CN111625010A - Spacecraft three-super near-zero error tracking control method based on combined filtering - Google Patents
Spacecraft three-super near-zero error tracking control method based on combined filtering Download PDFInfo
- Publication number
- CN111625010A CN111625010A CN202010350592.5A CN202010350592A CN111625010A CN 111625010 A CN111625010 A CN 111625010A CN 202010350592 A CN202010350592 A CN 202010350592A CN 111625010 A CN111625010 A CN 111625010A
- Authority
- CN
- China
- Prior art keywords
- load
- spacecraft
- star
- target
- attitude
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
- G05D1/0808—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
- G05D1/0816—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability
- G05D1/0833—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability using limited authority control
Abstract
A spacecraft three-ultra-near-zero error tracking control method based on combined filtering is suitable for the field of target tracking and having the requirement of determining ultrahigh precision of loads. Different from the single-stage attitude control of the traditional spacecraft star platform, the invention provides a star-load-fast reflecting mirror three-stage attitude cooperative control method based on combined filtering aiming at the spacecraft platform with three-stage and super control performances of ultra-high precision pointing, ultra-high stability control, hypersensitive control and the like, the attitude calculation of a target is improved by utilizing deep learning, the attitude control precision is improved step by step from a star, load and fast reflecting mirror three-stage system, and the high-precision attitude control is provided for optical load fast tracking and high-quality imaging. The method mainly comprises the following steps: establishing a three-level cooperative control system dynamic model; calculating the pose of the characteristic position of the target spacecraft based on deep learning; designing a multi-level system fusion filter; and designing a three-level cooperative control system controller, including bandwidth design.
Description
Technical Field
The invention belongs to the field of spacecraft attitude control, and relates to a spacecraft three-ultra-near-zero error tracking control method based on combined filtering.
Background
The spacecraft relative attitude tracking is a core problem in the control field, has important significance for the smooth implementation of many tasks such as space observation and the like, the current spacecraft provides the requirements of three-super (ultrahigh precision, ultrahigh stability and ultra-agility) for relative attitude tracking control, and the tracking precision and stability of a target are important problems to be researched for relative attitude tracking. The three-super platform is just based on the requirement of aiming at the optical load attitude with high precision control. At present, components containing high-speed rotors, such as flywheels, control moment gyros and the like, are generally adopted by spacecrafts as actuating mechanisms of attitude control systems. These high speed rotating components inevitably produce high frequency jitter and micro-vibrations that directly affect the performance of the load. Therefore, the requirement is provided by three-level cooperative control of the spacecraft. Meanwhile, when the satellite platform and the target spacecraft are in rapid intersection, the relative motion speed is high, the relative attitude change is also high, and how to rapidly and accurately measure the position attitude of the target is also a problem needing urgent research as the input of control.
The prior control system has the following defects:
1. cannot realize isolation and suppression of high-frequency micro vibration of star
In the attitude control system of the existing spacecraft, a load is rigidly connected with a star body. The flexible vibration and high-frequency micro-vibration existing in the spacecraft star body are directly transmitted to the load, so that the optical load cannot further improve the imaging quality. The traditional spacecraft attitude system is limited by the bandwidth of a controller and the precision of an actuating mechanism, and cannot realize active control on flexible vibration and high-frequency micro-vibration, so that the control precision and stability of the star body are further improved and limited.
2. The contradiction exists between the high precision and the high dynamic performance of the attitude determination, and the realization of the high precision and the high dynamic performance cannot be realized simultaneously
The traditional spacecraft control system only has a star body primary control loop and does not comprise an active pointing hyperstatic platform and a fast reflecting mirror mathematical model. Limited by the measurement principle of the optical sensor, the attitude sensor with higher precision has poorer dynamic characteristics, the attitude sensor with better dynamic characteristics has poorer measurement precision, and the high-precision tracking of a high-dynamic moving target requires a spacecraft control system to have high-dynamic and high-precision sight measurement capability at the same time. Therefore, it is necessary to develop a three-level cooperative target tracking control method, which makes the control system have high dynamic and high precision attitude determination capability by using the advantages of different sensors, and realizes three-super performance tracking control of the load.
3. It is difficult to achieve rapidity for the purpose.
At present, the source of the target characteristic data is mainly ground-based detection, including optical and radar equipment. However, the detection data of the ground-based equipment is limited by too many factors, so that the detection result has great uncertainty. In the aspect of target detection and identification at home and abroad, point target detection under a long distance condition and identification of an on-orbit motion state are mostly concentrated, and research on identification of important characteristic components such as a satellite body and a solar sailboard is less. In recent years, the development of deep learning technology greatly improves the effect of image target identification, which brings a new method means to the field of space target identification.
Disclosure of Invention
The technical problem solved by the invention is as follows: the method overcomes the defects of the prior art, provides a combined filtering-based spacecraft three-ultra-near-zero error tracking control method, can realize high-precision pointing control and high-stability control of the optical load of a spacecraft, and provides a technical basis for ultrahigh-precision pointing, ultrahigh-stability control and high-quality imaging of the optical load of the spacecraft in the future.
The technical solution of the invention is as follows:
a spacecraft three-ultra-near-zero error tracking control method based on combined filtering comprises the following steps:
(1) establishing a dynamic model of a spacecraft multilevel system;
(2) measuring the relative attitude of the target and the spacecraft;
(3) filtering the measured value obtained in the step (2);
(4) designing a controller for the spacecraft multilevel system in the step (1);
(5) and (4) designing controller parameters for the controller designed in the step (4), thereby realizing the spacecraft three-super-near-zero error tracking control based on the combined filtering.
Furthermore, the spacecraft multilevel system comprises a star platform, an active pointing hyperstatic platform, a load, a sensor system and a fast reflector;
the load is an optical system used for imaging the celestial body;
the quick reflector is arranged in the load and used for adjusting the direction of the optical axis of the load;
the sensor system is used for measuring data;
the star platform is used for supporting the active pointing hyperstatic platform and the load;
the active pointing hyperstatic platform is arranged between the load and the star platform, the upper plane of the active pointing hyperstatic platform is connected with the load, and the lower plane of the active pointing hyperstatic platform is connected with the star platform; the active pointing hyperstatic platform consists of six actuators, and each actuator comprises a displacement sensor, a spring-damping structure and a linear motor;
the displacement sensor is used for measuring the translational displacement of the linear motor; the spring-damping structure is used for isolating the high-frequency vibration of the star platform; the linear motor is used for providing main power and realizing the attitude control of the load.
Further, the linear motor is used as an active link, the spring-damping structure is used as a passive link, and the spring-damping structure and the linear motor are installed in parallel; the spring-damper structure includes a spring and a damper in parallel.
