CN101692315A - Method for analyzing high precision 4D flight trajectory of airplane based on real-time radar data - Google Patents

Abstract

The invention discloses a method for analyzing high precision 4D flight trajectory of an airplane based on real-time radar data, comprising the following steps: establishing a 4D fight trajectory theoretical model based on the airplane performance; establishing a 4D fight trajectory empirical model by mining and analyzing based on the historical fight trajectory data of the airplane types in the above step; combining the theoretical model and the empirical model and quantizing various influencing factors in the flight process into adjustable parameters; and correcting the 4D flight trajectory mixed models by radar data to form the final flight trajectory of the airplane. The invention takes the airplane performance into consideration and establishes the theoretical model by performance library parameters on the basis of a standard flight procedure, thus having high reliability; the invention forms the empirical model by analyzing the historical data, thus having high reality; the invention generates the 4D flight trajectory most similar to the actual flight and can adjust the parameters to form a plan trajectory aiming at the environment of each flight, thus high sensitivity; and the invention correct the plan trajectory by introducing the real-time radar data, thus having high accuracy.

Description

Method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data

Technical field

The present invention relates to a kind of Air Traffic Control Automation System and flight flow management system, the high precision 4D based on real-time radar data of the flight planning track analysis that particularly a kind of real-time acquisition is accurate, continuous, level and smooth (fourdimension) aircraft track analytical approach of being used for.

Background technology

At present, in the traffic control automated system aloft, the analytical approach of flight path mainly contains three kinds of great-circle track analysis and isogonism track analysis method, adaptive Kalman filter (Kalman) algorithm and interaction multi-model (IMM) algorithms, all is following position and the time by way point of analyzing in-flight in real time.

Great-circle track is analyzed and isogonism track analysis method, belongs to dynamic track analysis method, is applicable to air route and the termination environment course line of advancing to leave the theatre respectively.But under short-range situation, (as the distance of the great-circle track between Beijing and the Lhasa is 1380 nautical miles for great-circle track distance and isogonism track distance approximately equal, the isogonism track distance is 1386.3 nautical miles), therefore for convenience of calculation, adopt the isogonism track analysis usually.The aircraft of setting flight when analyzing uses following speed variation rule in addition: be to do with cruise speed to fly at a constant speed on the air route, do even decelerating flight on (the advancing near) course line of marching into the arena, do the even flight of quickening on the course line of leaving the theatre.Information such as the current location that provides according to real-time radar data, time, speed are obtained and following way point distance, and then were analyzed the way point time.

Kalman filtering algorithm (Kalman) is a kind of application algorithm very widely, and the criterion of filtering and analysis is the root-mean-square error minimum.Since it can based target motor-driven and observation noise Automatic Model Selection gain sequence, and can automatically adapt to the variation of testing process, can measure the precision of estimating easily by covariance matrix, by residual error (new breath wherein, be the poor of an observed reading and a step assay value) vectorial variation, can judge whether the object module of former supposition and the kinetic characteristic of realistic objective meet, and with this as a kind of means of motor-driven detection and motor-driven identification etc., therefore, it has obtained good development and application at trajectory analysis with in following the tracks of.

More than two kinds of algorithms all be method at single model, be that each has only a model in action constantly, the shortcoming of these methods is when object model is changed saltus step to take place, and causes motor-driven detection hysteresis or estimation analysis precision are not reached higher requirement.

Interaction multi-model algorithm (IMM), application in civil aviaton also is in the research and development, it is that a plurality of wave filter parallel runnings that are complementary with model are arranged simultaneously with the key distinction of in the past filtering algorithm, renewal by Markov chain and new breath implementation model probability reaches the requirement of the soft conversion of model.And in algorithm in the past, though also use the different motion state of the corresponding targets of different models, each has only a model filtering device in action constantly usually, according to statistical test dbjective state is monitored between the different model filtering devices and switches.The multi-model algorithm is a kind of recursive algorithm, the corresponding different process noise level of each model, promptly corresponding a kind of motion model.

And the IMM model algorithm is to move a plurality of models each moment simultaneously, it has solved the above two uncontinuities when model conversion, but the model number is fixed in this method, model description is constant, just structure is changeless, this causes unnecessary calculated amount to increase with regard to making the performance of algorithm not necessarily obviously improve with the increase of the number of model on the contrary.

Make a general survey of existing track analysis algorithm, the current motion state with the flight object is that basic point has been set up motion model mostly, the correction problem of also having considered real-time radar data that has, but weak point is all arranged separately.1, analysis precision is low, though dynamically plan way point information has been considered in track analysis, adopted dynamic data to revise in real time, but, do not consider the flight performance data of flight object, the 4D track analysis model idealization of setting up adopts the velocity amplitude of supposition to analyze again, causes analysis precision to reduce; 2, the analysis period is long more, analytical error is big more, Kalman filtering algorithm and IMM algorithm all are to be observation station with the real motion state, but with the complete renunciation of flight planning, do not consider the to fly performance of object, and can only estimate the value in a following sampling period each time, along with the growth of analytical cycle, error is increasing.

