic description in time from train operation plan, the geographic coordinate information utilizing track map to provide and trackage information map, and the plan mileage of one dimension can be converted to the three-dimensional train position that can match with co-ordinates of satellite position.Broken line shown on the right side of in Fig. 2 describes train and amount to S from arrival B station, A station ₀operational plan in=33km section, wherein, on broken line, the coordinate of any point P illustrates train according to plan at T _pit is S that moment moves to mileage _pone dimension orbital position, utilize mileage S _pin road map data base, search package contains the minimum broken line of this mileage in-orbit by key point Q _m, Q _nmileage inverse S in proportion _pcorresponding longitude λ _p, latitude φ _p, elevation H _p, as shown on the left of Fig. 2, utilize this mapping mode time, spatial information to be associated, the planning location of train and satellite position be converted to same coordinate frame.

The detailed process of step S2 is:

Off-line prediction is implemented with this form of spatial-temporal distribution characteristic storehouse of positioning feature point error off-line protection limit value; with the corresponding relation of each unique point simulated target train time and mileage position in train operation plan in feature database; the relevant information composition data collection of each unique point, each unique point P (S _i, T _i) corresponding information comprises its mileage value S _i, time value T _iand the positioning error protection limit value predicted value E of this point _l(S _i, T _i).Wherein, choosing of unique point distribution adopts default time parameter σ _t, spatial parameter σ _scontrol, the present embodiment selects constant basis as the Appreciation gist of this feature database time, spatial resolution level.

Step S2 comprises following sub-step further:

S2.1, A station is amounted to S to B station is interval ₀the mileage scope of kilometer is according to spatial parameter σ _scarry out equal interval sampling, so can amount to N _s=[S ₀/ σ _s]+1 mileage value, [] represents and rounds.

S2.2, the primary train diagram line provided according to train operation plan, utilize T initial and arrival time _a, T _bthe corresponding plan moment solved corresponding to each mileage value respectively.Because train actual motion time-division and actual motion speed have uncertainty, so need according to the given time window T of this uncertainty _m, in order to comprise this uncertainty, time window T in the present embodiment _m=(T _b-T _a)/4, according to time parameter σ _tto this time window T _mcarry out equal interval sampling, each mileage value can obtain N respectively _t=[T _m/ σ _t] individual time value, then for each mileage value, from the plan moment of its correspondence, select N respectively from positive and negative both direction respectively _tindividual duration is associated with this mileage value, and so, then the mileage value of each unique point all has 2N _t+ 1 time value associates with it, thus forms each unique point P (S by mileage value and time value _i, T _i), by each unique point P (S _i, T _i) { P}, is formed " period by band " of train operation plan composition characteristic point set, as shown in Figure 3, its role is to adapt to the actual moving process of train and possible the departing from property of basic plan figure line.

S2.3, basis amount to N _s× (2N _t+ 1) mileage value, the time value of individual unique point, navigation satellite information judges that each satellite is at corresponding moment T _w, orbital position S _was observability (condition: satellite elevation angle is greater than limit value), duty (condition: the non-reported failures of satellite) during observation end, obtain the usable satellite collection { ID of each unique point of observation _q, X _q(t), Y _q(t), Z _q(t); Q=1,2 ..., M _i,p(t) }, wherein, ID _qfor satellite number, { X _q(t), Y _q(t), Z _q(t) } be the satellite q three-dimensional coordinate of t under ECEF coordinate system (Earth-Centered Earth-Fixed, ECEF), M _i,pt usable satellite quantity that () is this Feature point correspondence, if will implement integrity monitoring, then usable satellite quantity should meet and is more than or equal to 5, i.e. M _i,p>=5, in the present embodiment, usable satellite quantity is satisfied is more than or equal to 5.