Furthermore, a sensor system of the spacecraft multilevel system comprises a star gyroscope, a load micrometer sensor, a load camera, an active pointing hyperstatic platform displacement measurement sensor and a fast reflector displacement measurement sensor;
the star gyroscope measures the angular speed of a star;
the load micro-sensor measures the angular speed of the load;
the load camera measures the relative posture of the load camera and the target sight;
actively pointing to the hyperstatic platform displacement measurement sensor to measure the relative attitude of the load and the star;
the fast reflector displacement measuring sensor measures the relative attitude of the fast reflector and the load;
the spacecraft realizes the attitude control of the load by applying the load active control moment; the load micrometer sensor is realized by adopting a micrometer optical fiber gyroscope, and the precision is one order of magnitude higher than that of a star gyroscope.
Further, establishing a dynamic model of the spacecraft multilevel system, specifically:
the kinetic equations for stars and loads are:
y＝Cx
wherein the state variable
XpFor load-dependent state quantities, XbThe expressions are respectively as follows:
Xp＝[xp,yp,zp,θxp,θyp,θzp]T，Xb＝[xb,yb,zb,θxb,θyb,θzb]T
wherein x isp、yp、zpRespectively the component of the position vector of the load from the nominal position in the inertial system, thetaxp、θyp、θzpRespectively the roll, pitch and yaw angles of the load relative to the inertial system; x is the number ofb、yb、zbRespectively the components of the position vector of the star with respect to the nominal position in the inertial system, thetaxb、θyb、θzbRolling, pitching and yawing angles of the star relative to an inertia system are respectively obtained;
the state matrices A, B, C are respectively represented as
Wherein M isp,MbGeneralized inertia matrices, M, for loads and stars, respectivelyp＝diag{mp,Ip}， Mb＝diag{mb,Ib}; wherein m isp，IpMass and moment of inertia, m, of the load, respectivelyb，IbThe mass and the moment of inertia of the star body respectively; kpp、Kpb、Kbp、KbbThe rigidity matrixes are respectively used for transferring the load to the load, transferring the load to the star and transferring the load to the star; cpp、Cpb、Cbp、CbbThe damping matrixes are respectively used for transmitting the load to the load, transmitting the load to the star body and transmitting the load to the star body; the conversion forms of the damping array and the rigidity array are respectively expressed
Wherein JpAnd JbRespectively a Jacobian matrix from a load real space and a star body real space to an actuator motion space, and K and C respectively a rigidity matrix and a damping matrix of the actuator;
the output variable is expressed as
The input variable is expressed as
u＝[udpudb+ucFa]T
Wherein l is an array of variable lengths of individual actuators, udp、udbDisturbance forces and moments, u, acting on the load and the star, respectivelycThe array is an array of control force and control moment acting on a star body, is a 6-dimensional array and comprises 3-dimensional control force and 3-dimensional control moment;
Fafor the control force of the actuator, In×nIs an n × n-dimensional unit matrix, wherein n is any positive integer;
the dynamic model of the fast mirror is expressed as:
wherein, TFSMIs the control period of the fast reflecting mirror.
Further, the step (2) measures the relative attitude of the target and the spacecraft, specifically:
(2-1) converting the ground satellite model and the satellite picture into an on-orbit observation image form, and removing noise by filtering:
wherein the content of the first and second substances,
a covariance matrix after pixel point x filtering is obtained; c (y) is a covariance matrix of pixel points y; s (x) is a search window of the pixel point x; ω (x, y) is the filter weight coefficient, C (x) is the covariance matrix of pixel point x, and the calculation formulas are:
ω(p,q)＝exp(-d(X,Y)/h)
d(X,Y)＝lq ln2+l(ln||C(x)||+ln||C(y)||-2ln||C(x)+C(y)||)
wherein l represents the view number, q represents the dimension of the covariance matrix, and h is the attenuation factor of the filter weight.
(2-2) marking the outer contour of the characteristic part of the spacecraft in the filtered image; the characteristic part comprises a solar panel and a star sensor;
(2-3) bringing the filtered image and the marked external contour into a deep neural network for training, and establishing a learning and parameter model of the characteristic part; the deep neural network in the invention can adopt a Deeplabv3 network.
(2-4) when the spacecraft runs in orbit, recognizing the contour of the characteristic part of the target spacecraft by using a learning and parameter model obtained by training, and calculating the centroid of the characteristic part according to the recognized contour, wherein the centroid is used as the target attitude of target tracking control;
and (2-5) calculating a relative attitude measurement value of the target and the spacecraft from the calculated attitude of the target and the attitude of the spacecraft.
Further, on the basis of obtaining a measurement value of the relative attitude of the target and the spacecraft by utilizing deep learning, the measurement value is filtered by combining the dynamic model of the spacecraft multilevel system in the step (1), and the method specifically comprises the following steps:
(3-1) filtering the relative attitude of the load and the target, and calculating according to the following formula:
(3-1-1) order
Filtering the system state for the load versus target relative attitude at time k-1, where θTP(k-1)Is the relative of the load and the target line of sight at the moment k-1Attitude, Δ ωTP(k-1)And (2) obtaining a state equation of a filtering system according to the dynamic model of the spacecraft multistage system in the step (1) if the relative angular speed of the load and the target sight line at the moment k-1 is represented by k 1,2
Wherein, ω isg(k)Is angular velocity of load, wk-1For system noise, the variance matrix is Qk-1，fk-1(xTP(k-1)) Is a filtering system function; delta T is the sampling period of the filter system;
(3-1-2) the measurement equation of the relative attitude of the load and the target is
Wherein the content of the first and second substances,
is a relative attitude measurement of the load to the target line of sight,
as a measure of angular velocity of the load, vkFor measuring noise, the variance matrix is Rk，hk(xTP(k)) For the measurement function, let Pk-1Is xTP(k-1)Is estimated error covariance matrix, Pk-1Cholesky decomposition, i.e.
The volume points are selected as follows:
wherein m is the number of volume points;
[1]iis a set of points [1]To (1) ai columns; the volume point after the state equation of the filtering system is propagated
The state prediction value is
(3-1-3) covariance matrix of state prediction errors of
(3-1-4) to Pk|k-1Cholesky decomposition, i.e.