Summary of the invention

Goal of the invention: technical matters to be solved by this invention is at the deficiencies in the prior art, provide on a kind of basis based on flight planning, aircraft performance parameter, historical flying quality, utilize real-time radar data that the flight planning flight path is analyzed, promptly have orientation, distance, highly, the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data of the flight aircraft movement locus of information such as speed, time.。

Technical scheme: the invention discloses a kind of method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data, this method may further comprise the steps:

Step (1) is set up the 4D flight path theoretical model based on aeroplane performance;

Step (2) based on the historical flight path data of aircraft model in the step (1), is set up 4D flight path empirical model by mining analysis;

Step (3) combines theoretical model and empirical model, and the various influence factors in the flight course are quantized to become adjustable parameter, promptly obtains an accurate 4D flight path mixture model of considering the extraneous factor variation based on aeroplane performance again;

Step (4) with the described 4D flight path of radar data correction mixture model, forms final aircraft flight flight path.

The flight path of 4D described in the present invention be have orientation, distance, highly, the flight aircraft movement locus of information such as speed, time.

Beneficial effect: the invention has the advantages that: at first, the present invention has considered the performance of aircraft, on the basis of standard mission program, has set up theoretical model with performance library parameters, this model and flight to as if be closely connected, reliable with respect to existing processing; The second, the present invention analyzes historical data, has formed the model of experience, and is relatively truer comparatively speaking; The 3rd, will integrate theory with practice, produce 4D track, and can regulate the formation desired track to parameter at the environment of each flight near Live Flying, relatively sensitiveer; At last, introduce real-time radar data desired track is revised, make the desired track accuracy improve.

Description of drawings

Below in conjunction with the drawings and specific embodiments the present invention is done further to specify, above-mentioned and/or otherwise advantage of the present invention will become apparent.

Fig. 1 is the method for analyzing high precision 4 D flight trajectory of airplane process flow diagram that the present invention is based on real-time radar data.

Fig. 2 sets up process flow diagram for 4D track analysis among the present invention.

Fig. 3 calculates synoptic diagram for point of contact, aircraft turning angle in the Turn Models among the present invention.

Fig. 4 is effect main external force aboard in the aircraft longitudinal movement process among the present invention.

Embodiment:

As depicted in figs. 1 and 2, the invention discloses a kind of method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data, this method may further comprise the steps:

Step (1) is set up the 4D flight path theoretical model based on aeroplane performance;

Step (2) based on the historical flight path data of aircraft model in the step (1), is set up 4D flight path empirical model by mining analysis;

Step (3) combines theoretical model and empirical model, and the various influence factors in the flight course are quantized to become adjustable parameter, promptly obtains an accurate 4D flight path mixture model of considering the extraneous factor variation based on aeroplane performance again;

Step (4) with the described 4D flight path of radar data correction mixture model, forms final aircraft flight flight path.

Particularly, the present invention comprises following content:

1, based on the 4D modelling of aeroplane performance

1.1 data are prepared

Set up accurate 4D flight path theoretical model, the information that relates to has:

1) flight planning information comprises flight number, type, E.T.D(estimated time of departure), landing airport etc.

2) spatial domain information, standard mission program, course line, air route and at the course line of advancing to leave the theatre on landing airport.

3) model-performance parameter, the performance data of type correspondence, this is that theoretical model is set up main factor, comprises aircraft basis geometric parameter, aircraft basic motive mathematic(al) parameter (lift coefficient, resistance coefficient, the climb rate), aircraft configuration information (industry dead weight capacity, center of gravity), engine parameter, speed envelope curve, fuel consume.

1.2 cross track model algorithm

Flight f is made up the 4D track, at first plan information is handled: at first according to landing airport and settle the standard instrument leave the theatre course line SID (Standard Instrument Departure) and Standard Instrument Approach course line STAR (Standard Terminal Arrival Route) of the course line distribution principle of advancing to leave the theatre of airport correspondence separately, secondly according to constraint condition refinements such as air route information, cruising altitude by way of way point and leap height, preliminary set up three-dimensional flight path framework, and cross track be exactly through the projection of way point at surface level.Here the cross track of obtaining only is a plan position information, and the temporal information of location point must be calculated after obtaining velocity profile.

The air route and the course line of advancing to leave the theatre all are to be made of a series of air routes landmark point, are WP with the sequence definition of putting _i, number is n, determines radius of turn r according to flying speed _i, and definition constrained type type _i: initial point constraint, general point constraint and terminal constraint, simplified model is:

G _f={ wp ₁(x ₁, y ₁, type ₁, r ₁) ..., wp _i(x _i, y _i, type _i, r _i) ..., wp _n(x _n, y _n, type _n, r _n) formula (1)

If the initial point constraint is then flown nonstop to down a bit; The general point constraint will be considered Turn Models; Terminal constraint is exactly that head must line up with runway during through this for aircraft.According to various constraints, it is as follows to make up Turn Models:

Radius of turn

G is an acceleration of gravity, and flying speed is v during turning, and the pitch angle is φ.As Fig. 3, the coordinate of known adjacent three way points and corresponding radius of turn r obtain turning point P, the Q coordinate gets final product.The course angle of aircraft flight is with reference to zero degree with direct north, increases along clockwise direction, is 359 degree to the maximum.Use α ₁₂The course angle of representing last leg, α ₂₃The course angle of one leg, expression back according to 3 line relation, judges that aircraft is to turn left, turn right or fly nonstop to; In conjunction with the course angle of two sections flights with turn to, obtain aircraft turning angle Δ α; According to turning angle and radius of turn, obtain the distance of P point-to-point i: D _{P, i}=r * tan (Δ α/2), the coordinate that draws turning point P is:

$\left\{\begin{array}{c}{x}_p=x_i-D_{p,i}\×{\cos \mathrm{\α}}_{12}\\ y_p=y_i-D_{P,i}\×{\sin \mathrm{\α}}_{12}\end{array}\right.$

The Q point coordinate is:

$\left\{\begin{array}{c}{x}_Q=x_i+D_{P,i}\×{\cos \mathrm{\α}}_{23}\\ y_Q=y_i+D_{P,i}\×{\sin \mathrm{\α}}_{23}\end{array}\right.$

Final horizontal trajectory is made up of a series of orderly points:

G _f={ wp ₁(x ₁, y ₁) ..., P _i(x _i, y _i), Q _i(x _i, y _i) ..., wp _n(x _n, y _n) formula (2)

1.3 path in elevation model algorithm

According to most typical five stages of flight: take off, climb, flatly fly, descend, land, we also are divided into the calculating that five submodels are handled above-mentioned five paragraphs respectively with whole path in elevation computation model accordingly.At each paragraph, the method that its submodel adopts Newton second law and conservation of energy principle to combine is listed equation, use infinitesimal knowledge and divided the segment of some five equilibriums each stage, the substitution aircraft is in the performance parameter of each stage correspondence, find the solution the rising of every segment or rate of descent, corresponding distance and time, adding up at last forms altitude profile and velocity profile, and obtains flight time accumulation.

Vertical mould mainly characterizes the lengthwise movement of aircraft, this process effect main external force such as Fig. 4 aboard: motor power F, direction becomes engine installation angle along engine axis with fuselage axis X-shaped

Lift L is perpendicular to flying speed v, and resistance D is parallel to flying speed, and gravity G is downward, and α is the angle of attack, and θ is a flight-path angle.The sliding race process of takeoff phase, sliding speed is parallel to surface level, and will consider the friction force f of tire and runway face.

Use the kinetics equation of Newton second law foundation in the sliding race stage along runway heading:

Formula (3)

Wherein μ is the runway friction factor, and m is an aircraft weight, and g is an acceleration of gravity,

Be percentage speed variation,

Be the runway gradient, be worth very little so gravity component in the horizontal direction

All stages after liftoff before land, but binding kinetics and conservation of energy principle are set up equation:

Formula (4)

$(F-D)v_{\mathrm{TAS}}=\mathrm{mg}\frac{\mathrm{dh}}{\mathrm{dt}}+{\mathrm{mv}}_{\mathrm{TAS}}\frac{{\mathrm{dv}}_{\mathrm{TAS}}}{\mathrm{dt}}$ Formula (5)

Wherein h is an aircraft place height, v _TASBe the flight true air speed.Above mention in all formula

C _DAnd C _LBe given resistance coefficient of aeroplane performance parameter and lift coefficient, ρ is an atmospheric density, and S is a wing area.

In the starting heats stage, fuel consume can think that seldom aircraft weight is approximate constant, and aircraft angle of attack is assumed to the shutdown angle of attack, so C _DAnd C _LCan think and immobilize.The flight performance parameter substitution formula (3) of this stage correspondence is obtained acceleration a, with given flying speed V _LOFThe substitution formula:

For improving precision, utilize numerical integrating to slide and sue for peace after the race section is divided into the independent integration of segment, draw ground roll-out distance and sliding race time.

(airbus flight operation is supported and the customer service of course line assistance portion according to reference " grasp aeroplane performance ", in January, 2002, http://www.docin.com/p-5457290.html) description, the standard section that climbs comprises: quicken to climb to from departing and specify indicated airspeed V1; Etc. the indicated airspeed specified altitude assignment H1 (being generally 10000 feet) that climbs; Put down and fly to accelerate to indicated airspeed V2; Etc. the indicated airspeed transition altitude H2 (this is highly located corresponding flight Mach number and equates with the true air speed of aircraft) that climbs; At last to wait Mach number cruising altitude of climbing.V1 wherein, V2, H1, the corresponding every kind of type of H2 can be from reference obtains in the tables of data (this partial data also is included in the performance parameter) of various type correspondences.From the type section regulation as can be seen, each segment that climbs has all provided concrete initial and stop value, be exactly and the starting condition of next section of end condition of the last period, so every section the free and distance of institute is calculated in segmentation, Zong last addition is with regard to T.T. and distance for the whole section of climbing.With the performance parameter formula (5) of ramp-up period correspondence, obtain escalating rate, obtain the used time behind certain height that rises with numerical integration.Utilize formula (4) to obtain percentage speed variation again, binding time and percentage speed variation are obtained the horizontal range of every segment.Consider the fuel consume coefficient, obtain the amount of fuel that per stage consumes, every segment weight is constant in approximate thinking, following a bit of weight is that the preceding paragraph weight deducts amount of fuel.At last, result's summation of every segment is obtained time, distance, the speed of ultimate demand.Certainly the Duan Yue that divides is many, and precision is high more.