S2.4, according to satellite position and unique point space time information, each unique point P (S distributed in space-time unique to be positioned according to following formula predictions _i, T _i) positioning error off-line protection limit value E _l(S _i, T _i):

${\mathrm{\&rho;}}_d\left(t\right)=\sqrt{{[X_q\left(t\right)-x_S\left(i\right)]}^2+{[Y_q\left(t\right)-y_S\left(i\right)]}^2+{[Z_q\left(t\right)-z_S\left(i\right)]}^2}+{\mathrm{\&xi;}}_q\left(t\right),q=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,M_{i,p}\left(t\right)$

$z_t={[{\mathrm{\&rho;}}_1\left(t\right),{\mathrm{\&rho;}}_2\left(t\right)\&CenterDot;\&CenterDot;\&CenterDot;,{\mathrm{\&rho;}}_{M_{i,p}\left(t\right)}\left(t\right)]}^T$

$x_t={[X\left(t\right),\stackrel{\&CenterDot;}{X}\left(t\right),\stackrel{\&CenterDot;\&CenterDot;}{X}\left(t\right),Y\left(t\right),\stackrel{\&CenterDot;}{Y}\left(t\right),\stackrel{\&CenterDot;\&CenterDot;}{Y}\left(t\right),Z\left(t\right),\stackrel{\&CenterDot;}{Z}\left(t\right),\stackrel{\&CenterDot;\&CenterDot;}{Z}\left(t\right)]}^T$

$S{E}_t=e_t^T{e}_t={\mathrm{\&epsiv;}}_t^T{R}_t^T{R}_t{\mathrm{\&epsiv;}}_t$

${\mathrm{\&epsiv;}}_t=z_t-H_t{\hat{x}}_{t,t-1}$

R _t＝I-H _tΦ _t

$H_t=\frac{\&PartialD;h\left(x_t\right)}{\&PartialD;x_t}$

${\mathrm{\&Phi;}}_t=P_{t,t-1}^{\mathrm{xz}}{\left(P_{t,t-1}^z\right)}^{-1}$

$E_L(S_i,T_i)={\mathrm{\&sigma;}}_0\&CenterDot;\underset{j}{\max }\sqrt{\frac{{\mathrm{\&Phi;}}_{t,1j}^2+{\mathrm{\&Phi;}}_{t,4j}^2+{\mathrm{\&Phi;}}_{t,7j}^2}{R_{t,\mathrm{jj}}}};$

In formula:

ρ _qt () is the pseudo range observed quantity of q satellite, [X _q(t), Y _q(t), Z _q(t)] be the satellite three-dimensional coordinate under ECEF coordinate system, [x _s(i), y _s(i), z _s(i)] be the three-dimensional position of unique point, can by its mileage value S _i, longitude λ _s(i), latitude elevation H _si () calculates through coordinate conversion;

ξ _qt () is the equivalent error of q satellite pseudorange observation, comprise ionospheric error, tropospheric error, clocking error, multipath error, receiver thermonoise aliquot;

Z _tfor the measurement vector that the pseudorange amount that current usable satellite is corresponding is formed;

X _tfor being led the filter status vector formed by three-dimensional coordinate position and single order thereof, second order;

SE _tfor test statistics, wherein, e _t, ε _tbe respectively the estimation of measuring value, predicated error;

I is unit matrix;

H _tfor equivalent measurement matrix, can be tried to achieve by nonlinear measurement function h (*) linearization approximate, for asking partial derivative symbol;

Φ _tfor filter gain matrix, be respectively the variance of measurement information, cross-covariance matrix, be respectively and x _tand z _trelevant matrix, by adopted nonlinear filter structures shape;

Unique point P (S _i, T _i) time value correspond to time variable t, σ ₀for the detection coefficient selected according to false dismissal probability, Φ _{t, 1j}represent filter gain matrix Φ _tthe 1st row, jth column matrix element, Φ _{t, 4j}represent filter gain matrix Φ _tthe 4th row, jth column matrix element, Φ _{t, 7j}represent filter gain matrix Φ _tthe 7th row, jth column matrix element, R _{t, jj}representing matrix R _tjth row, jth column matrix element.