The volume points are calculated as:
the volume point after propagation through the measurement equation is
The measured predicted value is
(3-1-5) calculating a covariance matrix of measurement errors
(3-1-6) calculating one-step predictive cross-correlation covariance matrix
(3-1-7) correcting the state predicted value according to the relative attitude measurement value obtained in the step (2) to obtain a state estimation value
Wherein the content of the first and second substances,
the state estimation error covariance matrix is
(3-2) filtering the relative attitude of the star and the target according to the following formula:
(3-2-1) order
The system state is filtered for the relative attitude of the star to the target at time k-1, where θTB(k-1)Is the relative attitude of the star and the target line of sight at the time of k-1, delta omegaTP(k-1)And (2) obtaining a state equation of a filtering system according to the dynamic model of the spacecraft multistage system in the step (1) if the relative angular speed of the star body and the target sight line at the moment k-1 is represented by k 1,2
Wherein, ω isgb(k)Is the angular velocity of the star, wTB(k-1)For system noise, the variance matrix is QTB(k-1)，fk-1(xTP(k-1)) Is a filtering system function;
(3-2-2) the relative attitude filtering measurement equation of the star and the target is
Wherein the content of the first and second substances,
the relative attitude measurement value of the star and the target sight line is obtained;
as measured value of a star gyro, vTB(k)For measuring noise, the variance matrix is RTB(k)；hTB(k)(xTB(k)) Is a measurement function; let PTB(k-1)Is xTB(k-1)Is estimated error covariance matrix, PTB(k-1)Cholesky decomposition, i.e.
The volume points are selected as follows:
wherein m isbCounting volume points;
[1]iis a set of points [1]The ith column; the volume point after propagation of the system state equation is
The state prediction value is
(3-2-3) State prediction error covariance matrix of
(3-2-4) to PTB(k|k-1)Cholesky decomposition, i.e.
The volume points are calculated as:
volume after propagation of the measurement equationIs characterized by
The measured predicted value is
(3-2-5) calculating a covariance matrix of measurement errors
(3-2-6) calculating one-step predictive cross-correlation covariance matrix
(3-2-7) correcting the state predicted value according to the relative attitude measurement value obtained in the step (2) to obtain a state estimation value
Wherein the content of the first and second substances,
the state estimation error covariance matrix is
Further, a controller design is performed on the spacecraft multilevel system, and the controller design specifically comprises the following steps:
(4-1) converting the kinetic model in the step (1) into a transfer function matrix form:
wherein b is the input variable dimension, a is the output variable dimension,
for each sheetChannel system Gji(s) performing controller design, wherein j is 1,2,. a, i is 1,2,. b;
(4-2) designing a PID controller for the star body, wherein the input of the controller is the correction value of the relative attitude of the star body and the target in the step (3)
Designing a PID controller for the load, wherein the input of the controller is the relative attitude correction value of the load and the target in the step (3)
The following form of controller is designed for the fast reflecting mirror:
wherein, Ki_FSMFor the integration constant, the closed loop transfer function of the fast reflecting mirror is calculated as:
wherein, ω isnξ is the damping ratio of the fast-reflecting mirror control loop, and the calculation formulas are respectively:
further, controller parameter design is performed on the controller, specifically:
(5-1) designing the bandwidth of a three-level control loop according to the coordination requirement of the spacecraft multilevel system;
the primary star control system is often low in control precision requirement, large in sampling delay and greatly influenced by low-frequency disturbance of flexible bodies such as sailboards, and therefore stability is often used as a main design factor, and low bandwidth is taken.
The secondary load attitude control system has the requirements of accurate pointing and rapid stability, requires the system to have better response speed and better inhibition capability on primary disturbance, and considers the influence of a flexible accessory of a load part on the stability, so the secondary load attitude control system is designed to have medium bandwidth.
The three-stage fast reflecting mirror control system is small in mass inertia and less influenced by flexibility, and the control bandwidth of the three-stage fast reflecting mirror control system is mainly influenced by system delay. Under the condition that the sampling rate is met, in order to realize response and disturbance suppression of higher bandwidth, the three-level controller adopts high-bandwidth design.
Therefore, from the first-level control loop to the third-level control loop, the mass inertia of the control object is changed from large to small, the bandwidth is designed from low bandwidth to high bandwidth, and specifically, the first-level star control bandwidth < the second-level load control bandwidth < the third-level fast reflection mirror control bandwidth.
And (5-2) obtaining the controller parameters of the three-level control loop according to the bandwidth of the three-level control loop.
Compared with the prior art, the invention has the advantages that:
1. the micro-vibration isolation inhibition of the satellite platform is realized, and the load stability index is improved.
The spacecraft three-level cooperative control system designed by the invention realizes the attenuation of 20dB of high-frequency disturbance (>10Hz) in the satellite platform through actively pointing to the passive link of the hyperstatic platform. And the attenuation of 10dB of low-frequency disturbance (<10Hz) in the satellite platform is further realized through active control of the active directional hyperstatic platform. Through a fast reflection mirror passive link, the attenuation of low-frequency micro-vibration (<0.1Hz) of the star platform and low-low disturbance 20dB in load is realized. The micro-vibration isolation and inhibition of the satellite platform are realized through two-stage passive links, and the load stability is improved.
2. The target posture is realized.
The method realizes the rapid resolving of the pose of the feature position of the target spacecraft based on the deep learning, integrates the feature extraction and the feature identification, avoids the manual design of the feature, and has stronger universality, thereby providing accurate control input for a three-level cooperative control system of the spacecraft and improving the tracking precision.
3. The high-precision and high-dynamic attitude determination of the spacecraft system is realized.
A load control loop and a fast reflecting mirror control loop are introduced into a spacecraft control system, so that the control system can realize high-dynamic high-precision attitude determination. The load control loop adopts the voice coil and the displacement sensor, realizes relatively large stroke and high dynamic characteristic, provides relatively stable working environment for the fast-response mirror control loop, adopts optical measurement for the fast-response mirror control loop, realizes high-precision measurement in a small range, and provides a basis for realizing load three-super-performance tracking control.
Drawings
FIG. 1 is a schematic diagram of a multi-stage control of the process of the present invention;
FIG. 2 is a process of feature position and pose solution based on deep learning;
FIG. 3 is a load and star tracking control error, where plot (a) is a load attitude error curve, plot (b) is a load angular velocity error curve, plot (c) is a star attitude error curve, and plot (d) is a star angular velocity error curve;
fig. 4 shows tracking control errors of the fast reflective mirrors, where fig. (a), (b), and (c) show pointing errors of optical axes of the fast reflective mirrors under a two-stage control strategy of the star-fast reflective mirrors with fast reflective mirror control bandwidths of 8Hz, 10Hz, and 15Hz, respectively, and fig. (d) shows pointing errors of optical axes of the fast reflective mirrors under a three-stage control strategy proposed by the method.
Detailed Description
The invention provides a combined filtering-based spacecraft three-ultra-near-zero error tracking control method, which is suitable for the field of target tracking and having the requirement of determining ultrahigh precision of loads. Different from the single-stage attitude control of the traditional spacecraft star platform, the invention provides a star-load-fast reflecting mirror three-stage attitude cooperative control method based on combined filtering aiming at the spacecraft platform with three-stage control performances of ultra-high precision pointing, ultra-high stability control, hypersensitive control and the like, the attitude calculation of a target is improved by utilizing deep learning, the attitude control precision is improved step by step from a star, load and fast reflecting mirror three-stage system, and high-precision attitude control is provided for optical load fast tracking and high-quality imaging.