The climb flight course of section of standard decline section and the standard of describing in this data is just in time opposite, at first to wait Mach number to drop to transition altitude, again with etc. indicated airspeed drop to the appointment approach altitude, then level decelerates to the regulation velocity of approach, slows down at last and lands.The same with climbing, the parameter that relates to all has regulation.Equally, utilize formula (5) to obtain rate of descent, obtain the used time by integration, utilize formula (4) to obtain percentage speed variation again, binding time is obtained distance.

Cruising phase is fairly simple, adopts equal altitudes substantially and waits the Mach number mode to carry out.Stipulated most economical cruise speed or Mach number in the performance parameter, directly used kinematical equation and just can obtain cruise time, distance, amount of fuel.But when this stage runs into course change, need the transition altitude layer, because the division of spatial domain height layer is relevant with the thing course.At this moment require aircraft to drop to specified altitude assignment, obtain voyage and time with the maximum gliding angle of regulation.

1.44D flight path is comprehensive

According to from liftoff to landing the horizontal model and the vertical mould of whole flight course, with aircraft the horizontal range s of process be basic parameter, can in cross track model of setting up and path in elevation model, find corresponding 2 points respectively.Promptly in the cross track model, represent aircraft to be the some p1 of s, and in the path in elevation model, represent aircraft to be the some p2 of s from the beginning horizontal flight distance of taking off from the beginning horizontal flight distance of taking off.With such two spot correlations connection, latitude and longitude coordinates and the height of p2 and the flight time at that time of obtaining p1 respectively, form 4D volume coordinate point.So just can set up out a complete 4D flight path.

2, the 4D experience track that excavates based on historical data

Corresponding weather information (GRTB message) is carried out for a long time the admission of (can be month or a year or longer) during at first to comprehensive flight path output information of control automated system (this paper adopts the main DOD information with system's output of present district pipe) and implementing plan, and these data are put in order as data source.

2.1 the experience track excavates

Every frame DOD information of district's guard system output is a collection of aerial target status information and following desired track information at that time.By playing the historical DOD information of a period of time (such as one month), Frame is contrasted one by one, flight path by flight number, landing airport, actual time of departure (form is the date Hour Minute Second), type and flight path information number all is encapsulated in this flight object set as the sign of unique object position sequence, height sequence, the time corresponding sequence with whole flight course, promptly forms the historical track of flight.To same flight object, in a period of time (such as one month), have repeatedly flight, so that the object set that encapsulates has is a plurality of, a plurality of historical tracks are arranged promptly:

T (obj, i)={ x _j, y _j, h _j, t _j, v _j(i=(1,2 ..., n), j=(1,2 ..., m)) and formula (6)

Wherein obj is the flight object, and n is the flight total degree, and equation the right is position, time, the velocity information of each tracing point, and m is the tracing point number of each flight path.To a plurality of track set of each flight object,, draw the experience track of this object with every data difference weighted.Processing procedure comprises parameter alignment and weighted.

1) parameter alignment

A plurality of track set to same target, at first calculate the distance between first tracing point and the original base in each set, starting point in the big set of selected distance is as the position reference point, and the point in other set before this position is rejected, and forms approximate unified origin information; Secondly the time of all starting points is located zero point, the time sequencing skew of back tracing point.

2) weighted

To single object, with the position in a plurality of historical tracks set after the parameter alignment, time, speed respectively weighted draw the experience track of this object.

$T_{\mathrm{obj}}^e=\{{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(w_i\×x_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(u_i\×y_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(r_i\×h_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(s_i\×t_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(z_i\×v_i)}{n}\right)}_j\}$ Formula (7);

W wherein _i, u _i, r _i, s _i, z _iBe the weights coefficient, j is a tracing point, adds up to the effective number after parameter is alignd.The method of determining is: determine good plan air route according to plan information, with the subpoint of tracing point in the plan air route is benchmark, it is the scope interval at center that reference point is set, each weights coefficient is got between 1, two times of interval each weights and is got 0.5 in one times of interval, and the like, weights are the inverse of interval multiple just in several times of intervals, calculate for simplifying, the coefficient outside 5 times of intervals gets 0.1, so just draws the experience track of this object.

2.2 weather data analysis

A GRIB message comprises the upper-level winds information of 1/8th earth, and each height layer sends a message, according to the GRIB message format historical data is resolved, and obtains temperature, wind direction and the wind speed of certain location point upper-level winds.

According to the experience track of certain object, the weather information of each tracing point of coupling mainly is wind direction and a wind speed on the way in weather data.These data are kept in the community set of this object, unite at theoretical track and experience track and use when forming high precision 4D prediction locus.

3, the foundation of high-precision 4D forecast model

The flight path theoretical model for predicting is based on aerodynamic force mathematic(al) parameter and kinetics equation foundation, and empirical model is then set up through data mining by every data of practical flight.Both respectively have superiority: the basic framework of theoretical model is systemic and complete, and initiative is strong, belongs to positive feedback system, therefore suitable basic solution as the flight path prediction; Empirical model then is succinct flexibly, and debugging is simple, and passivity is strong, belongs to feedback system, and is therefore suitable to verification and addition item in the total solution of flight path prediction.Therefore set up high precision 4D flight path prediction algorithm, mix and use theoretical model and empirical model.Its main body is a theoretical model, and empirical model mainly plays the effect that verification, supplementary parameter and part substitute.