Step S2 also comprises following sub-step:

The alarm threshold that S2.5, basis are preset, judges the availability effective range in the spatial-temporal distribution characteristic storehouse of each positioning feature point error off-line protection limit value, and as the reference of planning, adjustment train operation plan, detailed process is:

According to the application demand determination alarm threshold AL that train locational decision exports; control for the operation of train in interval; selection alarm threshold is AL=5.0m, will predict each positioning feature point error protection limit value E in gained spatial-temporal distribution characteristic storehouse in previous step _l(S _i, T _i) compare with AL, comparative result directly determines the availability attributes of unique point:

If (i) E _l(S _i, T _i)>=AL, this unique point mark availability attributes is V (unavailable);

(ii) if E _l(S _i, T _i) <AL, this unique point mark availability attributes is A (available).

As shown in Figure 3, by above-mentioned decision procedure, the property distribution space of integrity state can be determined in " period is by the band " of above-mentioned train operation plan, plays the effect predicted before train actual motion, as the reference of planning, adjustment train operation plan.

The detailed process of step S3 is:

Utilize the observation information of train-installed Facility location unit extract real-time Navsat, comprise the pseudorange etc. of satellite ephemeris, each satellites in view, adopt and each positioning feature point error protection limit value E in off-line predicted characteristics storehouse _l(S _i, T _i) the consistent computational logic of <AL, real-time online calculates T each time of running _tunder target train positioning error protect limit value E online _z(S _t, T _t), formula is as follows:

$E_Z(S_t,T_t)={\mathrm{\&sigma;}}_0\&CenterDot;\underset{j}{\max }\sqrt{\frac{{\mathrm{\&Phi;}}_{t,1j}^2+{\mathrm{\&Phi;}}_{t,4j}^2+{\mathrm{\&Phi;}}_{t,7j}^2}{R_{t,\mathrm{jj}}}};$

In formula, the T time of running of target train _tcorrespond to time variable t.

Each unique point protects the key distinction of limit value to be 2 points to protect limit value and off-line to predict online in line computation target column car positioning error: one is that the source of the information of satellitosis is different, is respectively online, is obtained by different pieces of information source under off-line mode; Two is that the moonscope amounts such as pseudorange are directly obtained by receiver, and does not need to adopt observational error model to carry out predicting and superposition calculation.

The detailed process of step S4 is:

In order to make at line computation E _z(S _t, T _t), off-line prediction E _l(S _i, T _i) realize in the process that compares the time, space attribute consistent, make this compare and there is practical significance, need to implement alignment function in conjunction with dedicated track database, wherein, onboard system carries out calculating in real time the dedicated track map data base that adopts and off-line predicts used being consistent.Concrete alignment procedures comprises the following steps:

S4.1, combination calculate E _z(S _t, T _t) process just calculate target train at current time T _tposition be projected on track map, tried to achieve projected position and then extrapolate projection mileage value corresponding to this projection coordinate position first calculation target train is at current time T _tposition method for application Nonlinear Kalman Filter solve filter status vector x _testimation solution, then therefrom propose the 1st, 4,7 vector element, form three-dimensional coordinate amount;

S4.2, with the mileage value that projects and the time of running T _tfor index, in off-line prediction gained spatial-temporal distribution characteristic storehouse, search for the unique point of mating with it:

If existing characteristics point P is (S _k, T _k) meet and T _t=T _k, then by this unique point P (S _k, T _k) positioning error protection limit value E _l(S _k, T _k) as protecting the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value E _l(S _t, T _t);

If not existing characteristics point P (S _k, T _k) meet and T _t=T _k, then determine to comprise in feature database t _tminimum quadrilateral, extract its four endpoint location (S _j', T _j'), (S _j'+ σ _s, T _j'), (S _j', T _j'± σ _t), (S _j'+ σ _s, T _j'± σ _t), wherein, endpoint location meets and meet T _j'≤ T _t≤ T _j'+ σ _tor T _j'-σ _t≤ T _t≤ T _j',

