Satellite inertial attitude high-precision prediction method suitable for large time delay

Abstract

The invention relates to a satellite inertial attitude high-precision prediction method suitable for large time delay, and the method comprises the steps: constructing a projection relation between the motion characteristics of a satellite and a prediction attitude, and carrying out the attitude extrapolation for a dynamic process and a steady-state process through employing a reference attitude linear difference value and adaptive polynomial fitting; high-precision attitude prediction at the dynamic in-place initial stage and in the stable tracking process is achieved, and satellite attitude maneuver and extrapolation precision of the stable tracking process under the condition of large time delay are guaranteed.

Description

Technical Field

[0001] The invention relates to a high-precision prediction method for satellite inertial attitude applicable to large time delays, belonging to the technical field of spacecraft control. Background Art

[0002] The inertial attitude of a satellite can be measured in real time, usually obtained by collaborative filtering of a star-sensitive gyro, for use by an attitude control unit, or as an output interface for use by other subsystems. For example, the information processing subsystem of a remote sensing satellite requires inertial attitude information for real-time image correlation matching. However, due to hardware limitations in the information transmission link, the attitude information obtained by the information processing subsystem may have a large time delay, for example, a delay of 2 seconds. In order to ensure the accuracy of image correlation matching, the accuracy of the satellite attitude needs to be in the order of arc seconds, that is, the information processing system needs to obtain a high-precision real-time attitude estimate, otherwise it will directly affect the matching and correlation accuracy of the target. Traditional attitude estimation methods are mainly applicable to situations where the extrapolation cycle is relatively short. Summary of the invention

[0003] The technical problem to be solved by the present invention is: in order to improve the extrapolation accuracy of satellite attitude maneuvers and stable tracking processes under large delay conditions, a high-precision prediction method for satellite inertial attitude suitable for large delays is proposed. In the process of satellite attitude maneuvers, linear difference extrapolation based on reference attitude is adopted, and in the case of unknown expected attitude in the steady-state process, attitude estimation is performed using adaptive polynomial fitting of order and historical data volume. The two are intelligently and autonomously switched through scheduling logic to ensure the attitude estimation accuracy of the entire stage of satellite motion.

[0004] The technical solution adopted by the present invention is:

[0005] A high-precision satellite inertial attitude prediction method suitable for large time delay includes the following steps:

[0006] Step 1: Read the inertial attitude quaternion of the star body, and convert the inertial attitude quaternion of the star body into the inertial attitude angle of the satellite;

[0007] Step 2: Perform polynomial adaptive fitting on the satellite inertial attitude angle, perform attitude extrapolation based on the fitting coefficient, and then proceed to step 4;

[0008] Step 3: Perform attitude extrapolation based on linear interpolation, and then proceed to step 4;

[0009] Step 4: Select the attitude extrapolation method based on the logical switching conditions to achieve high-precision prediction of the satellite inertial attitude.

[0010] Furthermore, the celestial body inertial attitude quaternion is converted into the satellite inertial attitude angle, specifically:

[0011] The inertial attitude quaternion of the star is converted into the three-axis inertial Euler angle, and the jump is processed accordingly to prevent the attitude from jumping by ±180 degrees. Specifically:

[0012] For the roll angle φ t :

[0013] φ t (k) = φ t (k-1)+f(φ t (k)-φ t (k-1))

[0014] Among them, φ t (k) is the measured rolling angle of the current shot, φ t (k-1) is the measured rolling angle of the previous shot. For any variable x∈R, x0=x-2nπ, where n is an integer obtained by rounding 2π upwards; the same is true for the pitch angle and the yaw angle.

[0015] Furthermore, the satellite inertial attitude angle is adaptively fitted with a polynomial, that is, a polynomial is fitted based on multi-shot three-axis inertial Euler angles and their time-scale historical data, and the order of the polynomial and the length of the historical data can be adjusted.

