CN102426025B - Simulation analysis method for drift correction angle during remote sensing satellite attitude maneuver - Google Patents

Abstract

The invention relates to a simulation analysis method for a drift correction angle during remote sensing satellite attitude maneuver. According to the present invention, a simulation tool is adopted to establish a satellite and a satellite sensor, and the longitude and the latitude of the intersection point of a sensor visual axis and an earth model are acquired in the given time period; according to the longitude and the latitude, feature points are established on the earth model; the coordinates of each feature point at the current time point and the next time point in a J2000 coordinate system are acquired; in the J2000 coordinate system, the following velocity vector of the current feature point at the current time in the J2000 coordinate system, and the movement velocity vector of the current feature point relative to the satellite in the J2000 coordinate system are respectively acquired; the vector composition of the velocity is adopted to acquire the total velocity vector of the movement of the current feature point relative to the satellite in the J2000 coordinate system; a CCD imaging plane is established in the satellite model; the projection of the total velocity vector in the CCD imaging plane of the satellite is acquired by projecting; the included angle formed by the projection vector and the normal line direction of the CCD linear array is the drift correction angle.

Description

The simulating analysis of drift correction angle during remote sensing satellite attitude maneuver

Technical field

The present invention relates to a kind of method that adopts simulation analysis to obtain drift correction angle degree on the motor-driven clock star of the attitude of satellite.

Background technology

The definition of drift angle stems from aeronautical terminology, and its definition is while there is crosswind, and the actual flight path of aircraft and the course of aircraft are inconsistent, and the angle of course line and track line, is called drift angle.At satellite when the earth turns round, because can making the object of earth surface, earth rotation there is a convected velocity along earth autobiography angular velocity direction, the moving direction (track line) of the relative subject of camera is now inconsistent with the projection line velocity reversal (course line) of camera motion, also referred to as drift angle.

The definition of drift correction angle comes from the satellite with attitude maneuver ability, when satellite carries out pitching and side-sway automotive, for guarantee satellite at the target directing of process Satellite of adjusting attitude crab the wind angle (along satellite Z _bdirection of principal axis) constant, only by adjusting around satellite Z _baxle rotates to carry out drift angle correction, so this is around Z _bjust there is certain deviation with the drift angle of the resulting photography point of actual computation in the angle that axle rotates, defines this around Z _bthe angle that axle rotates is that drift correction angle is to show difference.The Z here _baxle belongs to satellite body coordinate system (O _bx _by _bz _b), the initial point O of satellite body coordinate system _bat frame satellite-rocket docking face center, butt joint ring lower end, the axial line of butt joint ring is satellite X _baxle, points to load cabin direction for just by initial point, satellite Z _baxle is for being parallel to breviseptum direction, with installation of sensors side over the ground for just, Y _baxle is parallel to long dividing plate direction, perpendicular to X _baxle and Z _baxle and follow the right-hand rule and X _baxle, Z _baxle form right angle coordinate system.

For space, carry out the sensor of imaging, especially after the technology such as TDICCD are applied on satellite, in satellite task, just there is strict demand with the relative motion of satellite on a surface target, the impact that satellite be can not ignore generation to obtaining of information around the less anglec of rotation of axis.Particularly for TDICCD, its imageing sensor is compared with general line array CCD, profile is a line array CCD device, but its structure is as a face battle array device, the similar multiexposure, multiple exposure of its image-forming principle, therefore require same each pixel listing to same target exposure integration, the picture quality of guarantee output.If do not carry out drift angle control, will produce laterally picture in image planes and move, the pixel that makes same row is in the process of integration repeatedly, and what see is not same target.This picture moves and will cause the reduction of image quality, therefore need on star, to drift angle, revise.

Raising along with TDICCD remote sensing satellite imaging resolution and image quality requirement, on satellite, carry out accurate drift angle correction and become gradually a kind of inevitable task, and drift angle analytical algorithm on star is verified and also just become the overall work of TDICCD remote sensing satellite.On ground, in the face of satellite drift correction angle, carry out high-precision calculating not only can design of satellites and factory testing stage to the star of satellite on the real-time computational accuracy of drift angle verify, on its in-orbit imaging performance impact estimate, can also, for verifying the source of star epigraph influence factor after satellite is entered the orbit, for better, ensure that image imaging quality provides the most directly support.