Utilize the approximate alignment position, center of this minimum quadrilateral, the off-line predicted value under adopting mean value method to estimate alignment case namely

${\hat{E}}_L(S_t^{*},T_t)=\frac{1}{4}[E_L(S_{j^{\&prime;}},T_{j^{\&prime;}})+E_L(S_{j^{\&prime;}}+{\mathrm{\&sigma;}}_S,T_{j^{\&prime;}})+E_L(S_{j^{\&prime;}},T_{j^{\&prime;}}\&PlusMinus;{\mathrm{\&sigma;}}_T)+E_L(S_{j^{\&prime;}}+{\mathrm{\&sigma;}}_S,T_{j^{\&prime;}}\&PlusMinus;{\mathrm{\&sigma;}}_T)];$

By the positioning error protection limit value extrapolated as protecting the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value E _l(S _t, T _t).

The detailed process of step S5 is:

E _l(S _t, T _t) <AL, E _z(S _t, T _t) meet while <AL two conditions and can be used for more adequately determining whether the metrical information that current train Facility location unit adopts satellite positioning receiver to provide can safely for final decision process.Auxiliary positioning sensor further by forming odometer and gyro carries out state self-inspection; comprise output detections and output quantity judgement; be in normally with the form of status word mark aiding sensors, abnormality, and then protect the differentiation of limit value jointly as the basis for estimation of locating information process structural adjustment with satellite position error.As shown in Figure 4, preset four kinds of process structures: the decision-making of (1) BDS/ aiding sensors tight integration, the decision-making of (2) aiding sensors location-independent, the decision-making of (3) BDS location-independent, (4) decision-making was lost efficacy.E _l(S _t, T _t) <AL, E _z(S _t, T _t) verified in parallel of <AL two conditions and this condition of auxiliary positioning sensor states word, in conjunction with the processing scheme that a upper moment is adopted, judged current time T _twhich kind of information processing scheme should be adopted down for the generation of final position decision-making, specifically can be divided into:

If (i) E _l(S _t, T _t)>=AL, E _z(S _t, T _tany one situation of)>=AL occurs, then train position decision process isolation BDS observation data, will from a upper moment T _t-1scheme 1. process cooperating measure to scheme 2., or from a upper moment T _t-1scheme be 3. converted to failure mode 4.;

(ii) if E _l(S _t, T _t) <AL, E _z(S _t, T _t) <AL two kinds of situations meet simultaneously, then in train position decision process, BDS observation data is upstate, locating information process organization plan will from a upper moment be converted to scheme 1. or scheme 3..

When adopt scheme 1. time, auxiliary positioning sensor can calculate gained train warp by layout, Position Latitude information pushes away pseudorange amount in conjunction with BDS satellitosis is counter, and then resolve pseudorange amount with BDS receiver and carry out error reckoning, its margin of error can be adopted to form state vector and to set up system equation and measurement equation, adopt Nonlinear Kalman Filter to realize state estimation under tight integration pattern and decision making package.

When the scheme of employing 2. time, directly the layout result of calculation of employing auxiliary positioning sensor is as train position decision-making and output.

When the scheme of employing 3. time, directly can to adopt in step S4 positioning error protection limit value E _z(S _t, T _t) when calculating the state estimation result in advance that carries out for train position decision-making and output.

When proceed to scheme 4. time, show that current BDS receiver or auxiliary positioning sensor information all exist security, available risk, do not reoffer Facility location and export, but send corresponding failure state alarm mark.

