Satellite attitude collaborative calibration method based on inter-satellite link for Internet constellation
Technical Field
The invention relates to the technical field of satellite attitude estimation, in particular to a satellite attitude collaborative calibration method based on an inter-satellite link for an internet constellation.
Background
In recent years, low-orbit internet satellite constellation is becoming a hotspot in the field of global communications. The low-orbit satellite constellation provides low-delay, high-bandwidth communication services for global users by deploying a large number of low-orbit satellites. In order to achieve stable communication between satellites, particularly, a laser link (inter-satellite link) between low-orbit satellites is widely used. Precise control of satellite attitude is critical to the alignment and communication quality of the inter-satellite links, as high-speed motion and orbital variation of satellites can cause attitude errors, thereby affecting the stability and communication efficiency of the links. Therefore, the satellite constellation needs to be subjected to high-precision attitude calibration correction regularly to ensure alignment of the laser link.
The existing satellite attitude correction method mainly depends on an attitude sensor (such as a star sensor, a gyroscope and the like) of a satellite and a ground measurement and control instruction. However, these methods have certain limitations in facing a large number of low-orbit satellites and complex operating environments. The existing satellite attitude calibration technology mainly has the following defects:
1) Limitations of relying on-board sensors
Single star attitude determination methods typically rely on-board sensors (e.g., star sensors, gyroscopes, etc.), but these sensors suffer from measurement errors and drift problems. When the star sensor fails, it is difficult for a single star gesture to maintain high accuracy, which can directly affect the alignment accuracy of the laser link.
2) High dependence on ground measurement and control
The conventional satellite attitude calibration method generally relies on measurement and control instructions of a ground station to correct the attitude of a satellite. However, due to the large number of low-orbit constellation satellites and the short orbit period, the ground station is difficult to cover efficiently and calibrate the attitude in real time, and the problems of delay and poor timeliness exist. This approach is particularly inapplicable in low-orbit constellations and cannot meet real-time and efficient pose correction requirements.
3) Lack of multi-satellite collaborative calibration mechanism
Neither the calibration method based on the data of the single satellite attitude sensor nor the ground dominant calibration method fully utilizes the attitude data of other satellites in the constellation to carry out collaborative calibration. In the prior art, inter-satellite links are mainly used for data transmission, and attitude information of a plurality of satellites in a constellation cannot be effectively utilized for cooperative work, so that an efficient attitude error correction mechanism is lacked.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a satellite attitude collaborative calibration method based on inter-satellite links for an internet constellation, which utilizes inter-satellite link multi-vector joint attitude determination of adjacent satellites, and when an attitude sensor is effective, attitude errors are calibrated through link vectors, so that the attitude calibration precision is improved; in the event of a failure of the attitude sensor, reliable attitude estimation and maintenance functions are provided through link assistance.
The technical scheme of the invention is that the satellite attitude collaborative calibration method based on the inter-satellite link for the Internet constellation comprises the following steps:
S1), acquiring node satellite and adjacent satellite data;
s2), calculating a link inertia vector in an ideal state through the orbit vector;
S3), calculating an attitude estimation value through the relation between the link inertia vector and the actual link vector in an ideal state;
s4), correcting the estimated attitude based on the error attitude correction amount when the satellite attitude sensor works normally;
s5), according to the attitude estimation or correction, the alignment or alignment optimization of the link is completed.
Preferably, in the step S1), the link is built according to a cross configuration, and the acquired node satellite and adjacent satellite data comprise satellite orbit positions, link load optical axis vectors and satellite self-attitude data.
Preferably, in step S2), the calculation formula of the link inertia vector in the ideal state is:
Wherein l i-1,i eci is a link inertia vector of S i-1,j pointing to S i,j, l i+1,i eci is a link inertia vector of S i+1,j pointing to S i,j, l j-1,j eci is a link inertia vector of S i,j-1 pointing to S i,j, l j+1,j eci is a link inertia vector of S i,j+1 pointing to S i,j, for satellite S i,j on constellation, i is a track plane number, j is a track in-plane satellite number, S i-1,j、Si+1,j、Si,j-1、Si,j+1 is a neighboring satellite of satellite S i,j, r i,j is an orbit position of satellite S i,j, r i-1,j、ri+1,j、ri,j-1、ri,j+1 is an orbit position of satellite S i-1,j、Si+1,j、Si,j-1、Si,j+1, upper label b is an entity coordinate system, and upper label eci is an inertia coordinate system.