1) verification

Because the flight path that empirical model is set up is directed to the practical flight data,, is well suited for as the comprehensive and approximate of actual model theoretical flight path is evaluated and tested so its precision is very high under the data source reliable conditions.Mainly come two technical indicators of verification herein: time error and space error.

If theoretical flight path is

Wherein n is the prediction track points number of theoretical flight path;

And the experience flight path is Wherein m is the prediction track points data of experience flight path;

Generally get m=n.

And P wherein _i ^p, P _i ^eAll be space-time four-dimentional space point, promptly contain { x, y, z, t} 4-coordinate.

Then the unit interval error is:

Δ t wherein _iBe the timeslice length between two track points, T _TatalBe flight time accumulation; The unit space error is:

Δ s wherein _iBe the spatial pieces length between two track points, S _TatalBe total distance of flying.These two indexs have reflected theoretical flight path and the experience flight path average integral gap on space-time and space respectively.As long as the both less than predetermined threshold, just can think that the reliability of theoretical model has reached acceptable level.

2) supplementary parameter

In the middle of theoretical model, owing to be subjected to the restriction of physical condition, possible some crucial model parameter can't be obtained, and this will directly cause model can't carry out prediction work.Therefore for the whole model of closure, must replenish initial parameter.And this is finished by empirical model usually.Here needing to replenish initialized important parameter comprises:

Takeoff speed:

I=1 wherein, 2,3...n, j=1,2,3 ... m;

M is the section of the taking off track points number of definition, and n is the flight path number.

The control intention: the control intention mainly comprises horizontal control intention and two kinds of vertical control intentions.It has reflected guiding and the flight orders operation that the controller carries out aircraft.

The level intention mainly is the formed actual flight path Control Node of flight paragraph of standoff on cross track.Comprise that flight path deflects away from a little, flight path virtual navigation point sequence, flight path reentry point.

Vertical intention mainly is the rise and fall control command of carrying out on vertical section.Comprise and highly regulate starting point, highly regulate terminating point, escalating rate/rate of descent.

Average escalating rate: the weighted mean by the escalating rate of ascent stage is obtained.

Average rate of decrease: the weighted mean by the rate of descent of descending branch is obtained.

Weather data: wind speed on the way that historical analysis is obtained and the speed in the experience track are asked phasor difference, this difference is added adjustment factor add to speed term in the theoretical model, can regulate this coefficient according to meteorology at that time before taking off.

3) substitution effect

Generally, whole flight course is divided into the indicated airspeed sections of climbing such as quickening the section of climbing, low latitude, the flat section of flying for the Accurate Analysis flight path, the indicated airspeed sections of climbing such as high-altitude, the Mach number sections of climbing such as high-altitude, flat each the air route section that flies in air route, initial descending branch, middle descending branch, paragraphs such as last descending branch.

In fact first above-mentioned proof procedure should all be carried out in superincumbent each stage, if can't fit tightly with empirical model at certain paragraph through the theoretical model of revising, then can directly give theoretical model with relevant empirical parameter.Also promptly substitute theoretical model and finish high-precision forecast with empirical model.

4, radar data is to the correction of accurate 4D prediction locus

After taking off, secondary radar promptly may detect target and transmits definite positional information (longitude, latitude and height), temporal information, velocity information (size and Orientation), and upgrades with radar period.At this, real radar data can be fed back in the model and go, the 4D track is revised.

Consider that commercial airliner substantially all is highly flight of air route, plan according to the rules, the predicted value of position in the 4D track and height and practical flight value difference are apart from little thus, subject matter is exactly the forecasting problem of time dimension, so this paper only provides the modification method about predicted time.

Correction has three kinds of situations opportunity, as long as select wherein a kind of: a) press the radar period correction, promptly receive new data (position, time, highly, speed) just to the prediction locus correction afterwards of this position, at this moment the meticulousst modification method, the correction number of times is maximum; Arrive the correction of plan way point, be near certain reporting point when radar data shows target, the prediction locus correction after this point; The correction of deviation thresholding is set on arrival plan way point correction and the air route to be carried out simultaneously.

Correcting process process each time is basic identical, is that example deals with second kind of situation mainly at this.Target data Data (the x that the up-to-date radar data frame that receives is formed after too much Radar Data Fusion _r, y _r, t _r, h _r, v _r), utilize following formula to judge whether this target is near the plan way point of prediction locus, the way point scope is set for being 10 kilometers of center radius (variable element) with this point:

$\sqrt{{(x_r-{\mathrm{wp}}_i)}^2-{(y_r-{\mathrm{wp}}_i)}^2}\≤10$ Formula (8);

Wherein i when satisfying following formula, illustrates that target is at way point i place for planning any one in the way point.Obtain the mistiming Δ t in way point i predicted time and the target data _r=t _r-wp _i(t), as the reference data of revising.

Predicted time in the 4D prediction locus is revised, two kinds of methods are arranged: the one, directly the mistiming that will obtain above affacts this way point on the prediction way point time afterwards, and just whole time dimension has been made drift correction promptly:

Wp _{M, i+1}(t)=wp _I+1(t)+Δ t _rFormula (9);

I is all way points that target flight will be passed through in the formula.This method is simple and easy to calculate.But when the current flying speed of this point and predetermined speed difference were very big, predicated error was also big.