The method that the present embodiment provides adds off-line, static information of forecasting to the effect of contraction of decision process in train position decision-making treatment process, the integrity monitoring degree of depth in train location Calculation is got a promotion, embodies off-line procedure and working in coordination with at line process.Given output is tending towards conservative, but can guarantee that satnav observation information that is invalid or that may there is risk is not introduced into Facility location process, thus effectively promote autonomy-oriented adjustment capability and the level of security of train location, provide powerful support for Train Detection and Identification, train is followed the trail of close to the multiple system relevant to traffic safety application such as early warning.

As shown in Figure 5, the train global position system based on collaborative integrity monitoring of the execution said method that the present embodiment provides, comprise: surface-mounted satellite data obtains server, integrity status predication server, track map data server, information of forecasting distributing terminals, and the diagnostic process terminal of settling ON TRAINS and train position decision package;

Satellite data obtains server, for inquiring about satellite almanac data and satellite health status data by network request, detecting that the situation of Data Update downloads satellite almanac and satellite health status data, and framing is stored to internal storage unit according to agreement again.

Track map data server, store for the track map data base that obtained by manual operation corresponding to each train number, tracking road map carries out upgrading and consistency maintenance.

Integrity status predication server, for by the network port to existing railway dispatching and command system, satellite data obtains server, track map data server solicited message, to obtain train operation plan data, satellite health status data, the writing data into memory such as dedicated track map datum, by arithmetic logic unit according to time parameter, the settings such as spatial parameter, calculated off-line generates each positioning feature point error off-line protection limit value of distribution in space-time unique to be positioned, by each positioning feature point error off-line protection limit value composition spatial-temporal distribution characteristic storehouse distributed in space-time unique to be positioned, and send forwarding instruction to information of forecasting distributing terminals.

Information of forecasting distributing terminals, for according to received forwarding instruction, utilize the Bus Schedules of resolving as foundation, the spatial-temporal distribution characteristic storehouse generated by integrity status predication server sends to the level diagnosis processing terminal of target train by cordless communication network.

Level diagnosis processing terminal, for receiving spatial-temporal distribution characteristic storehouse by cordless communication network before target train departure, after launch train runs, Navsat almanac data is extracted from satellite navigation receiver in each sampling period, the status word information of observation information and auxiliary positioning sensor, adopt the computational logic consistent with integrity status predication server to calculate current time of running target train positioning error and protect limit value online, and with current mileage, temporal information is reference, carry out searching for and aiming in offline feature storehouse, extract characteristic of correspondence point location error off-line protection limit value to compare, current locating information processing scheme is determined based on predetermined switching law according to comparative result, generate project setting instruction, and then real-time adjustment aim train locational decision scheme.

Train position decision package, for the target train locational decision scheme after the adjustment that real-time reception diagnostic process terminal sends, complete train position decision-making computing and utilize the quantity of state such as current location, speed of target train to generate position output data frame, externally being sent by data-interface.

Wherein:

As shown in Figure 6, diagnostic process terminal comprises:

Moonscope messaging interface module, for receive from satellite navigation receiver send satellite to the observation information of the target train operated in space-time unique to be positioned;

Information of forecasting receiver module, for receiving by cordless communication network and storing the spatial and temporal distributions storehouse sent by information of forecasting distributing terminals before train departure;

Integrity monitoring computing module, for according to the observation information of satellite to the target train operated in space-time unique to be positioned, calculates target train positioning error in real time and protects limit value online;

Information calculates module, protects the time of running of limit value online, mileage value characteristic of correspondence point location error off-line protects limit value for calculating in spatial-temporal distribution characteristic storehouse with target train positioning error;

Diagnosis decision-making module, protect limit value online for utilizing target train positioning error and protect limit value to compare with alarm threshold respectively, according to comparative result real-time adjustment aim train locational decision scheme with its time of running, mileage value characteristic of correspondence point location error off-line;

Information injection module, for carrying out interface with the location processor of train position decision package, sends to it locating information process structural adjustment instruction generated.