Preferably, in step S3), the attitude estimation value is calculated by a relation between the link inertia vector and the actual link vector in the ideal state, and specifically includes the steps of:
s31), the relationship between the link inertia vector and the actual link vector in the ideal state is expressed as:
Lb=AkjLeci (1);
Wherein L b is an actual link vector matrix under an ontology, L eci is a link inertia vector matrix under an ideal state, A kj is an attitude estimation matrix, and subscript kj represents an index of the attitude estimation matrix;
S32), constructing a pseudo inverse solution equation (1) by minimizing the error square sum, to obtain an inverse solution matrix a *:
A*=Lb(Leci)T(Leci(Leci)T)-1 (2);
Wherein A * represents an inverse solution matrix, L eci)T represents a transpose matrix of a link inertial vector matrix L eci in an ideal state, T represents a transpose operation;
S33), orthogonalizing the inverse solution matrix A * to enable the inverse solution matrix A * to meet the orthogonalization constraint condition of the rotation matrix, and optimizing to obtain a final attitude estimation matrix A kj;
S34), converting the final attitude estimation matrix A kj into an estimated attitude quaternion
Preferably, in step S4), the estimated attitude is corrected based on the error attitude correction amount, and specifically includes the steps of:
S41), combining the laser load error and the attitude determination error into a correction quantity Q e;
S42), estimating attitude quaternion based on the error attitude correction amount Q e Correcting to obtain a corrected satellite attitude quaternion
Preferably, in step S5), when the attitude sensor fails, the estimated attitude quaternion is utilizedMaintaining alignment of the laser link.
Preferably, in step S5), when the attitude sensor is active, the corrected satellite attitude quaternion is appliedThe input control system further adjusts the gesture to realize link alignment optimization.
The beneficial effects of the invention are as follows:
1. The invention utilizes the relation between the theoretical inertia vector and the actual load vector of the inter-satellite link, constructs a high-precision attitude estimation matrix through pseudo-inverse solution and orthogonal optimization, and further converts the attitude estimation matrix into quaternion for attitude correction;
2. The invention fully utilizes the inter-satellite link vector information, provides a reliable auxiliary mechanism for attitude estimation, and can ensure higher attitude precision even under the condition of complex environment or increased sensor error;
3. According to the invention, the attitude of the target satellite is estimated by combining the attitude and the orbit data of the adjacent satellites through the orbit vector and the link load information, so that the alignment and the communication stability of the laser link in a short period are ensured;
4. According to the invention, a distributed attitude calibration architecture is constructed through the synergistic effect of a plurality of satellites, and when a star sensor is effective, the sensor error is corrected through link attitude estimation;
5. the invention adopts an optimization mechanism for minimizing the square sum of link errors in the distributed architecture, and through comprehensive equalization of link performance, the condition that certain links have better alignment effect and other links are worse is effectively avoided in the system calibration process, and the equalization optimization characteristic ensures the consistency of the overall precision of the attitude of the target satellite and the link performance, so that the overall stability of the link is obviously improved;
6. the distributed architecture of the invention has higher robustness in calibration precision and link alignment effect, and is particularly suitable for a high dynamic environment such as a low-orbit constellation.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention;
Fig. 2 is a schematic diagram of a cross-configured inter-satellite link for a satellite in accordance with an embodiment of the present invention.
Detailed Description
The following is a further description of embodiments of the invention, taken in conjunction with the accompanying drawings:
As shown in fig. 1, the embodiment provides a satellite attitude collaborative calibration method based on an inter-satellite link for an internet constellation, which includes the following steps:
S1), acquiring node satellite and adjacent satellite data;
In this embodiment, as shown in fig. 2, when the link is built according to the cross configuration, the adjacent satellite of the satellite S i,j is S i-1,j,Si+1,j,Si,j-1,Si,j+1.
The satellite data acquired by the embodiment comprises an orbit position, a link load optical axis vector and satellite self-attitude data.
The node satellite acquires an orbit vector r of the satellite under an inertial system according to own orbits, acquires orbit vectors of adjacent satellites through inter-satellite links, and acquires a load optical axis vector matrix under the satellite system, namely an actual link vector matrix L b:
Lb=[pi,i-1 b pj,j-1 b pi,i+1 b pj,j+1 b];
Wherein,
Wherein, Q= [ Q 0,q1,q2,q3]T ] represents a satellite attitude inertial quaternion, an upper mark T represents a transposition operator, an upper mark b represents an ontology coordinate system, an upper mark eci represents an inertial coordinate system, Q 0,q1,q2,q3 respectively represents each element of the quaternion, o represents a rotation operation of the quaternion and a vector, a subscript i is a track plane number, and a subscript j is a track plane inner satellite number;
When the attitude sensor (such as a star sensor) of the satellite works normally, the attitude data of the satellite is acquired. S2), calculating a link inertia vector in an ideal state through the orbit vector, wherein the calculation formula is as follows:
Wherein l i-1,i eci is a link inertia vector of S i-1,j pointing to S i,j, l i+1,i eci is a link inertia vector of S i+1,j pointing to S i,j, l j-1,j eci is a link inertia vector of S i,j-1 pointing to S i,j, l j+1,j eci is a link inertia vector of S i,j+1 pointing to S i,j, for satellite S i,j on constellation, i is a track plane number, j is a track in-plane satellite number, S i-1,j、Si+1,j、Si,j-1、Si,j+1 is a neighboring satellite of satellite S i,j, r i,j is an orbit position of satellite S i,j, r i-1,j、ri+1,j、ri,j-1、ri,j+1 is an orbit position of satellite S i-1,j、Si+1,j、Si,j-1、Si,j+1, upper label b is an entity coordinate system, and upper label eci is an inertia coordinate system.
S3) calculating an attitude estimation value through the relation between the link inertia vector and the actual link vector in an ideal state, wherein the method specifically comprises the following steps of:
s31), the relationship between the link inertia vector and the actual link vector in the ideal state is expressed as:
Lb=AkjLeci (1);
Wherein L b is an actual link vector matrix under an ontology, L eci is a link inertia vector matrix under an ideal state, A kj is an attitude estimation matrix, and subscript kj represents an index of the attitude estimation matrix.
Wherein,
Lb=[pi,i-1 b pj,j-1 b pi,i+1 b pj,j+1 b];
Leci=[li-1,i eci lj-1,j eci li+1,i eci lj+1,j eci];
S32), constructing a pseudo inverse solution equation (1) by minimizing the error square sum, to obtain an inverse solution matrix a *:
A*=Lb(Leci)T(Leci(Leci)T)-1 (2);
Wherein A * represents an inverse solution matrix, and L eci)T represents a transpose matrix of a link inertia vector matrix L eci in an ideal state;
S33), orthogonalizing the inverse solution matrix A * to enable the inverse solution matrix A * to meet the orthogonalization constraint condition of the rotation matrix, and optimizing to obtain a final attitude estimation matrix A kj;
Wherein a 11~a33 represents each element of the 3x3 matrix A kj, respectively;
S34), converting the final attitude estimation matrix A kj into an estimated attitude quaternion
The quaternion solution formula is as follows:
Q 0,q1,q2,q3 is calculated based on the first line of formulas (7) - (10), the absolute value of q 0,q1,q2,q3 is compared, and the formula corresponding to the maximum absolute value is selected to calculate the estimated gesture
S4), when the satellite attitude sensor works normally, correcting the estimated attitude based on the error attitude correction amount, and specifically comprising the following steps:
S41), combining the laser load error and the attitude determination error into an error attitude correction quantity Q e, wherein the expression of the error attitude correction quantity Q e is as follows:
In the formula, Respectively represent quaternionsQ s is the attitude quaternion determined by the satellite attitude sensor; To estimate the pose;
S42), estimating attitude quaternion based on the error attitude correction amount Q e Correcting to obtain a corrected satellite attitude quaternionNamely:
wherein W e is the error attitude correction amount, and W s is the attitude quaternion determined by the satellite attitude sensor.
S5), according to the attitude estimation or correction, the alignment or alignment optimization of the link is completed.
In this embodiment, when the attitude sensor fails, the estimated attitude quaternion is utilizedMaintaining alignment of the laser link.
When the attitude sensor is effective, the corrected satellite attitude quaternion is used forThe input control system further adjusts the gesture to realize link alignment optimization.
The foregoing embodiments and description have been provided merely to illustrate the principles and best modes of carrying out the invention, and various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.