In the analytical approach of traditional drift angle, all use geometric analysis and the derivation of equation to calculate, the shortcoming of this mode maximum is exactly to depend on too much hypothesis, therefore directly cause computational accuracy not enough, can only, as analyzing qualitatively, cannot verify by the result of this low precision correctness and the precision of algorithm on star.And along with the use of satellite pitch attitude maneuverability, adopt the feature that derivation is complicated, precision is on the low side of the method for geometric analysis just more to display.The correction precision of drift angle and image quality have closely and contact, thereby the mode of traditional qualitative analysis formed not controlled leak in the imaging link design proof procedure of satellite, be unfavorable for from the quantitative assurance satellite of the angle of system image quality in-orbit.

Summary of the invention

Technology of the present invention is dealt with problems and is: overcome the deficiencies in the prior art, provide a kind of based on high precision model, use numerical computation method, the simulating analysis of drift correction angle during to the adaptable attitude maneuver of the pitching side-sway attitude maneuver of TDICCD remote sensing satellite.

Technical solution of the present invention is: the simulating analysis of drift correction angle during remote sensing satellite attitude maneuver, and step is as follows:

(1) use emulation tool to set up satellite and star upper sensor, the attitude of satellite and sensor parameters are set;

(2) within the given time cycle, obtain the longitude and latitude of the sensor optical axis and earth model intersection point;

(3) longitude and latitude obtaining according to step (2) is set up successively unique point on earth model;

(4) obtain the coordinate of each unique point under the J2000 of current point in time and next time point coordinate system;

(5) under J2000 coordinate system, calculate current unique point the coordinate of next time point with respect to current unique point the differential at current time point coordinate, obtain the convected velocity vector of current unique point in the J2000 of current time coordinate system;

(6) under J2000 coordinate system, calculate current unique point the coordinate of current point in time with respect to next unique point the differential at next time point coordinate, obtain current unique point movement velocity vector with respect to satellite in J2000 coordinate system;

(7), according to the result of step (5) and step (6), utilize the vector of speed to synthesize the general speed vector that the current unique point of acquisition is moved in J2000 coordinate system with respect to satellite;

(8) in dummy satellite, set up CCD imaging plane, by projection, obtain general speed vector that current unique point moves in J2000 coordinate system with respect to the satellite projection vector in Satellite CCD imaging plane;

(9) ask for the angle of described projection vector and CCD linear array normal direction, be the drift correction angle of current unique point;

(10) repeating step (5)～(9), obtain the drift correction angle of all unique points.

The present invention's advantage is compared with prior art: during attitude maneuver that the present invention proposes, the simulating analysis of drift correction angle has been broken away from the dependence to hypothesis too much in conventional geometric analytical approach, in method, merged High Precision Simulation, space vector computing and TDICCD remote sensor imaging geometry characteristic, overcome the deficiency of existing methods analyst precision, obtained the motion vector generated data of actual photography point, by vector projection in star upper sensor imaging plane, and further combined with the imaging direction vector of actual TDICCD device, obtained the drift correction angle on photography moment star.Because drift correction angle is parameter very crucial on star, its parameter will be directly connected to the ground integrated index of star such as image quality, image position accuracy, camera imaging spatial direction precision of TDICCD remote sensing satellite.Therefore the inventive method efficiently solves the high precision validation problem to drift angle correction algorithm on star, for the ground test of satellite imagery link is verified the important means of having supplemented.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the inventive method;

Fig. 2 is emulation and the optical axis schematic diagram of satellite motion imaging of the present invention;

Fig. 3 is that the present invention is synthesized and obtained general speed by vector

schematic diagram;

Fig. 4 is Satellite CCD imaging plane of the present invention and general speed model schematic diagram;

Fig. 5 is general speed projection of the present invention and CCD normal direction dummy satellite schematic diagram;

Fig. 6 is the Calculation results schematic diagram of the present invention to different side-swaies angle;

Fig. 7 is simulation analysis of the present invention and geometric analysis result contrast schematic diagram.

Embodiment

The definition of given first J2000 inertial coordinates system describes the concept of the optical axis and photography point simultaneously.

J2000 inertial coordinates system (O _jx _jy _jz _j) be an inertial space coordinate system, this coordinate system be take the earth's core as initial point O _j, X _jthe direction in average first point of Aries of the earth of measuring when axle forward points to UTC Universal Time Coordinated 12:00 on the 1st January in 2000, Z _jaxle forward points to the average axis of rotation the North that the earth is measured when UTC Universal Time Coordinated 12:00 on the 1st January in 2000, Y _jaxle and X _j, Z _jaxle is vertical, X _j, Y _j, Z _jthree axles form right-handed coordinate system.

The definition of the optical axis is exactly the axis of sensor field of view, owing to being difficult to that in imaging simulation the whole imageable target in whole sensor field of view are analyzed, so generally all adopt relatively little in visual field in the situation that the analysis of visual axis imaging to characterize the imaging performance of whole sensor.When analyzing, the point that the optical axis points to earth surface is just called as photography point.

For one, typically take the remote sensing satellite that TDICCD camera is load, the simulating analysis of its drift correction angle as shown in Figure 1, is mainly divided into and sets up realistic model, obtains optical axis intersection point, obtains after photography point motion vector, projection and calculate and obtain drift correction angle Four processes.Below just using N time point as an emulation cycle, Δ t, as the interval between each time point, describes embodiment with this.

1. set up realistic model

In this step, used STK as emulation tool.Open STK software, newly-built satellite, input orbit parameter, comprises time epoch, semi-major axis, excentricity, orbit inclination, the angle of depression, perigee, ascending node precision, mean anomaly.Select the J4 Disturbance Model of the earth as the deduction model of satellite orbit, and select EOPv1.1 as the sensing parameter of earth model.

Newly-built sensor on satellite, sets the visual field of sensor according to the field angle of actual TDICCD.Open the attribute of satellite, in attitude, the attitude maneuver parameter of satellite is set, if side-sway automotive only, select " ECI velocity alignment with nadir constraint ", input side swinging in " Constraint ", if have side-sway and pitching motor-driven simultaneously, select " Aligned and Constrained ", in " Aligned Vector " and " Constrained Vector ", according to the sequencing of side-sway and pitching, distinguish input side swinging and luffing angle afterwards, complete the setting of attitude.

2. obtain optical axis intersection point

In simulation software, carry out the emulation of satellite motion imaging, as shown in Figure 2, obtain the intersection point on the star upper sensor optical axis (visual field central authorities vector in Fig. 2) and earth model ground, within the emulation cycle of N time point, obtain the longitude and latitude data of satellite and earth model ground intersection point.

In this note, represent that the data of intersection point are: (Lat ₁, lon ₁), (Lat ₂, lon ₂) ...., (Lat _n, lon _n), wherein subscript has indicated the moment that obtains this intersection point longitude and latitude, and the longitude and latitude of the intersection point of i time point generation is (Lat _i, lon _i).

3. obtain photography point motion vector

Owing to being all object-based for most of emulation tools, and not an object for the optical axis of satellite and the intersection point on ground, so be difficult to obtain the detailed data of its position of intersecting point.Therefore in order to obtain the motion vector of photography point, just must be first photography point (characterizing with optical axis intersection point) objectification, the longitude and latitude data of the intersection point obtaining in utilizing are 2. set up unique point at correspondence position, and obtain needed coordinate data information, and concrete operations are as follows:

To i=1～N, at coordinate, be (Lat successively _i, lon _i) landscape position set up unique point, utilize emulation tool STK to obtain the coordinate data of this unique point under i time point and the J2000 coordinate system in two moment of i+1 time point, be expressed as:

$\left\{\begin{array}{c}({\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="HSA00000559954200031.png"\; state="\{\{state\}\}">X</figure-callout>}}_{1,t=1},{\mathrm{<figure-callout\; id="1"\; label="Y"\; filenames="HSA00000559954200031.png"\; state="\{\{state\}\}">Y</figure-callout>}}_{1,t=1},{\mathrm{<figure-callout\; id="1"\; label="Z"\; filenames="HSA00000559954200031.png"\; state="\{\{state\}\}">Z</figure-callout>}}_{1,t=1}),({\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="HSA00000559954200031.png"\; state="\{\{state\}\}">X</figure-callout>}}_{1,t=2},{\mathrm{<figure-callout\; id="1"\; label="Y"\; filenames="HSA00000559954200031.png"\; state="\{\{state\}\}">Y</figure-callout>}}_{1,t=2},{\mathrm{<figure-callout\; id="1"\; label="Z"\; filenames="HSA00000559954200031.png"\; state="\{\{state\}\}">Z</figure-callout>}}_{1,t=2})\\ (X_{2,t=2},Y_{2,t=2},Z_{2,t=2}),(X_{2,t=3},Y_{2,t=3},Z_{2,t=3})\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ (X_{i,t=i},Y_{i,t=i},Z_{i,t=i}),(X_{i,t=i+1},Y_{i,t=i+1},Z_{i,t=i+1})\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\end{array}\right.$

Use the position coordinates obtaining, first adopt the mode of position differential to obtain convected velocity V _evector Message, circular is as follows:

Coordinate information (X to satellite in i unique point of i time point and i+1 time point acquisition _{i, t=i}, Y _{i, t=i}, Z _{i, t=i}) and (X _{i, t=i+1}, Y _{i, t=i+1}, Z _{i, t=i+1}) time diffusion is obtained to the convected velocity V that unique point is caused by earth rotation _e.Here use location, to Δ t differential (unique point at the earth's surface self convected motion is exactly the motion with respect to satellite, so convected velocity is pointed to the position of t=i+1 from the position of t=i), therefore has:

$\stackrel{\&RightArrow;}{V_{e,t=i}}=\frac{\mathrm{\&Delta;}(X,Y,Z)}{\mathrm{\&Delta;t}}=\frac{{(X_i,Y_i,Z_i)}_{t=i+1}-{(X_i,Y_i,Z_i)}_{t=i}}{\mathrm{\&Delta;t}}=\frac{(X_{i,t=i+1}-X_{i,t=i},Y_{i,t=i+1}-Y_{i,t=i},Z_{i,t=i+1}-Z_{i,t=i})}{\mathrm{\&Delta;t}}$

Secondly the same mode of position differential that adopts is obtained the movement velocity V with respect to satellite _ivector Message:

Coordinate information (the X of i the unique point that satellite is obtained at i time point _{i, t=i}, Y _{i, t=i}, Z _{i, t=i}) and the coordinate information (X of i+1 unique point obtaining of i+1 time point _{i+1, t=i+1}, Y _{i+1, t=i+1}, Z _{i+1, t=i+1}) time diffusion is obtained to the speed of related movement V that unique point is caused by satellite motion _ihere use location difference to Δ t differential (not in the same time unique point be not both kinetic by satellite, therefore certain unique point is contrary with the direction of motion of satellite itself with respect to the movement velocity of satellite, so relative velocity points to the position of i the unique point of t=i from i+1 the characteristic point position of t=i+1), therefore have:

$\stackrel{\&RightArrow;}{V_{i,t=i}}==\frac{{(X_i,Y_i,Z_i)}_{t=i}-{(X_{i+1},Y_{i+1},Z_{i+1})}_{t=i+1}}{\mathrm{\&Delta;t}}=\frac{(X_{i,t=i}-X_{i+1,t=i+1},Y_{i,t=i}-Y_{i+1,t=i+1},Z_{i,t=i}-Z_{i+1,t=i+1})}{\mathrm{\&Delta;t}}$

Finally by the synthetic mode of vector, obtain unique point with respect to the general speed of satellite flight, as shown in Figure 3, $\stackrel{\&RightArrow;}{V_{a,t=i}}=\stackrel{\&RightArrow;}{V_{e,t=i}}+\stackrel{\&RightArrow;}{V_{i,t=i}}.$

Obtained thus under J2000 inertial coordinates system, synthetic general speed vector at the photography point of i time point with respect to satellite, successively to i=1,2, ... N altogether N time point carries out above-mentioned analysis, obtain the satellite aggregate velocity vector of all N time point, and then can carry out thus the calculating of drift correction angle.

4. after projection, calculate and obtain drift correction angle

In order to carry out projection, first need to determine the imaging plane of CCD, in " Vector Geometry Tool (how much instruments of vector) " of STK, set up CCD imaging plane, the concrete steps of setting up are model 3-D walls and floors, according to the design of satellite by CCD imaging plane the X based on satellite body coordinate system _bo _by _bplane defines in the mode of Eulerian angle rotation, secondly the coordinate axis defining is defined as to a coordinate system, the initial point of system is used the initial point of satellite body coordinate system, and finally the mode of use " Quadrant " selects to define the XOY plane of coordinate system above, sets up CCD imaging plane.By obtained above

centered by the initial point of satellite body coordinate system, show as shown in Figure 4 with CCD imaging plane simultaneously, as can be seen from the figure,

not on the imaging plane of CCD.

Directly right in model

projection on the imaging plane of CCD, obtains projection vector

and try to achieve this projection vector

normal direction vector with CCD device

angle, as shown in Figure 5, the value at this angle is drift correction angle.

By what obtain in 3. in step

(i=1,2 ... the projection steps that N) the satellite aggregate velocity vector of N time point is brought respectively face into, can obtain all drift correction angle data of N time point altogether.Due to actual bias current angle less (less than 5 degree), in order to make the characteristics such as the projection of vector and angle more clear in legend, in Fig. 4 and Fig. 5, vector angle is amplified.

Embodiment

For the TDICCD remote sensing satellite of the sun synchronous orbit of 650 km height, orbit parameter is: semi-major axis=7023.14 km; Excentricity=0; Orbit inclination=97.9708 degree; The angle of depression=0, perigee degree; Longitude of ascending node=337.752 degree; Mean anomaly=0 degree.According to the design of general satellite, the X of satellite body coordinate system is set _bo _by _bplane is exactly CCD imaging plane.In the situation that only considering satellite side-sway attitude maneuver, get N=5000s, Δ t=1s, each step through the inventive method, (is provided with 0 ° here by different side-sway angles is set, 15 °, 25 °, 35 °), can obtain high-precision drift correction angle data by simulation analysis, concrete result as shown in Figure 6, has provided the relation of drift correction angle and satellite latitude in figure.In the motor-driven angle of the attitude of satellite hour, the precision of geometric analysis method is just slightly high, and the inventive method is not just too obvious on precision improvement, and in the situation that not carrying out any attitude maneuver, average behavior only promotes 2%.And side-sway angle is larger, the computational accuracy of method of geometry is just lower, in side-sway 35 degree, the data of the data of geometric analysis and simulation analysis is contrasted as shown in Figure 7.As can be seen from the figure, pure geometric analysis is because hypothesis is more, also there is larger deviation, geometric analysis data when to 35 degree side-sway contrast statistics with the simulation analysis data of using the method to obtain, use this emulation mode can on average improve 0.67 ° of the calculation deviation of drift angle, improve precision approximately 30%.

The content not being described in detail in instructions of the present invention belongs to those skilled in the art's known technology.

Claims (1)

1. the simulating analysis of drift correction angle during remote sensing satellite attitude maneuver, is characterized in that step is as follows:

(1) use emulation tool to set up satellite and star upper sensor, the attitude of satellite and sensor parameters are set; Described emulation tool is STK;

(2) within the given time cycle, obtain the longitude and latitude of the sensor optical axis and earth model intersection point; In this note, represent that the data of intersection point are: (Lat ₁, lon ₁), (Lat ₂, lon ₂) ...., (Lat _n, lon _n), wherein subscript has indicated the moment that obtains this intersection point longitude and latitude, and the longitude and latitude of the intersection point of i time point generation is (Lat _i, lon _i);

(3) longitude and latitude obtaining according to step (2) is set up successively unique point on earth model;

(4) obtain the coordinate of each unique point under the J2000 of current point in time and next time point coordinate system, be expressed as

(5) under J2000 coordinate system, calculate current unique point the coordinate of next time point with respect to current unique point the differential at current time point coordinate, obtain the convected velocity vector of current unique point in the J2000 of current time coordinate system; Circular is as follows:

Coordinate information (X to satellite in i unique point of i time point and i+1 time point acquisition _{i, t=i}, Y _{i, t=i}, Z _{i, t=i}) and (X _{i, t=i+1}, Y _{i, t=i+1}, Z _{i, t=i+1}) time diffusion is obtained to the convected velocity V that unique point is caused by earth rotation _e, have

(6) under J2000 coordinate system, calculate current unique point the coordinate of current point in time with respect to next unique point the differential at next time point coordinate, obtain current unique point movement velocity vector with respect to satellite in J2000 coordinate system; Circular is as follows:

Coordinate information (the X of i the unique point that satellite is obtained at i time point _{i, t=i}, Y _{i, t=i}, Z _{i, t=i}) and the coordinate information (X of i+1 unique point obtaining of i+1 time point _{i+1, t=i+1}, Y _{i+1, t=i+1}, Z _{i+1, t=i+1}) time diffusion is obtained to the speed of related movement V that unique point is caused by satellite motion _i, have

(7), according to the result of step (5) and step (6), utilize the vector of speed to synthesize the general speed vector that the current unique point of acquisition is moved in J2000 coordinate system with respect to satellite

successively to i=1,2 ... N altogether N time point analyzes, and obtains the general speed vector of all N time point;

(8) in dummy satellite, set up CCD imaging plane, by projection, obtain general speed vector that current unique point moves in J2000 coordinate system with respect to the satellite projection vector in Satellite CCD imaging plane;

(9) ask for the angle of described projection vector and CCD linear array normal direction, be the drift correction angle of current unique point;

(10) repeating step (5)～(9), obtain the drift correction angle of all unique points.

Images