As shown in fig. 1, a three-super-near-zero error tracking control method for a spacecraft based on combined filtering specifically includes:
(1) the spacecraft multilevel system is a three-level cooperative system and comprises a star platform, an active pointing hyperstatic platform, a load, a sensor system and a fast reflector.
The load is an optical system for high quality imaging of a target celestial body.
The sensor system is used for measuring data;
the star platform is mainly used for supporting the active pointing hyperstatic platform and loads.
The active directional hyperstatic platform is arranged between the load and the star platform, the upper plane of the active directional hyperstatic platform is connected with the load, and the lower plane of the active directional hyperstatic platform is connected with the star platform. The active directional hyperstatic platform consists of six actuators. Each actuator comprises a spring-damping passive link, a linear motor active link and a displacement sensor which are arranged in parallel: the displacement sensor is used for measuring the translational displacement of the linear motor; the spring-damping structure is used for isolating the high-frequency vibration of the star platform; the linear motor is used for providing main power and realizing the attitude control of the load.
The linear motor is used as an active link, the spring-damping structure is used as a passive link, and the spring-damping structure is installed in parallel with the linear motor; the spring-damper structure includes a spring and a damper in parallel.
The quick reflection mirror is arranged in the load and used for adjusting the direction of the optical axis of the load.
The sensor system of the spacecraft multilevel system comprises a star gyroscope, a load micrometer sensor, a load camera, an active pointing hyperstatic platform displacement measurement sensor and a fast reflector displacement measurement sensor;
the star gyroscope measures the angular speed of a star;
the load micro-sensor measures the angular speed of the load;
the load camera measures the relative posture of the load camera and the target sight;
actively pointing to the hyperstatic platform displacement measurement sensor to measure the relative attitude of the load and the star;
the fast reflector displacement measuring sensor measures the relative attitude of the fast reflector and the load;
the spacecraft realizes the attitude control of the load by applying the load active control moment; the load micrometer sensor is realized by adopting a micrometer optical fiber gyroscope, and the precision is one order of magnitude higher than that of a star gyroscope.
According to the three-level cooperative control structure, a dynamic model for establishing a spacecraft multistage system (namely a spacecraft three-level cooperative system dynamic model) is established as follows:
the kinetic equations for stars and loads are:
y＝Cx
wherein the state variable
XpFor load-dependent state quantities, XbThe expressions are respectively as follows:
Xp＝[xp,yp,zp,θxp,θyp,θzp]T，Xb＝[xb,yb,zb,θxb,θyb,θzb]T
wherein x isp、yp、zpRespectively the component of the position vector of the load from the nominal position in the inertial system, thetaxp、θyp、θzpRespectively the roll, pitch and yaw angles of the load relative to the inertial system; x is the number ofb、yb、zbRespectively the components of the position vector of the star with respect to the nominal position in the inertial system, thetaxb、θyb、θzbRolling, pitching and yawing angles of the star relative to an inertia system are respectively obtained;
the state matrices A, B, C are respectively represented as
Wherein M isp,MbGeneralized inertia matrices, M, for loads and stars, respectivelyp＝diag{mp,Ip}， Mb＝diag{mb,Ib}; wherein m isp，IpMass and moment of inertia, m, of the load, respectivelyb，IbThe mass and the moment of inertia of the star body respectively; kpp、Kpb、Kbp、KbbThe rigidity matrixes are respectively used for transferring the load to the load, transferring the load to the star and transferring the load to the star; cpp、Cpb、Cbp、CbbThe damping matrixes are respectively used for transmitting the load to the load, transmitting the load to the star body and transmitting the load to the star body; the conversion forms of the damping array and the rigidity array are respectively expressed
Wherein JpAnd JbRespectively a Jacobian matrix from a load real space and a star body real space to an actuator motion space, and K and C respectively a rigidity matrix and a damping matrix of the actuator;
the output variable is expressed as
The input variable is expressed as
u＝[udpudb+ucFa]T
Wherein l is an array of variable lengths of individual actuators, udp、udbDisturbance forces and moments, u, acting on the load and the star, respectivelycIs an array of control forces and control moments acting on the stars, is a 6-dimensional array, and comprises3-dimensional control force and 3-dimensional control moment; faFor the control force of the actuator, In×nIs an n × n-dimensional unit matrix, wherein n is any positive integer;
the dynamic model of the fast mirror is expressed as:
wherein, TFSMIs the control period of the fast reflecting mirror.
According to the embodiment of the invention, the spacecraft model parameters are substituted into the above formula mb＝765kg， Ib＝diag(692.0,643.1,685.0)kgm2，mp＝510kg，Ip＝diag(200.0,200.0,80.0)kgm2Stiffness k of the actuatorj15000.0N/m, actuator damping cjAnd obtaining a spacecraft three-level cooperative control system model, wherein the speed is 500.0N/m.
(2) And measuring the relative attitude of the target and the spacecraft.
Under a complex space environment, typical parts and motion characteristics of a target spacecraft are accurately positioned, intelligent classification, identification and segmentation are realized on operating parts, measurement data are provided for a target tracking control task, and the relative posture of a target and the target is provided.
(2-1) converting the ground satellite model and the satellite picture into an on-orbit observation image form, and removing noise by filtering:
wherein the content of the first and second substances,
a covariance matrix after pixel point x filtering is obtained; c (y) is a covariance matrix of pixel points y; s (x) is a search window of the pixel point x; ω (x, y) is the filter weight coefficient, C (x) is the covariance matrix of pixel point x, and the calculation formula is:
ω(x,y)＝exp(-d(X,Y)/h)
d(X,Y)＝lqln2+l(ln||C(x)||+ln||C(y)||-2ln||C(x)+C(y)||)
wherein l represents the view number, q represents the dimension of the covariance matrix, and h is the attenuation factor of the filter weight.
(2-2) marking the outer contour of the characteristic part of the spacecraft in the filtered image; the characteristic part comprises a solar panel and a star sensor;
(2-3) bringing the filtered image and the marked external contour into a deep neural network for training, and establishing a learning and parameter model of the characteristic part; the deep neural network can adopt a Deeplab v3 network;
(2-4) when the spacecraft runs in orbit, recognizing the contour of the characteristic part of the target spacecraft by using a learning and parameter model obtained by training, and calculating the centroid of the characteristic part according to the recognized contour, wherein the centroid is used as the target attitude of target tracking control;
and (2-5) calculating a relative attitude measurement value of the target and the spacecraft from the calculated attitude of the target and the attitude of the spacecraft.
(3) And (3) filtering the measured value by combining the dynamic model of the spacecraft multilevel system in the step (1) on the basis of obtaining the relative attitude measured value of the target and the spacecraft by utilizing deep learning.