The satellite data acquisition server that ground subsystem comprises, integrity status predication server, track map data server, information of forecasting distributing terminals all can adopt two cover configuration, the reliability of increase system and failure tolerant ability, thus deposit in the case of a fault at individual equipment, system still can adopt backup device to continue to realize off-line prediction and Information issued according to above-mentioned flow process.

The diagnostic process terminal that onboard subsystem comprises can adopt two cover configuration, two terminal concurrent workings, adopt on all four hardware configuration and software logic, train position decision package receives the injection information of two terminals simultaneously, adopt consistance decision logic, only in the on all four situation of locating information process structural adjustment instruction received from two terminals, perform corresponding project setting and operate and carry out information processing and framing output.

Track map data server and the vehicle-mounted train position decision package on ground store on all four dedicated track map data base, be respectively used to implement off-line prediction and calculation and inline diagnosis calculating, car-ground is regularly carried out database maintenance and verification, gives a warning and carry out synchronous in time when verifying unsuccessfully.Because of the reasons such as track circuit construction, transformation cause cartographic information to change when, map data base will carry out upgrade maintenance in time by operating personnel, guarantee that system provides effective service.

Obviously; the above embodiment of the present invention is only for example of the present invention is clearly described; and be not the restriction to embodiments of the present invention; for those of ordinary skill in the field; can also make other changes in different forms on the basis of the above description; here cannot give exhaustive to all embodiments, every belong to technical scheme of the present invention the apparent change of extending out or variation be still in the row of protection scope of the present invention.

Claims (10)

1., based on a train satellite positioning method for collaborative integrity monitoring, it is characterized in that, the method comprises the steps:

S1, collection satellite almanac data, satellite health status data, dedicated track map datum and train operation plan data;

S2, each data gathered according to step S1, predict the positioning error off-line protection limit value of each unique point distributed in space-time unique to be positioned, composition spatial-temporal distribution characteristic storehouse;

S3, extract real-time Navsat to the observation information of the target train operated in space-time unique to be positioned, and calculate target train positioning error in real time according to described observation information and protect limit value online;

S4, calculate in described spatial-temporal distribution characteristic storehouse protect the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value;

S5, utilize target train positioning error to protect limit value online and protect limit value to compare with alarm threshold respectively with its time of running, mileage value characteristic of correspondence point location error off-line, and according to comparative result real-time adjustment aim train locational decision scheme.

2. the train satellite positioning method based on collaborative integrity monitoring according to claim 1, it is characterized in that, dedicated track map datum in step S1 is obtained by map data base, and this map data base comprises positional information, attributive character information, the terrain information of each track key point.

3. the train satellite positioning method based on collaborative integrity monitoring according to claim 1, it is characterized in that, step S2 comprises following sub-step further:

S2.1, according to train operation plan to the mileage scope S in space-time unique to be positioned ₀according to spatial parameter σ _scarry out equal interval sampling, obtain S ₀/ σ _s+ 1 mileage value;

S2.2, be engraved in given time window T to during plan corresponding to each mileage value respectively _minterior according to time parameter σ _tcarry out equal interval sampling, obtain each mileage value distinguish 2 corresponding (T _m/ σ _t)+1 time value;

S2.3, be based upon feature point set { P}, wherein i-th unique point P (S of distribution in space-time unique to be positioned _i, T _i) by the mileage value S of this unique point _iwith time value T _iform;

S2.4, determine each unique point P (S _i, T _i) available observation satellite;

S2.5, predict each unique point P (S distributed in space-time unique to be positioned _i, T _i) positioning error off-line protection limit value E _l(S _i, T _i), formula is as follows:

R _t＝I-H _tΦ _t；

$E_L(S_i,T_i)={\mathrm{\&sigma;}}_0\&CenterDot;\underset{j}{\max }\sqrt{\frac{{\mathrm{\&Phi;}}_{t,1j}^2+{\mathrm{\&Phi;}}_{t,4j}^2+{\mathrm{\&Phi;}}_{t,7j}^2}{R_{t,\mathrm{jj}}}};$