[0016] Furthermore,

[0017] The specific process of polynomial fitting is as follows:

[0018] (1) For the rolling axis, the calculation formula of the fitting parameter matrix p1 is:

[0019]

[0020] in,

[0021]

[0022]

[0023] φ t (k) is the measured rolling angle value corresponding to the current shot, and its corresponding time scale is t(k), k s is the number of historical data, k r is the order of the polynomial;

[0024] (2) The initial value uses the bound order and the number of historical data beats, saves the extrapolated posture and compares it with the posture at the corresponding moment, and performs accuracy statistics. If the accuracy does not meet the requirements, modify and adjust the polynomial order and the number of historical data beats until the accuracy meets the requirements;

[0025] (3) Similarly, perform polynomial fitting on the pitch axis and yaw axis according to steps (1) to (2).

[0026] Furthermore, the posture extrapolation based on the fitting coefficient is specifically as follows:

[0027] Assume the measured delay is t d seconds, the predicted value of the current real rolling angle is

[0028]

[0029] The current true pitch angle and the current true yaw angle are processed in the same way.

[0030] Furthermore, the posture extrapolation based on linear interpolation is specifically as follows:

[0031]

[0032] Among them, ΔT is the time difference between the current beat and its previous row.

[0033] Furthermore, the posture extrapolation method is selected based on the logic switching condition, specifically:

[0034] Make logical judgments based on whether the satellite is in a dynamic process or a steady-state process. If the satellite is in a dynamic process, introduce attitude extrapolation based on linear interpolation; if the satellite is in a steady-state process, introduce attitude extrapolation based on polynomial fitting to achieve high-precision attitude extrapolation for the entire process.

[0035] The dynamic process refers to the stage of rapid posture maneuvering capture; the steady-state process refers to the stage of stable tracking after the posture maneuvering is in place.

[0036] In a second aspect, the present invention further proposes a processor, which is used to run a program, wherein the program executes the high-precision prediction method for satellite inertial attitude applicable to large time delays when running.

[0037] In a third aspect, the present invention further proposes a non-volatile storage medium, comprising: a computer program product, which, when executed, executes the high-precision prediction method for satellite inertial attitude applicable to large time delays.

[0038] In a fourth aspect, the present invention further proposes a computer program product, which, when executed by a processor, implements the high-precision prediction method for satellite inertial attitude applicable to large time delays.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] (1) The existing satellite attitude extrapolation method is not suitable for the situation with long extrapolation period. The present invention proposes a new estimation method for the problem of satellite real-time inertial attitude determination under large transmission delay. The method fully exploits the attitude characteristics of the dynamic process and steady-state process of attitude maneuvering, and autonomously schedules the reference attitude linear difference and adaptive polynomial fitting extrapolation algorithm through kinematics and working mode characteristics to ensure high-precision attitude prediction in the initial stage of dynamic positioning and stable tracking process. This method has not appeared in existing papers or patents and is novel.

[0041] (2) The present invention proposes a high-precision inertial attitude extrapolation algorithm applicable to the entire process under large time delay. This method can effectively solve the problem of high-precision attitude estimation under long extrapolation periods through autonomous on-board optimization and adaptive estimation. The entire algorithm is simple in design and requires little parameter debugging.

[0042] (3) The present invention proposes a new solution to the problem of high-precision attitude estimation under long extrapolation periods. It does not require additional data input and is computationally simple. The algorithm can be adapted to a large class of satellite systems with similar attitude extrapolation requirements and has strong engineering practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the method of the present invention;

[0044] Figure 2 It is a schematic diagram of the satellite inertial attitude angle (deg);

[0045] Figure 3 Schematic diagram of the satellite attitude estimation error (deg). DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0047] like Figure 1 As shown, the present invention relates to a high-precision prediction method for satellite inertial attitude applicable to large time delay, comprising the following steps:

[0048] Step 1: Read the inertial attitude quaternion of the star body, and convert the inertial attitude quaternion of the star body into the inertial attitude angle of the satellite;

[0049] Step 2: Perform polynomial adaptive fitting on the satellite inertial attitude angle, perform attitude extrapolation based on the fitting coefficient, and then proceed to step 4;

[0050] Step 3: Perform attitude extrapolation based on linear interpolation, and then proceed to step 4;

[0051] Step 4: Select the attitude extrapolation method based on the logical switching conditions to achieve high-precision prediction of the satellite inertial attitude.