Filtering the relative attitude of the load to the target may be calculated as follows:
(3-1) filtering the relative attitude of the load and the target, and calculating according to the following formula:
(3-1-1) order
Filtering the system state for the load versus target relative attitude at time k-1, where θTP(k-1)Is the relative attitude of the load and the target line of sight at the time of k-1, delta omegaTP(k-1)And (2) obtaining a state equation of a filtering system according to the dynamic model of the spacecraft multistage system in the step (1) if the relative angular speed of the load and the target sight line at the moment k-1 is represented by k 1,2
Wherein, ω isg(k)Is angular velocity of load, wk-1For system noise, the variance matrix is Qk-1，fk-1(xTP(k-1)) Is a filtering system function; and delta T is the sampling period of the filtering system.
(3-1-2) the measurement equation of the relative attitude of the load and the target is
Wherein the content of the first and second substances,
is a relative attitude measurement of the load to the target line of sight,
as a measure of angular velocity of the load, vkFor measuring noise, the variance matrix is Rk，hk(xTP(k)) For the measurement function, let Pk-1Is xTP(k-1)Is estimated error covariance matrix, Pk-1Cholesky decomposition, i.e.
The volume points are selected as follows:
wherein m is the number of volume points;
[1]iis a set of points [1]The ith column; the volume point after the state equation of the filtering system is propagated
The state prediction value is
(3-1-3) covariance matrix of state prediction errors of
(3-1-4) to Pk|k-1Cholesky decomposition, i.e.
The volume points are calculated as:
the volume point after propagation through the measurement equation is
The measured predicted value is
(3-1-5) calculating a covariance matrix of measurement errors
(3-1-6) calculating one-step predictive cross-correlation covariance matrix
(3-1-7) correcting the state predicted value according to the relative attitude measurement value obtained in the step (2) to obtain a state estimation value
Wherein the content of the first and second substances,
the state estimation error covariance matrix is
(3-2) filtering the relative attitude of the star and the target according to the following formula:
(3-2-1) order
The system state is filtered for the relative attitude of the star to the target at time k-1, where θTB(k-1)Is the relative attitude of the star and the target line of sight at the time of k-1, delta omegaTP(k-1)And (2) obtaining a state equation of a filtering system according to the dynamic model of the spacecraft multistage system in the step (1) if the relative angular speed of the star body and the target sight line at the moment k-1 is represented by k 1,2
Wherein, ω isgb(k)Is the angular velocity of the star, wTB(k-1)For system noise, the variance matrix is QTB(k-1)，fk-1(xTP(k-1)) Is a filtering system function;
(3-2-2) the relative attitude filtering measurement equation of the star and the target is
Wherein the content of the first and second substances,
the relative attitude measurement value of the star and the target sight line is obtained;
as measured value of a star gyro, vTB(k)For measuring noise, the variance matrix is RTB(k)；hTB(k)(xTB(k)) Is a measurement function; let PTB(k-1)Is xTB(k-1)Is estimated error covariance matrix, PTB(k-1)Cholesky decomposition, i.e.
The volume points are selected as follows:
wherein m isbCounting volume points;
[1]iis a set of points [1]The ith column; the volume point after propagation of the system state equation is
The state prediction value is
(3-2-3) State prediction error covariance matrix of
(3-2-4) to PTB(k|k-1)Cholesky decomposition, i.e.
The volume points are calculated as:
the volume point after propagation through the measurement equation is
The measured predicted value is
(3-2-5) calculating a covariance matrix of measurement errors
(3-2-6) calculating one-step predictive cross-correlation covariance matrix
(3-2-7) correcting the state predicted value according to the relative attitude measurement value obtained in the step (2) to obtain a state estimation value
Wherein the content of the first and second substances,
the state estimation error covariance matrix is
(4) And designing a controller for the spacecraft multilevel system.
(4-1) converting the kinetic model in the step (1) into a transfer function matrix form:
where b is the input variable dimension and a is the output variable dimension, for each single channel system Gji(s) performing controller design, wherein j is 1,2,. a, i is 1,2,. b;
(4-2) designing a PID controller for the star body, wherein the input of the controller is the correction value of the relative attitude of the star body and the target in the step (3)
Designing a PID controller for the load, wherein the input of the controller is the relative attitude correction value of the load and the target in the step (3)
The following form of controller is designed for the fast reflecting mirror:
wherein, Ki_FSMFor the integration constant, the closed loop transfer function of the fast reflecting mirror is calculated as:
wherein, ω isnξ is the damping ratio of the fast-reflecting mirror control loop, and the calculation formulas are respectively:
(5) and designing the parameters of the controller.
(5-1) designing the bandwidth of a three-level control loop according to the coordination requirement of the spacecraft multilevel system;
the first-level star control bandwidth < the second-level load control bandwidth < the third-level fast reflection mirror control bandwidth;
the primary star control system is often low in control precision requirement, large in sampling delay and greatly influenced by low-frequency disturbance of flexible bodies such as sailboards, and therefore stability is often used as a main design factor, and low bandwidth is taken.
The secondary load attitude control system has the requirements of accurate pointing and rapid stability, requires the system to have better response speed and better inhibition capability on primary disturbance, and considers the influence of a flexible accessory of a load part on the stability, so the secondary load attitude control system is designed to have medium bandwidth.
The three-stage fast reflecting mirror control system is small in mass inertia and less influenced by flexibility, and the control bandwidth of the three-stage fast reflecting mirror control system is mainly influenced by system delay. Under the condition that the sampling rate is met, in order to realize response and disturbance suppression of higher bandwidth, the three-level controller adopts high-bandwidth design.
And (5-2) obtaining the controller parameters of the three-level control loop according to the bandwidth of the three-level control loop. Therefore, from one level to three levels, the mass inertia of the control object is reduced from large, the bandwidth design is changed from low bandwidth to high bandwidth, and multi-level multi-bandwidth cooperative control is realized.
The embodiment of the invention comprises the following steps:
the MEV1 satellite, manufactured by noroplap grulman, is intended to interface with the Intelsat901, an international communications satellite organization communications satellite 3.6 kilometers from the earth. MEV1 docked with Intelsat901 satellite at month 2 of 2020, an in-orbit observation of Intelsat901 was obtained during the docking process. As shown in fig. 2, satellite feature components are carried out on the on-orbit captured images obtained by the MEV1, and the spacecraft body attitude fusion filtering estimation is carried out through the identified feature information, so that the attitude pointing control accuracy is improved. The test result shows that: the method designed by the invention can identify target components such as a sailboard, an engine nozzle and the like.