The 2nd, speed is considered in the correction model, with following formula the time is revised:

${\mathrm{wp}}_{m,i+1}\left(t\right)=({\mathrm{wp}}_{i+1}\left(t\right)-{\mathrm{wp}}_i\left(t\right))\×\frac{{\mathrm{wp}}_i\left(v\right)}{v_r}+t_r$ Formula (10);

I is all way points that target flight will be passed through in the formula, the accuracy height of revising in this way when velocity perturbation is big.

In flight course, constantly prediction locus is revised according to the method described above, desired track accurately is provided, can make things convenient for the controller to understand following air traffic situation.

Core of the present invention is based on the basis of flight planning, aircraft performance parameter, historical flying quality, create the model that a kind of high precision 4D (four dimension) flight track is calculated, utilize real-time radar data that the flight planning flight path is predicted reckoning.

This method is mainly used in Air Traffic Control Automation System and flight flow management system, obtains accurate, continuous, level and smooth flight planning flight path prediction in real time.

The present invention is based on flight planning information, consider performance data, environmental data and the historical flying experience data of flight object, set up high-precision 4D flight model, and introduce real-time radar data on this basis and constantly proofread and correct, so that form desired track accurately, the objective of the invention is to improve the precision of flight planning prediction flight path, predicated error is controlled at bottom line.

The invention provides a kind of thinking and method of the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data; the method and the approach of this technical scheme of specific implementation are a lot; the above only is a preferred implementation of the present invention; should be understood that; for those skilled in the art; under the prerequisite that does not break away from the principle of the invention, can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.The all available prior art of each component part not clear and definite in the present embodiment is realized.

Claims (9)

1. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data is characterized in that, this method may further comprise the steps:

Step (1) is set up the 4D flight path theoretical model based on aeroplane performance;

Step (2) based on the historical flight path data of aircraft model in the step (1), is set up 4D flight path empirical model by mining analysis;

Step (3), the empirical model that theoretical model that step (1) is obtained and step (2) obtain combines, various influence factors in the flight course are quantized to become adjustable parameter, promptly obtain an accurate 4D flight path mixture model of considering the extraneous factor variation based on aeroplane performance again;

Step (4) with the described 4D flight path of radar data correction mixture model, forms final aircraft flight flight path.

2. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 1 is characterized in that, described foundation may further comprise the steps based on the 4D flight path theoretical model of aeroplane performance:

Step (a) is extracted flight planning information, spatial domain information and model-performance parameter: described flight planning information comprises flight number, type, E.T.D(estimated time of departure), the landing airport; Described spatial domain information comprises standard mission program, course line, air route, at advance the to leave the theatre course line and the restriction of various spatial domain on landing airport; Described model-performance parameter comprises the performance data of type correspondence, and promptly aircraft basis geometric parameter, aircraft basic motive mathematic(al) parameter comprise that lift coefficient, resistance coefficient and the climb rate, aircraft configuration information comprise the industry dead weight capacity of aircraft and center of gravity, engine parameter, speed envelope curve, fuel consume;

Step (b) is set up aircraft cross track model:

Utilize flight planning information, spatial domain information in the step (a), advance departure procedure information, for each flight determine its flight the course line, air route, determine its approach course and the course line of leaving the theatre, thereby form the complete horizontal flight flight path of certain flight;

Step (c) is set up aircraft path in elevation model:

Utilize step (a) in model-performance information set up the motion model of the climbing of aircraft, decline and air route flight, and be incorporated into departure procedure and carry out the flight path constraint, form the vertical flight flight path of certain complete flight;

Step (d) with aircraft the horizontal range s of process be basic parameter, in cross track model that step (b) and step (c) are set up and path in elevation model, find corresponding 2 points respectively, promptly in the cross track model, represent aircraft to be the some p1 of s from the beginning horizontal flight distance of taking off, and in the path in elevation model, represent aircraft to be the some p2 of s from the beginning horizontal flight distance of taking off, two spot correlations are joined, obtain latitude and longitude coordinates and the height of p2 and the flight time at that time of p1 respectively, form 4D volume coordinate point, thereby form 4D flight path theoretical model based on aeroplane performance.

3. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 2 is characterized in that, the described horizontal model of aircraft track of setting up may further comprise the steps:

Step (a) definition cross track, from the plan information of landing airport, Standard Instrument Approach and/or leave the theatre course line and air route information, cruising altitude extract aircraft by way of way point and leap height, set up three-dimensional flight track framework, cross track promptly through the projection line of way point at surface level;

The air route and the course line of advancing to leave the theatre are made of a series of air routes landmark point, are WP with the sequence definition of putting _i, number is n, determines radius of turn r according to flying speed _i, definition constrained type type _i: initial point constraint, general point constraint and terminal constraint, simplified model is:

G _f＝{wp ₁(x ₁，y ₁，type ₁，r ₁)，…，wp _i(x _i，y _i，type _i，r _i)，…，wp _n(x _n，y _n，type _n，r _n)}；

Step (b) is handled turning point, if the initial point constraint is then flown nonstop to down a bit; Turn Models is considered in the general point constraint; The necessary line up with runway of head when end is constrained to aircraft through this; According to the above constraint, structure comprises radius of turn r, the Turn Models of turning point P, Q coordinate:

Radius of turn $r=\frac{v^2}{g\×\tan \mathrm{\φ}},$ G is an acceleration of gravity, and flying speed is v during turning, and the pitch angle is φ; The course angle of aircraft flight is with reference to zero degree with direct north, increases along clockwise direction, is 359 degree to the maximum; Use α ₁₂The course angle of representing last leg, α ₂₃The course angle of one leg, expression back according to 3 line relation, judges that aircraft is to turn left, turn right or fly nonstop to; In conjunction with the course angle of two sections flights with turn to, obtain aircraft turning angle Δ α; According to turning angle and radius of turn, obtain the distance of P point-to-point i: D _{P, i}=r * tan (Δ α/2), the coordinate that draws turning point P is:

$\left\{\begin{array}{c}{x}_p=x_i-D_{P,i}\×{\cos \mathrm{\α}}_{12}\\ y_P=y_i-D_{P,i}\×{\sin \mathrm{\α}}_{12}\end{array}\right.$

The Q point coordinate is:

$\left\{\begin{array}{c}{x}_Q=x_i+D_{P,i}\×{\cos \mathrm{\α}}_{23}\\ y_Q=y_i+D_{P,i}\×{\sin \mathrm{\α}}_{23}\end{array}\right.$

Final cross track mark is made up of a series of orderly points:

G _f＝{wp ₁(x ₁，y ₁)，…，P _i(x _i，y _i)，Q _i(x _i，y _i)，…，wp _n(x _n，y _n)}。

4. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 2 is characterized in that, the described aircraft track vertical mould of setting up may further comprise the steps:

Step (a) is divided the paragraph of flight path in elevation model, according to five stages of flight: take off, climb, flatly fly, descend, land, whole path in elevation computation model is divided into the calculating that five submodels are handled above-mentioned five paragraphs respectively;

Step (b) is determined flight path in elevation model basic parameter, described vertical mould be aircraft in height-time or height-range coordinate system motion model, definite effect external force aboard: motor power F, direction becomes engine installation angle along engine axis with fuselage axis X-shaped

Lift L is perpendicular to flying speed v; Resistance D is parallel to flying speed; Gravity G is downward; α is the angle of attack, and θ is a flight-path angle; The method that adopts Newton second law and conservation of energy principle to combine is listed the submodel equation of each inflight phase correspondence, use imfinitesimal method and divided the segment of some five equilibriums each stage, the substitution aircraft is in the performance parameter of each stage correspondence, find the solution the rising/rate of descent of every segment, corresponding distance and time, adding up at last forms altitude profile and velocity profile, and obtains flight time accumulation;

Step (c) is determined the model of the sliding race process of takeoff phase:

The sliding race process of takeoff phase, sliding speed is parallel to surface level, the friction force f of tire and runway face;

Sliding the kinetics equation of race stage foundation along runway heading:

Wherein μ is the runway friction factor, and m is an aircraft weight, and g is an acceleration of gravity, Be percentage speed variation,

Be the runway gradient, gravity component in the horizontal direction:

In the starting heats stage, aircraft angle of attack is assumed to the shutdown angle of attack, draws acceleration a, with given flying speed VLOF substitution formula: $S_G={\∫}_0^{V_{\mathrm{LPF}}}\frac{\mathrm{vdv}}{a},$ $t_G={\∫}_0^{V_{\mathrm{LOF}}}\frac{\mathrm{dv}}{a},$ Adopt numerical integrating to slide and sue for peace after the race section is divided into the independent integration of segment, draw ground roll-out distance and sliding race time;

Step (d) is determined aerial each inflight phase model:

In all stages after liftoff before land, use equation:

$(F-D)v_{\mathrm{TAS}}=\mathrm{mg}\frac{\mathrm{dh}}{\mathrm{dt}}+{\mathrm{mv}}_{\mathrm{TAS}}\frac{{\mathrm{dv}}_{\mathrm{TAS}}}{\mathrm{dt}};$

Wherein h is an aircraft place height, v _TASBe the flight true air speed; In the top formula

Described C _DAnd C _LBe given resistance coefficient of aeroplane performance parameter and lift coefficient, ρ is an atmospheric density, and S is a wing area;

According to the thrust magnitude of ramp-up period correspondence, obtain escalating rate, obtain the used time behind certain height that rises with numerical integration; Obtain percentage speed variation, binding time and percentage speed variation are obtained the horizontal range of every segment; Add the fuel consume coefficient, obtain the amount of fuel that per stage consumes, every segment weight is constant in approximate thinking, following a bit of weight is that the preceding paragraph weight deducts amount of fuel; Result's summation of every segment is obtained time, distance, the speed of ultimate demand;

Cruising phase adopts equal altitudes and waits the Mach number mode to carry out; Stipulated most economical cruise speed or Mach number in the described aeroplane performance parameter, used kinematical equation and obtain cruise time, distance, amount of fuel; Aircraft drops to specified altitude assignment with the maximum gliding angle of regulation, obtains voyage and time;

The performance parameter of substitution decline stage correspondence is obtained rate of descent, obtains the used time by integration; Obtain percentage speed variation, binding time is obtained distance, has formed the vertical mould of whole airflight.

5. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 1, it is characterized in that, described step (2) during to the comprehensive flight path output information of control automated system and implementing plan corresponding weather information carry out the admission of certain hour, these data are put in order as data source;

Specifically may further comprise the steps:

Step (a) generates the experience flight path:

The comprehensive output information of every frame of district's guard system output is a collection of aerial target status information and following desired track information at that time; By playing the historical comprehensive output information of a period of time, Frame is contrasted one by one, flight path by flight number, landing airport, actual time of departure, type and flight path information number all is encapsulated in this flight object set as the sign of unique object position sequence, height sequence, the time corresponding sequence with whole flight course, promptly forms the historical track of flight; To same flight object, in the repeatedly flight of a period of time, the object set that encapsulates has a plurality of, and a plurality of historical tracks are promptly arranged:

T(obj，i)＝{x _j，y _j，h _j，t _j，v _j}(i＝(1，2，…，n)，j＝(1，2，…，m))；

Wherein obj is the flight object, and n is the flight total degree, and equation the right is position, time, the velocity information of each tracing point, and m is the tracing point number of each flight path; To a plurality of track set of each flight object,, draw the experience track of this object with every data difference weighted; Processing procedure comprises parameter alignment and weighted;

Described parameter is aligned to a plurality of track set to same target, comprise and calculate the distance between first tracing point and original base in each set, starting point in the big set of selected distance is as the position reference point, point in other set before this position is rejected, and forms approximate unified origin information; The time of all starting points is located zero point, the time sequencing skew of back tracing point;

Described weighted is to single object, comprise with the position in a plurality of historical tracks set after the parameter alignment, time, speed respectively weighted draw the experience track of this object;

$T_{\mathrm{obj}}^e=\{{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(w_i\×x_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(u_i\×y_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(r_i\×h_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(s_i\×t_i)}{n}\right)}_j,{\left(\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(z_i\×v_i)}{n}\right)}_j\};$ W wherein _i, u _i, r _i, s _i, z _iBe the weights coefficient, j is a tracing point, adds up to the effective number after parameter is alignd; Determine good plan air route according to meter information, with the subpoint of tracing point in the plan air route is benchmark, it is the scope interval at center that reference point is set, each weights coefficient gets 1 in one times of interval, each weights gets 0.5 between two times of intervals, weights are the inverse of interval multiple just in several times of intervals, and the coefficient outside 5 times of intervals gets 0.1, draws the experience track of this object;

Step (b) is handled the upper-level winds meteorology:

Described weather data analysis comprises: a GRIB message comprises the upper-level winds information of 1/8th earth, each height layer sends a message, according to the GRIB message format historical data is resolved, obtain temperature, wind direction and the wind speed of certain location point upper-level winds; According to the experience track of certain object, the weather information of each tracing point of coupling comprises on the way wind direction and wind speed in weather data; These data are kept in the community set of this object.

6. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 1 is characterized in that, accurate 4D flight path mixture model carries out verification and supplementary parameter to the described theoretical model of step (1) in the described step (3).

7. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 6 is characterized in that, described verification comprises time error and space error verification;

If theoretical flight path is $T^p=\left\{P_i^p\right\}(i=\mathrm{1,2,3}...n),$ Wherein n is the prediction track points number of theoretical flight path;

And the experience flight path is $T^e=\left\{P_i^e\right\}(i=\mathrm{1,2,3}...m),$ Wherein m is the prediction track points data of experience flight path;

Get m=n;

P wherein _i ^p, P _i ^eAll be space-time four-dimentional space point, promptly contain { x, y, z, t} 4-coordinate;

Then the unit interval error is:

Δ t wherein _iBe the timeslice length between two track points, T _TatalBe flight time accumulation; The unit space error is:

Δ s wherein _iBe the spatial pieces length between two track points, S _TatalBe total distance of flying.

8. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 6 is characterized in that described supplementary parameter comprises: takeoff speed $V=\frac{\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}{V}_{i,j}}{m\·n},$ I=1 wherein, 2,3...n, j=1,2,3 ... m; M is the section of the taking off track points number of definition, and n is the flight path number; Control intention, control intention comprise horizontal control intention and vertical control intention; Level is intended that the formed actual flight path Control Node of flight paragraph of standoff on cross track, comprises that flight path deflects away from a little, flight path virtual navigation point sequence, flight path reentry point; Vertically be intended that the rise and fall control command of on vertical section, carrying out, comprise and highly regulate starting point, highly regulate terminating point, escalating rate or rate of descent; Average escalating rate is obtained by the weighted mean of the escalating rate of ascent stage; Average rate of decrease: the weighted mean by the rate of descent of descending branch is obtained; Weather data, wind speed on the way that historical analysis is obtained and the speed in the experience track are asked phasor difference, this difference is added adjustment factor add to speed term in the theoretical model, and basis meteorology is at that time regulated this coefficient taking off before.

9. the method for analyzing high precision 4 D flight trajectory of airplane based on real-time radar data according to claim 1, it is characterized in that, described step (4) comprises that secondary radar promptly may detect Aircraft Target and transmits definite positional information, temporal information, velocity information, and upgrade, and the time is revised with radar period; The described time is modified to by on radar period correction or arrival plan way point correction or arrival plan way point correction and the air route correction of deviation thresholding being set carries out simultaneously.

Images