In formula, I is unit matrix; H _tfor equivalent measurement matrix; Φ _tfor filter gain matrix; Unique point P (S _i, T _i) time value T _icorrespond to time variable t; σ ₀for the detection coefficient selected according to false dismissal probability; Φ _{t, 1j}for filter gain matrix Φ _tthe 1st row, jth column matrix element; Φ _{t, 4j}for filter gain matrix Φ _tthe 4th row, jth column matrix element; Φ _{t, 7j}for filter gain matrix Φ _tthe 7th row, jth column matrix element; R _{t, jj}for matrix R _tjth row, jth column matrix element.

4. the train satellite positioning method based on collaborative integrity monitoring according to claim 3, is characterized in that, also comprise the steps: after step S2.5

The alarm threshold AL that S2.6, basis are preset, judges the availability effective range in the spatial-temporal distribution characteristic storehouse of each positioning feature point error off-line protection limit value, and as the reference of planning, adjustment train operation plan, formula is:

If unique point P _ipositioning error off-line protection limit value E _l(S _i, T _i)>=AL, then this unique point P (S _i, T _i) mark availability be invalid;

If unique point P _ipositioning error off-line protection limit value E _l(S _i, T _i) <AL, then this unique point P (S _i, T _i) mark availability be effective.

5. the train satellite positioning method based on collaborative integrity monitoring according to claim 1, is characterized in that, calculates target train positioning error and protect the formula of limit value to be online described in step S3:

R _t＝I-H _tΦ _t；

$E_Z(S_t,T_t)={\mathrm{\&sigma;}}_0\&CenterDot;\underset{j}{\max }\sqrt{\frac{{\mathrm{\&Phi;}}_{t,1j}^2+{\mathrm{\&Phi;}}_{t,4j}^2+{\mathrm{\&Phi;}}_{t,7j}^2}{R_{t,\mathrm{jj}}}};$

In formula, E _z(S _t, T _t) protect limit value online for target train positioning error; I is unit matrix; H _tfor equivalent measurement matrix; Φ _tfor filter gain matrix; The T time of running of target train _tcorrespond to time variable t; σ ₀for the detection coefficient selected according to false dismissal probability; Φ _{t, 1j}represent filter gain matrix Φ _tthe 1st row, jth column matrix element; Φ _{t, 4j}represent filter gain matrix Φ _tthe 4th row, jth column matrix element; Φ _{t, 7j}represent filter gain matrix Φ _tthe 7th row, jth column matrix element; R _{t, jj}representing matrix R _tjth row, jth column matrix element.

6. the train satellite positioning method based on collaborative integrity monitoring according to claim 1, it is characterized in that, step S4 comprises following sub-step further:

S4.1, just calculation target train operationally carve T _tposition this position is projected on track map, tries to achieve projected position and then extrapolate projection mileage value corresponding to this projection coordinate position

S4.2, with the mileage value that projects and the time of running T _tfor index, in described spatial-temporal distribution characteristic storehouse, search for the unique point of mating with it:

If existing characteristics point P is (S _k, T _k) meet and T _t=T _k, then by this unique point P (S _k, T _k) positioning error protection limit value E _l(S _k, T _k) as protecting the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value E _l(S _t, T _t);

If not existing characteristics point P (S _k, T _k) meet and T _t=T _k, then mean value method is adopted to calculate off-line predicted value as protecting the time of running of limit value online with target train positioning error, mileage value characteristic of correspondence point location error off-line protects limit value E _l(S _t, T _t), prediction equation is as follows:

${\hat{E}}_L(S_t^{*},T_t)=\frac{1}{4}[E_L(S_{j\&prime;},T_{j\&prime;})+E_L(S_{j\&prime;}+{\mathrm{\&sigma;}}_S,T_{j\&prime;})+E_L(S_{j\&prime;},T_{j\&prime;}\&PlusMinus;{\mathrm{\&sigma;}}_T)+E_L(S_{j\&prime;}+{\mathrm{\&sigma;}}_S,T_{j\&prime;}\&PlusMinus;{\mathrm{\&sigma;}}_T)];$

$S_{j\&prime;}\&le;S_t^{*}\&le;S_{j\&prime;}+{\mathrm{\&sigma;}}_S;$

T _j'≤ T _t≤ T _j'+ σ _tor T _j'-σ _t≤ T _t≤ T _j';

In formula, σ _tfor time parameter, σ _sfor spatial parameter.

7. the train satellite positioning method based on collaborative integrity monitoring according to claim 1, is characterized in that, comprises following sub-step further described in step S5 according to comparative result real-time adjustment aim train locational decision scheme:

If target train positioning error protects limit value online and protect arbitrary value in limit value to be more than or equal to default alarm threshold with its time of running, mileage value characteristic of correspondence point location error off-line, then target train locational decision scheme is the observation data of isolation satnav, directly adopts auxiliary positioning sensing data to calculate target column truck position;

If target train positioning error protects limit value online and protect limit value to be all less than default alarm threshold with its time of running, mileage value characteristic of correspondence point location error off-line; then target train locational decision scheme is comprehensive employing satellite positioning receiver and auxiliary positioning sensors observe data, merges estimation target column truck position by tight coupling.

8. perform the train global position system based on collaborative integrity monitoring as method as described in arbitrary in claim 1-7, it is characterized in that, this system comprises:

Satellite data obtains server, for gathering satellite almanac data and satellite health status data;

Integrity status predication server, for receiving the train operation plan data that TDCS sends, according to train operation plan data, satellite health status data, dedicated track map datum, each positioning feature point error off-line protection limit value that prediction distributes in space-time unique to be positioned, composition spatial-temporal distribution characteristic storehouse;

Be placed in the level diagnosis processing terminal on target train, for according to the observation information of satellite to the target train operated in space-time unique to be positioned, real-time calculating target train positioning error protects limit value online, the time of running protecting limit value with target train positioning error is online calculated in spatial-temporal distribution characteristic storehouse, mileage value characteristic of correspondence point location error off-line protection limit value, and utilize target train positioning error protect online limit value and with its time of running, mileage value characteristic of correspondence point location error off-line protection limit value compares with alarm threshold respectively, according to comparative result real-time adjustment aim train locational decision scheme,

Be placed in the train position decision package on target train, for gathering the metrical information of satellite navigation receiver and aiding sensors, according to the target train locational decision scheme implementation train locational decision after adjustment in real time, export current location and the speed of target train in real time.

9. the train global position system based on collaborative integrity monitoring according to claim 8, it is characterized in that, this system also comprises: information of forecasting distributing terminals, for spatial-temporal distribution characteristic storehouse to be sent to the level diagnosis processing terminal of target train by cordless communication network.

10. the train global position system based on collaborative integrity monitoring according to claim 8, is characterized in that, described level diagnosis processing terminal comprises:

Moonscope messaging interface module, for receiving the observation information of satellite to the target train operated in space-time unique to be positioned;

Information of forecasting receiver module, for receiving spatial-temporal distribution characteristic storehouse;

Integrity monitoring computing module, for according to the observation information of satellite to the target train operated in space-time unique to be positioned, calculates target train positioning error in real time and protects limit value online;

Information calculates module, protects the time of running of limit value online, mileage value characteristic of correspondence point location error off-line protects limit value for calculating in spatial-temporal distribution characteristic storehouse with target train positioning error;

Diagnosis decision-making module, protect limit value online for utilizing target train positioning error and protect limit value to compare with alarm threshold respectively, according to comparative result real-time adjustment aim train locational decision scheme with its time of running, mileage value characteristic of correspondence point location error off-line;

Information injection module, for being sent to train position decision package by the target train locational decision scheme after adjustment in real time.