[0052] The specific process of step one is:

[0053] The inertial attitude quaternion of the star is converted into the three-axis inertial Euler angle, and the jump is processed accordingly to prevent the attitude from jumping by ±180 degrees. Specifically:

[0054] For the roll angle φ t :

[0055] φ t (k) = φ t (k-1)+f(φ t (k)-φ t (k-1))

[0056] Among them, φ t (k) is the measured rolling angle of the current shot, φ t (k-1) is the measured rolling angle of the previous shot. For any variable x∈R, x0=x-2nπ, where n is an integer obtained by rounding 2π upwards; the same is true for the pitch angle and the yaw angle.

[0057] In the step 2, the satellite inertial attitude angle is adaptively fitted with a polynomial, that is, a polynomial fitting is performed based on multi-shot three-axis inertial Euler angles and their time-scale historical data, and the order of the polynomial and the length of the historical data can be adjusted.

[0058] Take the scroll axis as an example.

[0059] (2.1) For the rolling axis, the calculation formula of the fitting parameter matrix p1 is

[0060]

[0061] in,

[0062]

[0063] φ t (k) is the measured rolling angle value corresponding to the current shot, and its corresponding time scale is t(k), k s is the number of historical data, k r is the order of the polynomial;

[0064] (2.2) The initial value uses the bound order and the number of historical data beats, saves the extrapolated posture and compares it with the posture at the corresponding moment, and performs accuracy statistics. If the accuracy does not meet the requirements, modify and adjust the polynomial order and the number of historical data beats until the accuracy meets the requirements;

[0065] (2.3) Similarly, perform polynomial fitting for the pitch axis and yaw axis according to steps (2.1) to (2.2).

[0066] In the step three, posture extrapolation is performed based on the fitting coefficients.

[0067] Assume the measured delay is t d seconds, the predicted value of the current real rolling angle is

[0068]

[0069] The same process can be done for the pitch axis and yaw axis.

[0070] In the step 4, the posture extrapolation based on linear interpolation is performed, specifically:

[0071]

[0072] Among them, ΔT is the time difference between the current beat and its previous row.

[0073] In the step five, posture extrapolation is selected based on the logic switching condition.

[0074] The polynomial fitting extrapolation method has high steady-state estimation accuracy, but poor dynamic characteristics. Combined with the kinematic characteristics of the satellite, linear differences are introduced in the dynamic process through logical switching conditions, and polynomial fitting is autonomously introduced in the steady-state process to achieve high-precision attitude extrapolation of the entire process.

[0075] Specifically: logical judgment is made according to whether the satellite is in a dynamic process or a steady-state process. If the satellite is in a dynamic process, attitude extrapolation based on linear interpolation is introduced; if the satellite is in a steady-state process, attitude extrapolation based on polynomial fitting is introduced to achieve high-precision attitude extrapolation of the whole process.

[0076] The dynamic process refers to the stage of rapid posture maneuvering capture; the steady-state process refers to the stage of stable tracking after the posture maneuvering is in place.

[0077] The present invention will be further described below in conjunction with the embodiments.

[0078] Example:

[0079] Taking a certain satellite as the background, a set of typical on-orbit conditions are used to simulate and verify the method proposed in this patent application. The satellite is initially in the normal ground-to-earth mode, and the attitude maneuver begins at t=200s. The target is captured through rapid maneuvers. It is in place at t=248s. At this time, high-precision attitude extrapolation is required for image matching and association. At t=250s, the information processing system successfully outputs image information, and the target enters the stable tracking stage based on image feedback. The system uses a 4th-order polynomial by default, with 5 seconds of historical data. Through self-correction of prediction accuracy feedback, the polynomial order is adjusted to 5th order. The attitude estimation accuracy of the maneuvering process is better than 0.003 degrees, and the attitude estimation accuracy of the stable tracking process is better than 2e-4 degrees. See the specific simulation curves. Figure 2-Figure 3 .

[0080] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

[0081] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A high-precision satellite inertial attitude prediction method suitable for large time delay, characterized in that: include: Step 1: Read the inertial attitude quaternion of the star body, and convert the inertial attitude quaternion of the star body into the inertial attitude angle of the satellite; Step 2: Perform polynomial adaptive fitting on the satellite inertial attitude angle, perform attitude extrapolation based on the fitting coefficient, and then proceed to step 4; Step 3: Perform attitude extrapolation based on linear interpolation, and then proceed to step 4; Step 4: Select the attitude extrapolation method based on the logical switching conditions to achieve high-precision prediction of the satellite inertial attitude.

2. The high-precision satellite inertial attitude prediction method suitable for large time delay according to claim 1 is characterized in that: The inertial attitude quaternion of the star body is converted into the inertial attitude angle of the satellite, specifically: The inertial attitude quaternion of the star is converted into the three-axis inertial Euler angle, and the jump is processed accordingly to prevent the attitude from jumping by ±180 degrees. Specifically: For the roll angle φ t : f t (k)=φ t (k-1)+f(φ t (k)-φ t (k-1)) Among them, φ t (k) is the measured rolling angle of the current shot, φ t (k-1) is the measured rolling angle of the previous shot. For any variable x∈R, n is an integer obtained by rounding up 2π; the same is true for the pitch angle and yaw angle.

3. The high-precision satellite inertial attitude prediction method suitable for large time delay according to claim 2 is characterized in that: The polynomial adaptive fitting of the satellite inertial attitude angle is performed based on the multi-shot three-axis inertial Euler angle and its time-scale historical data. The order of the polynomial and the length of the historical data can be adjusted.

4. The high-precision satellite inertial attitude prediction method applicable to large time delay according to claim 3 is characterized in that: The specific process of polynomial fitting is: (1) For the rolling axis, the calculation formula of the fitting parameter matrix p1 is: in, φ t (k) is the measured rolling angle value corresponding to the current shot, and its corresponding time scale is t(k), k s is the number of historical data, k r is the order of the polynomial; (2) The initial value uses the bound order and the number of historical data beats, saves the extrapolated posture and compares it with the posture at the corresponding moment, and performs accuracy statistics. If the accuracy does not meet the requirements, modify and adjust the polynomial order and the number of historical data beats until the accuracy meets the requirements; (3) Similarly, perform polynomial fitting on the pitch axis and yaw axis according to steps (1) to (2).

5. The high-precision satellite inertial attitude prediction method applicable to large time delay according to claim 4 is characterized in that: The posture extrapolation based on the fitting coefficient is specifically as follows: Assume the measured delay is t d seconds, the predicted value of the current real rolling angle is The current true pitch angle and the current true yaw angle are processed in the same way.

6. The high-precision satellite inertial attitude prediction method applicable to large time delay according to claim 4 is characterized in that: The posture extrapolation based on linear interpolation is specifically as follows: Among them, ΔT is the time difference between the current beat and its previous row.

7. The high-precision satellite inertial attitude prediction method applicable to large time delay according to claim 1 is characterized in that: The posture extrapolation method is selected based on the logic switching condition, specifically: Make logical judgments based on whether the satellite is in a dynamic process or a steady-state process. If the satellite is in a dynamic process, introduce attitude extrapolation based on linear interpolation; if the satellite is in a steady-state process, introduce attitude extrapolation based on polynomial fitting to achieve high-precision attitude extrapolation for the entire process. The dynamic process refers to the stage of rapid posture maneuvering capture; the steady-state process refers to the stage of stable tracking after the posture maneuvering is in place.

8. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the high-precision satellite inertial attitude prediction method applicable to large time delay as described in any one of claims 1 to 7.

9. A non-volatile storage medium, characterized in that: include: A computer program product, when executed, performs the high-precision prediction method for satellite inertial attitude applicable to large time delay as described in any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is executed by a processor, the high-precision prediction method for satellite inertial attitude applicable to large time delay as described in any one of claims 1 to 7 is implemented.

Images