The control method and the filtering method are brought into spacecraft multilevel system dynamics for moving target attitude tracking control, and the performance of a three-level cooperative tracking control system is inspected. From the simulation results, fig. 3 shows that the maximum error value of the star point control is less than 10 "at the peak of the target angular acceleration, and the maximum error value of the load point control is less than 0.2" at the peak of the target angular velocity. Therefore, the secondary load control system reduces the error of the primary load from 10 'to within 0.2', and greatly improves the tracking performance of the optical load. On one hand, compared with the traditional two-stage control, the stroke requirement of the fast reflecting mirror is reduced to 1' order from 10 ", so that the field of view of the tracking camera can be further optimized to be reduced, and the measurement noise of the camera can be further reduced under the condition that the resolution of the camera is fixed. On the other hand, on the basis of two-stage control, three-stage cooperative control can further improve the tracking performance of the high-dynamic target. As can be seen from fig. 4(a), 4(b) and 4(c), when the control bandwidth of the fast reflective mirror is low, a large error exists in the tracking process, and the maximum tracking error is 0.8 "at a bandwidth of 8 Hz. Although increasing the control bandwidth increases the tracking accuracy and also causes a decrease in the damping ratio, the measurement noise increases the amount of jitter, and at a bandwidth of 15Hz, the amount of jitter error is 0.5 ". Comparing fig. 4(b) and fig. 4(d), it can be known that after the three-stage cooperative control is adopted, the maximum error in the tracking process is reduced to be within 0.2 ″, and meanwhile, the jitter amount caused by the noise in the whole process is also obviously reduced, which indicates that the three-stage cooperative moving target tracking control method provided by the present invention has an obvious effect on improving the accuracy and stability of the moving target tracking control.
Those skilled in the art will appreciate that those matters not described in detail in the present specification are well known in the art.
Claims (10)
1. A spacecraft three-ultra-near-zero error tracking control method based on combined filtering is characterized by comprising the following steps:
(1) establishing a dynamic model of a spacecraft multilevel system;
(2) measuring the relative attitude of the target and the spacecraft;
(3) filtering the measured value obtained in the step (2);
(4) designing a controller for the spacecraft multilevel system in the step (1);
(5) and (4) designing controller parameters for the controller designed in the step (4), thereby realizing the spacecraft three-super-near-zero error tracking control based on the combined filtering.
2. The combined filtering-based spacecraft three-ultra-near-zero error tracking control method according to claim 1, characterized in that: the spacecraft multilevel system comprises a star platform, an active pointing hyperstatic platform, a load, a sensor system and a fast reflector;
the load is an optical system used for imaging the celestial body;
the quick reflector is arranged in the load and used for adjusting the direction of the optical axis of the load;
the sensor system is used for measuring data;
the star platform is used for supporting the active pointing hyperstatic platform and the load;
the active pointing hyperstatic platform is arranged between the load and the star platform, the upper plane of the active pointing hyperstatic platform is connected with the load, and the lower plane of the active pointing hyperstatic platform is connected with the star platform; the active pointing hyperstatic platform consists of six actuators, and each actuator comprises a displacement sensor, a spring-damping structure and a linear motor;
the displacement sensor is used for measuring the translational displacement of the linear motor; the spring-damping structure is used for isolating the high-frequency vibration of the star platform; the linear motor is used for providing main power and realizing the attitude control of the load.
3. The spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 2, characterized in that: the linear motor is used as an active link, the spring-damping structure is used as a passive link, and the spring-damping structure is installed in parallel with the linear motor; the spring-damper structure includes a spring and a damper in parallel.
4. The spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 2, characterized in that: the sensor system of the spacecraft multilevel system comprises a star gyroscope, a load micrometer sensor, a load camera, an active pointing hyperstatic platform displacement measurement sensor and a fast reflector displacement measurement sensor;
the star gyroscope measures the angular speed of a star;
the load micro-sensor measures the angular speed of the load;
the load camera measures the relative posture of the load camera and the target sight;
actively pointing to the hyperstatic platform displacement measurement sensor to measure the relative attitude of the load and the star;
the fast reflector displacement measuring sensor measures the relative attitude of the fast reflector and the load;
the spacecraft realizes the attitude control of the load by applying the load active control moment; the load micrometer sensor is realized by adopting a micrometer optical fiber gyroscope, and the precision is one order of magnitude higher than that of a star gyroscope.
5. The spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 2, characterized in that: establishing a dynamic model of a spacecraft multilevel system, which specifically comprises the following steps:
the kinetic equations for stars and loads are:
y＝Cx
wherein the state variable
XpFor load-dependent state quantities, XbThe expressions are respectively as follows:
Xp＝[xp,yp,zp,θxp,θyp,θzp]T，Xb＝[xb,yb,zb,θxb,θyb,θzb]T
wherein x isp、yp、zpRespectively the component of the position vector of the load from the nominal position in the inertial system, thetaxp、θyp、θzpRespectively the roll, pitch and yaw angles of the load relative to the inertial system; x is the number ofb、yb、zbRespectively the components of the position vector of the star with respect to the nominal position in the inertial system, thetaxb、θyb、θzbRolling, pitching and yawing angles of the star relative to an inertia system are respectively obtained;
the state matrices A, B, C are respectively represented as
Wherein M isp,MbGeneralized inertia matrices, M, for loads and stars, respectivelyp＝diag{mp,Ip}，Mb＝diag{mb,Ib}; wherein m isp，IpMass and moment of inertia, m, of the load, respectivelyb，IbThe mass and the moment of inertia of the star body respectively; kpp、Kpb、Kbp、KbbThe rigidity matrixes are respectively used for transferring the load to the load, transferring the load to the star and transferring the load to the star; cpp、Cpb、Cbp、CbbThe damping matrixes are respectively used for transmitting the load to the load, transmitting the load to the star body and transmitting the load to the star body; the conversion forms of the damping array and the rigidity array are respectively expressed
Wherein JpAnd JbRespectively a Jacobian matrix from a load real space and a star body real space to an actuator motion space, and K and C respectively a rigidity matrix and a damping matrix of the actuator;
the output variable is expressed as
The input variable is expressed as
u＝[udpudb+ucFa]T
Wherein l is an array of variable lengths of individual actuators, udp、udbDisturbance forces and moments, u, acting on the load and the star, respectivelycThe array is an array of control force and control moment acting on a star body, is a 6-dimensional array and comprises 3-dimensional control force and 3-dimensional control moment;
Fafor the control force of the actuator, In×nIs an n × n-dimensional unit matrix, wherein n is any positive integer;
the dynamic model of the fast mirror is expressed as:
wherein, TFSMIs the control period of the fast reflecting mirror.
6. The spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 5, characterized in that: the step (2) of measuring the relative attitude of the target and the spacecraft is specifically as follows:
(2-1) converting the ground satellite model and the satellite picture into an on-orbit observation image form, and removing noise by filtering:
wherein the content of the first and second substances,
a covariance matrix after pixel point x filtering is obtained; c (y) is a covariance matrix of pixel points y; s (x) is a search window of the pixel point x; ω (x, y) is the filter weight coefficient, C (x) is the covariance matrix of pixel point x, and the calculation formulas are:
ω(x,y)＝exp(-d(X,Y)/h)
d(X,Y)＝lqln2+l(ln||C(x)||+ln||C(y)||-2ln||C(x)+C(y)||)
wherein l represents the view number, q represents the dimension of the covariance matrix, and h is the attenuation factor of the filter weight.
(2-2) marking the outer contour of the characteristic part of the spacecraft in the filtered image; the characteristic part comprises a solar panel and a star sensor;
(2-3) bringing the filtered image and the marked external contour into a deep neural network for training, and establishing a learning and parameter model of the characteristic part;
(2-4) when the spacecraft runs in orbit, recognizing the contour of the characteristic part of the target spacecraft by using a learning and parameter model obtained by training, and calculating the centroid of the characteristic part according to the recognized contour, wherein the centroid is used as the target attitude of target tracking control;
and (2-5) calculating a relative attitude measurement value of the target and the spacecraft from the calculated attitude of the target and the attitude of the spacecraft.
7. The spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 6, characterized in that: on the basis of obtaining a relative attitude measurement value of a target and a spacecraft by utilizing deep learning, the measurement value is filtered by combining the dynamic model of the spacecraft multilevel system in the step (1), and the method specifically comprises the following steps:
(3-1) filtering the relative attitude of the load and the target, and calculating according to the following formula:
(3-1-1) order
Filtering the system state for the load versus target relative attitude at time k-1, where θTP(k-1)Is the relative attitude of the load and the target line of sight at the time of k-1, delta omegaTP(k-1)And (2) obtaining a state equation of a filtering system according to the dynamic model of the spacecraft multistage system in the step (1) if the relative angular speed of the load and the target sight line at the moment k-1 is represented by k 1,2
Wherein, ω isg(k)Is angular velocity of load, wk-1For system noise, the variance matrix is Qk-1，fk-1(xTP(k-1)) Is a filtering system function; delta T is the sampling period of the filter system;
(3-1-2) the measurement equation of the relative attitude of the load and the target is
Wherein the content of the first and second substances,
is a relative attitude measurement of the load to the target line of sight,
as a measure of angular velocity of the load, vkFor measuring noise, the variance matrix is Rk，hk(xTP(k)) For the measurement function, let Pk-1Is xTP(k-1)Is estimated error covariance matrix, Pk-1Cholesky decomposition, i.e.
The volume points are selected as follows:
wherein m is the number of volume points;
[1]iis a set of points [1]The ith column; the volume point after the state equation of the filtering system is propagated
The state prediction value is
(3-1-3) covariance matrix of state prediction errors of
(3-1-4) to Pk|k-1Cholesky decomposition, i.e.
The volume points are calculated as:
the volume point after propagation through the measurement equation is
The measured predicted value is
(3-1-5) calculating a covariance matrix of measurement errors
(3-1-6) calculating one-step predictive cross-correlation covariance matrix
(3-1-7) correcting the state predicted value according to the relative attitude measurement value obtained in the step (2) to obtain a state estimation value
Wherein the content of the first and second substances,
the state estimation error covariance matrix is
(3-2) filtering the relative attitude of the star and the target according to the following formula:
(3-2-1) order
The system state is filtered for the relative attitude of the star to the target at time k-1, where θTB(k-1)Is the relative attitude of the star and the target line of sight at the time of k-1, delta omegaTP(k-1)And (2) obtaining a state equation of a filtering system according to the dynamic model of the spacecraft multistage system in the step (1) if the relative angular speed of the star body and the target sight line at the moment k-1 is represented by k 1,2
Wherein, ω isgb(k)Is the angular velocity of the star, wTB(k-1)For system noise, the variance matrix is QTB(k-1)，fk-1(xTP(k-1)) Is a filtering system function;
(3-2-2) the relative attitude filtering measurement equation of the star and the target is
Wherein the content of the first and second substances,
the relative attitude measurement value of the star and the target sight line is obtained;
as measured value of a star gyro, vTB(k)For measuring noise, the variance matrix is RTB(k)；hTB(k)(xTB(k)) Is a measurement function; let PTB(k-1)Is xTB(k-1)Is estimated error covariance matrix, PTB(k-1)Cholesky decomposition, i.e.
The volume points are selected as follows:
wherein m isbCounting volume points;
[1]iis a set of points [1]The ith column; the volume point after propagation of the system state equation is
The state prediction value is
(3-2-3) State prediction error covariance matrix of
(3-2-4) to PTB(k|k-1)Cholesky decomposition, i.e.
The volume points are calculated as:
the volume point after propagation through the measurement equation is
The measured predicted value is
(3-2-5) calculating a covariance matrix of measurement errors
(3-2-6) calculating one-step predictive cross-correlation covariance matrix
(3-2-7) correcting the state predicted value according to the relative attitude measurement value obtained in the step (2) to obtain a state estimation value
Wherein the content of the first and second substances,
the state estimation error covariance matrix is
8. The spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 7, characterized in that: the method comprises the following steps of designing a controller for a spacecraft multilevel system, specifically:
(4-1) converting the kinetic model in the step (1) into a transfer function matrix form:
wherein b is the input variable dimension, a is the output variable dimension,
the following is for each single channel system Gji(s) performing controller design, wherein j is 1,2,. a, i is 1,2,. b;
(4-2) designing a PID controller for the star body, wherein the input of the controller is the correction value of the relative attitude of the star body and the target in the step (3)
Designing a PID controller for the load, wherein the input of the controller is the relative attitude correction value of the load and the target in the step (3)
The following form of controller is designed for the fast reflecting mirror:
wherein, Ki_FSMFor the integration constant, the closed loop transfer function of the fast reflecting mirror is calculated as:
wherein, ω isnξ is the damping ratio of the fast-reflecting mirror control loop, and the calculation formulas are respectively:
9. the spacecraft three-ultra-near-zero error tracking control method based on the combined filtering according to claim 7, characterized in that: the method comprises the following steps of designing controller parameters of a controller, specifically:
(5-1) designing the bandwidth of a three-level control loop according to the coordination requirement of the spacecraft multilevel system;
and (5-2) obtaining the controller parameters of the three-level control loop according to the bandwidth of the three-level control loop.
10. The combined filtering-based spacecraft three-ultra-near-zero error tracking control method according to claim 9, characterized in that: from the first-level control loop to the third-level control loop, the mass inertia of a control object is changed from large to small, the bandwidth is designed from low bandwidth to high bandwidth, and specifically, the first-level star control bandwidth < the second-level load control bandwidth < the third-level fast reflection mirror control bandwidth.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010350592.5A CN111625010B (en) | 2020-04-28 | 2020-04-28 | Combined filtering-based spacecraft three-ultra-near-zero error tracking control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010350592.5A CN111625010B (en) | 2020-04-28 | 2020-04-28 | Combined filtering-based spacecraft three-ultra-near-zero error tracking control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111625010A true CN111625010A (en) | 2020-09-04 |
| CN111625010B CN111625010B (en) | 2023-04-14 |
Family
ID=72271764
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010350592.5A Active CN111625010B (en) | 2020-04-28 | 2020-04-28 | Combined filtering-based spacecraft three-ultra-near-zero error tracking control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111625010B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113110537A (en) * | 2021-03-31 | 2021-07-13 | 北京控制工程研究所 | Cooperative control method for relative sight tracking of spacecraft |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5400033A (en) * | 1994-02-07 | 1995-03-21 | Rockwell International Corporation | Tracking system for tracking targets with a spacecraft |
| US7835826B1 (en) * | 2005-12-13 | 2010-11-16 | Lockheed Martin Corporation | Attitude determination system for yaw-steering spacecraft |
| CN102519470A (en) * | 2011-12-09 | 2012-06-27 | 南京航空航天大学 | Multi-level embedded integrated navigation system and navigation method |
| CN107036761A (en) * | 2016-11-11 | 2017-08-11 | 大连理工大学 | A kind of large angle maneuver lower band flexible appendage spacecraft rotary inertia in-orbit identification method |
| CN108646775A (en) * | 2018-06-08 | 2018-10-12 | 北京控制工程研究所 | One kind three surpassing the quick motor-driven and fast and stable control method of platform |
| CN108762285A (en) * | 2018-05-25 | 2018-11-06 | 北京控制工程研究所 | A kind of targeted attitude collaborative planning method and system of spacecraft multistage composite control |
| CN108801270A (en) * | 2018-06-08 | 2018-11-13 | 北京控制工程研究所 | A kind of superhigh precision attitude determination method of spacecraft multistage composite control |
| CN109002047A (en) * | 2018-06-08 | 2018-12-14 | 北京控制工程研究所 | A kind of coarse-fine layering speed of spacecraft combines main by integrated multi-stage composite control method |
| CN110466806A (en) * | 2019-07-24 | 2019-11-19 | 北京控制工程研究所 | A method of the attitude of satellite is controlled using CMG |
-
2020
- 2020-04-28 CN CN202010350592.5A patent/CN111625010B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5400033A (en) * | 1994-02-07 | 1995-03-21 | Rockwell International Corporation | Tracking system for tracking targets with a spacecraft |
| US7835826B1 (en) * | 2005-12-13 | 2010-11-16 | Lockheed Martin Corporation | Attitude determination system for yaw-steering spacecraft |
| CN102519470A (en) * | 2011-12-09 | 2012-06-27 | 南京航空航天大学 | Multi-level embedded integrated navigation system and navigation method |
| CN107036761A (en) * | 2016-11-11 | 2017-08-11 | 大连理工大学 | A kind of large angle maneuver lower band flexible appendage spacecraft rotary inertia in-orbit identification method |
| CN108762285A (en) * | 2018-05-25 | 2018-11-06 | 北京控制工程研究所 | A kind of targeted attitude collaborative planning method and system of spacecraft multistage composite control |
| CN108646775A (en) * | 2018-06-08 | 2018-10-12 | 北京控制工程研究所 | One kind three surpassing the quick motor-driven and fast and stable control method of platform |
| CN108801270A (en) * | 2018-06-08 | 2018-11-13 | 北京控制工程研究所 | A kind of superhigh precision attitude determination method of spacecraft multistage composite control |
| CN109002047A (en) * | 2018-06-08 | 2018-12-14 | 北京控制工程研究所 | A kind of coarse-fine layering speed of spacecraft combines main by integrated multi-stage composite control method |
| CN110466806A (en) * | 2019-07-24 | 2019-11-19 | 北京控制工程研究所 | A method of the attitude of satellite is controlled using CMG |
Non-Patent Citations (1)
| Title |
|---|
| 张科备 等: "一种航天器星体和快反镜两级姿态估计方法" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113110537A (en) * | 2021-03-31 | 2021-07-13 | 北京控制工程研究所 | Cooperative control method for relative sight tracking of spacecraft |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111625010B (en) | 2023-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108803649B (en) | A kind of VTOL Reusable Launch Vehicles active disturbance rejection sliding-mode control | |
| CN107655485B (en) | Cruise section autonomous navigation position deviation correction method | |
| CN109002047B (en) | Coarse-fine layering speed and speed combined main-quilt integrated multi-stage composite control method for spacecraft | |
| Lin et al. | Recent advances in precision measurement & pointing control of spacecraft | |
| CN107458630B (en) | Suspension point cross-scale tracking method for suspension type active gravity compensation system | |
| CN111625010B (en) | Combined filtering-based spacecraft three-ultra-near-zero error tracking control method | |
| CN108082538B (en) | Multi-body system low-energy track capturing method considering initial and final constraints | |
| Lampater et al. | Preparation of the pointing and control system of the SOFIA Airborne Telescope for early science missions | |
| CN115892519A (en) | Spacecraft control method for short-distance spacecraft orbit pulse game | |
| CN113359444B (en) | Flexible spacecraft rigid-flexible coupling characteristic intelligent identification method based on neural network | |
| CN112393835B (en) | Small satellite on-orbit thrust calibration method based on extended Kalman filtering | |
| Anthony et al. | On-orbit modal identification of the Hubble space telescope | |
| CN115062416A (en) | Fast regression orbit design method based on discrete space combination optimization | |
| Khoroshylov et al. | Dynamics and attitude control of space-based synthetic aperture radar | |
| CN111578934A (en) | Refraction star optimization method and system based on inertia/astronomical combined navigation application | |
| CN116300417B (en) | Large-scale distributed space telescope primary and secondary mirror formation control method | |
| CN111854765A (en) | Medium-orbit navigation satellite orbit long-term forecasting method | |
| Quadrelli | Dynamics and control of novel orbiting formations with internal dynamics | |
| Prieto et al. | Orbit and attitude control for the European satellite GOCE | |
| CN111781943B (en) | Three-override control method for distributed load pose of spacecraft | |
| CN111536983B (en) | Spacecraft triple-control broadband multi-source multi-stage collaborative attitude determination method and system | |
| Woffinden | On-orbit satellite inspection: navigation and [Delta] v analysis | |
| CN113602533B (en) | Ultra-low orbit satellite orbit control method based on aerodynamic assistance | |
| Yao et al. | Implementation of three dofs small satellite ground simulation system | |
| CN111580532B (en) | Polymerization separation three-phase control method for spacecraft multistage system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |