A short wave satellite attitude control system

Abstract

The application provides a short-wave satellite attitude control system and relates to the technical field of data processing, and comprises the following steps: inputting a three-axis combined control torque instruction into a reaction flywheel controller, converting the three-axis combined control torque instruction into a flywheel motor speed regulation instruction and driving the reaction flywheel to execute; based on actual attitude changes after driving the reaction flywheel to execute, adjusting and analyzing a continuously updated attitude deviation angle to obtain updated sliding mode controller gain parameters and feeding back the parameters of the updated sliding mode controller; based on the parameters of the updated sliding mode controller, controlling, verifying the direction consistency of newly output sun direction vector measurement values, and verifying the actual included angle; if the actual included angle is less than a preset precision threshold, triggering a new round of data acquisition cycle to realize closed-loop control. The application guarantees the timeliness and effectiveness of attitude control and improves the efficiency and reliability of control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a short-wave satellite attitude control system. BACKGROUND

[0002] As an important communication method, short-wave communication plays an irreplaceable role in the fields of aerospace, emergency communication, etc. The stability and accurate control of the attitude of short-wave satellite, as a key platform for realizing long-distance short-wave communication, directly determines the communication quality and task completion efficiency. With the continuous expansion of short-wave communication business and the increasing requirement for communication quality, the traditional satellite attitude control system may not meet the precision attitude control requirements of short-wave satellite in complex space environment.

[0003] When processing the measurement data such as the sun direction vector, the traditional attitude solving method often uses a simple vector comparison method, which fails to fully consider the nonlinear characteristics of the spherical coordinates, resulting in a large error in the calculation of the attitude deviation angle, and it is difficult to meet the requirements of short-wave satellite for attitude accuracy. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a short-wave satellite attitude control system, which ensures the timeliness and effectiveness of attitude control and improves the efficiency and reliability of control.

[0005] To solve the above technical problems, the technical scheme of the present application is as follows:

[0006] In a first aspect, a short-wave satellite attitude control system comprises:

[0007] An acquisition module is configured to obtain angular velocity measurement values and sun direction vector measurement values in a satellite body coordinate system;

[0008] A calculation module is configured to perform vector difference operation on the sun direction vector measurement values and the sun reference direction predicted by ephemeris through an orbit attitude solver, and convert the vector difference into spherical azimuth deviation and elevation angle deviation through a spherical coordinate mapping algorithm to obtain an attitude deviation angle;

[0009] A control module is configured to input the attitude deviation angle into a sliding mode controller and generate a three-axis attitude correction torque command through a variable structure control algorithm;

[0010] A compensation module is configured to calculate a gravity gradient interference torque and a solar pressure interference torque based on a preset orbit perturbation model, generate a compensation torque after compensation processing, and superimpose the compensation torque on the three-axis attitude correction torque command to generate a three-axis combined control torque command;

[0011] The execution module is configured to input the three-axis combined control torque instruction into the reaction flywheel controller, convert the three-axis combined control torque instruction into a flywheel motor speed regulation instruction, and drive the reaction flywheel to execute;

[0012] The analysis module is configured to adjust and analyze the continuously updated attitude deviation angle based on an actual attitude change after the reaction flywheel is driven to execute, to obtain updated gain parameters of the sliding mode controller, and feed back the parameters of the updated sliding mode controller;

[0013] The verification module is configured to perform direction consistency verification on a newly output sun direction vector measurement value based on the parameters of the updated sliding mode controller, verify an actual included angle, and trigger a new round of data acquisition period to realize closed-loop control if the actual included angle is less than a preset precision threshold.

[0014] Further, the sun direction vector measurement value is subjected to vector difference operation with a sun reference direction predicted by ephemeris through an orbit attitude solver, and the vector difference is converted into a spherical azimuth angle deviation and an elevation angle deviation through a spherical coordinate mapping algorithm to obtain an attitude deviation angle, including:

[0015] In the orbit attitude solver, a sun direction vector measurement value in a satellite body coordinate system and a sun reference direction vector predicted by ephemeris are obtained, component values of the two on X, Y and Z axes are extracted respectively, subtraction operation on corresponding coordinate axes is performed based on the component values to generate axis component difference values, the component difference values are reorganized according to orthogonal directions of the satellite body coordinate system to generate a sun direction vector difference; and based on the satellite body coordinate system, an orbit normal axis is selected as an elevation angle reference axis, and an orbit plane reference surface is established;

[0016] The sun direction vector difference is projected onto the orbit plane reference surface to obtain a plane projection vector; an azimuth angle of the plane projection vector relative to a preset reference direction axis in the orbit plane reference surface is calculated to obtain a spherical azimuth angle deviation; and an included angle between the sun direction vector difference and the elevation angle reference axis is calculated to obtain a spherical elevation angle deviation;

[0017] The spherical azimuth angle deviation and the spherical elevation angle deviation are fused to finally obtain the attitude deviation angle.

[0018] Further, the attitude deviation angle is input into the sliding mode controller, and a three-axis attitude correction torque instruction is generated through a variable structure control algorithm, including:

[0019] Based on the attitude deviation angle and the historically stored attitude deviation angle, the sliding mode controller calculates a rate of change of the attitude deviation angle with time to generate an attitude deviation rate; the attitude deviation angle and the attitude deviation rate are weighted and combined according to a preset proportional coefficient inside the sliding mode controller to generate a sliding mode surface input value; and the sliding mode surface input value is input into the sliding mode controller to obtain a sliding mode surface state value.

[0020] The equivalent control component is calculated based on the input value of the sliding mode surface and a preset equivalent control gain parameter; the switching control component is calculated based on the sign of the state value of the sliding mode surface and a preset switching control gain parameter; the equivalent control component and the switching control component are subjected to scalar addition operation, and a sliding mode control signal is generated by fusion;

[0021] Based on the sliding mode control signal and the control distribution matrix pre-stored in the sliding mode controller, a three-axis control torque reference vector is calculated and generated, and the reference vector is distributed to each control axis according to the three orthogonal directions of the roll axis, the pitch axis and the yaw axis of the satellite body coordinate system, so as to generate a three-axis attitude correction torque instruction.

[0022] Further, based on a preset orbit perturbation model, the gravity gradient interference torque and the solar pressure interference torque are calculated, a compensation torque is generated after compensation processing, and the compensation torque is superimposed on the three-axis attitude correction torque instruction to generate a three-axis combined control torque instruction, including:

[0023] Based on the real-time orbit data of the satellite and the preset orbit perturbation model, the orbit position parameters of the satellite are obtained; based on the orbit position parameters, the gravity gradient interference torque component in the orbit coordinate system is calculated and converted to the satellite body coordinate system; based on the solar direction vector measurement value, the solar pressure interference torque component in the satellite body coordinate system is calculated;

[0024] The gravity gradient interference torque component and the solar pressure interference torque component are subjected to vector addition operation in the satellite body coordinate system to generate a total interference torque vector;

[0025] The total interference torque vector is subjected to a negative operation to generate a compensation torque vector; the three-axis attitude correction torque instruction and the compensation torque vector are subjected to vector addition operation in the satellite body coordinate system to generate a three-axis combined control torque instruction.

[0026] Further, the three-axis combined control torque instruction is input into the reaction flywheel controller to be converted into a flywheel motor speed regulation instruction and drive the reaction flywheel to execute, including:

[0027] The three-axis combined control torque instruction is subjected to decomposition processing to generate target control torque components of the roll, pitch and yaw axes; based on the target control torque components of the axes, the target angular accelerations of the axes are calculated in combination with the moment of inertia parameters; and based on the preset flywheel configuration information, the axes responsible for each flywheel and the distribution proportion thereof are determined;

[0028] Based on the target angular accelerations of the control axes, the flywheel configuration and the distribution proportion, the target torque required by each reaction flywheel is calculated; based on the target torque in combination with the moment of inertia parameters of the flywheel, the target speed change amount required by the flywheel motor is calculated;

[0029] Based on the target speed change amount, combined with the real-time speed feedback value of each flywheel motor currently obtained from the flywheel motor, the flywheel motor speed adjustment instruction value is calculated and generated; the flywheel motor speed adjustment instruction value is changed through the driving reaction flywheel execution speed.

[0030] Further, based on the actual attitude change after driving the reaction flywheel to execute, the continuously updated attitude deviation angle is adjusted and analyzed to obtain the updated sliding mode controller gain parameter, and the parameter of the updated sliding mode controller is fed back, including:

[0031] Based on the actual attitude change of the satellite after driving the reaction flywheel to execute, the updated angular velocity measurement value and the sun direction vector measurement value in the satellite body coordinate system are continuously obtained; the updated sun direction vector measurement value is input into the orbit attitude solver to generate the updated attitude deviation angle;

[0032] Based on the updated attitude deviation angle, combined with the historically stored attitude deviation angle data, a time series data is established; based on the time series data, the statistical characteristics of the attitude deviation angle in the preset time window are calculated;

[0033] Based on the statistical characteristics of the attitude deviation angle, combined with the preset sliding mode controller gain adjustment strategy rule library for analysis; according to the analysis result, the gain adjustment factor for the equivalent control gain parameter and the switching control gain parameter of the sliding mode controller is generated;

[0034] The gain adjustment factor is multiplied by the equivalent control gain parameter and the switching control gain parameter stored in the sliding mode controller respectively, and the updated equivalent control gain parameter value and the updated switching control gain parameter value are calculated;

[0035] The equivalent control gain parameter value and the switching control gain parameter value are fed back to the internal parameter storage of the sliding mode controller, and the online update of the sliding mode controller parameter is completed.

[0036] Further, based on the updated parameters of the sliding mode controller, the newly output sun direction vector measurement value is verified for direction consistency, and the actual included angle is verified; if the actual included angle is less than the preset precision threshold, a new round of data acquisition period is triggered, realizing closed-loop control, including:

[0037] Based on the updated sliding mode controller parameters, a complete control chain is executed to drive the satellite attitude adjustment, and the newly output sun direction vector measurement value under the new parameter control is obtained;

[0038] The newly output sun direction vector measurement value is dot product operated with the sun reference direction vector predicted by ephemeris, to calculate the actual direction included angle therebetween; the actual direction included angle is compared with the preset threshold to determine whether the actual direction included angle is less than the threshold.

[0039] According to the judgment result, if the direction included angle is less than the threshold value, a new round of data acquisition period is triggered, and if the direction included angle is not less than the threshold value, the control flow is maintained to continue control and verification.

[0040] Further, based on the satellite real-time orbit data and the preset orbit perturbation model, the orbit position parameters of the satellite are obtained; based on the orbit position parameters, the gravity gradient interference torque component in the orbit coordinate system is calculated and converted to the satellite body coordinate system; based on the solar direction vector measurement value, the solar pressure interference torque component in the satellite body coordinate system is calculated, including:

[0041] The orbit coordinate system position coordinate value extracted based on the satellite real-time orbit data is calculated to obtain the geocentric modulus scalar value, the gravity field strength scalar value is generated in combination with the pre-stored earth gravity constant, and the three-axis gravity gradient torque component is generated by calling the satellite moment of inertia main diagonal element value;

[0042] According to the three-axis gravity gradient torque component, the satellite real-time attitude quaternion is combined to construct a rotation transformation matrix, and the vector coordinate system conversion operation is performed through the matrix to obtain the three-axis gravity gradient interference torque component value;

[0043] Based on the solar incident direction unit vector, the effective projection area and the reflection coefficient are output from the satellite surface optical parameter database, the light pressure scalar value is calculated in combination with the solar pressure intensity constant, and the solar pressure interference torque component is generated through the torque fusion operation of the light pressure scalar value and the pre-calibration position deviation vector.

[0044] In a second aspect, a computing device includes:

[0045] One or more processors;

[0046] A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, so that the one or more processors implement the system.

[0047] In a third aspect, a computer readable storage medium stores a program, which is executed by a processor to implement the system.

[0048] The above-mentioned scheme of the present application at least includes the following beneficial effects:

[0049] The orbit attitude solver performs vector difference operation on the solar direction vector measurement value and the ephemeris prediction reference value, and obtains the attitude deviation angle in combination with the spherical coordinate mapping algorithm, which realizes accurate quantization of the attitude deviation, can quickly respond to the attitude deviation and generate a correction torque, effectively suppresses system disturbance, and improves the dynamic stability of the attitude control.

[0050] The synthetic control torque is directly converted into a motor speed regulation instruction by the reaction flywheel controller, the delay of intermediate conversion links is reduced, and the execution speed of the control instruction is improved. The link design of the overall process from measurement, calculation, control to feedback is simple and efficient, and the overall complexity is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 is a short-wave satellite attitude control system schematic diagram provided by an embodiment of the application.

[0052] Figure 2 is a three-axis synthetic control torque instruction flow schematic diagram of an embodiment of the application. DETAILED DESCRIPTION

[0053] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0054] As Figure 1 shown, an embodiment of the application proposes a short-wave satellite attitude control system, comprising:

[0055] The acquisition module is configured to obtain angular velocity measurement values and sun direction vector measurement values in a satellite body coordinate system.

[0056] The calculation module is configured to perform vector difference operation on the sun direction vector measurement values and the sun reference direction predicted by the ephemeris through an orbit attitude solver, and convert the vector difference into a spherical azimuth angle deviation and an elevation angle deviation through a spherical coordinate mapping algorithm to obtain an attitude deviation angle.

[0057] The control module is configured to input the attitude deviation angle into a sliding mode controller and generate a three-axis attitude correction torque instruction through a variable structure control algorithm.

[0058] The compensation module is configured to calculate a gravity gradient interference torque and a solar pressure interference torque based on a preset orbit perturbation model, generate a compensation torque after compensation processing, and superimpose the compensation torque on the three-axis attitude correction torque instruction to generate a three-axis synthetic control torque instruction.

[0059] The execution module is configured to input the three-axis synthetic control torque instruction into a reaction flywheel controller, convert the three-axis synthetic control torque instruction into a flywheel motor speed regulation instruction, and drive the reaction flywheel to execute.

[0060] The analysis module is configured to adjust and analyze the continuously updated attitude deviation angle based on the actual attitude change after the driving reaction flywheel is executed, to obtain updated sliding mode controller gain parameters, and feed back the parameters of the updated sliding mode controller;

[0061] The verification module is configured to verify the direction consistency of the newly output sun direction vector measurement value based on the parameters of the updated sliding mode controller, and verify the actual included angle. If the actual included angle is less than a preset precision threshold, a new round of data acquisition period is triggered to realize closed-loop control.

[0062] In the embodiment of the present application, the sun direction vector measurement value and the ephemeris prediction reference value are subjected to vector difference operation by the orbit attitude solver, and the attitude deviation angle is obtained by combining the spherical coordinate mapping algorithm, so that the attitude deviation is accurately quantized, the correction torque is quickly generated in response to the attitude deviation, the system disturbance is effectively suppressed, and the dynamic stability of the attitude control is improved.

[0063] The reaction flywheel controller directly converts the synthesized control torque into a motor speed adjustment instruction, reducing the delay of the intermediate conversion link and improving the execution speed of the control instruction. The link design from measurement, calculation, control to feedback is simple and efficient, reducing the overall complexity.

[0064] In a preferred embodiment of the present application, the sun direction vector measurement value is subjected to vector difference operation with the sun reference direction predicted by ephemeris by the orbit attitude solver, and the vector difference is converted into spherical azimuth angle deviation and elevation angle deviation by the spherical coordinate mapping algorithm to obtain the attitude deviation angle, including:

[0065] In the orbit attitude solver, the sun direction vector measurement value in the satellite body coordinate system and the sun reference direction vector predicted by ephemeris are obtained, the component values of the two on the X, Y and Z axes are extracted, subtraction operation on the corresponding coordinate axes is performed based on the component values to generate the component difference values of each axis, the component difference values are reorganized according to the orthogonal direction of the satellite body coordinate system to generate the sun direction vector difference; based on the satellite body coordinate system, the orbit normal axis is selected as the elevation angle reference axis, and an orbit plane reference surface is established;

[0066] The sun direction vector difference is projected onto the orbit plane reference surface to obtain a plane projection vector; the azimuth angle of the plane projection vector relative to a preset reference direction axis in the orbit plane reference surface is calculated to obtain the spherical azimuth angle deviation; the included angle between the sun direction vector difference and the elevation angle reference axis is calculated to obtain the spherical elevation angle deviation;

[0067] The spherical azimuth angle deviation and the spherical elevation angle deviation are fused to finally obtain the attitude deviation angle.

[0068] In the embodiment of the present application, from the satellite body coordinate system (right-handed coordinate system, X axis along the satellite longitudinal axis forward, Y axis points to the right side of the satellite, Z axis is perpendicular to the orbital plane), read the three components of the sun direction vector measurement value (denoted as A x , A y , A_z) and the sun reference direction vector components predicted by ephemeris (denoted as B x , B y , B_z); the component values are in the range of [-1, 1].

[0069] Perform subtraction operation on the components of the same coordinate axis: X axis difference: Δ x =A x -B x ; Y axis difference: Δ y =A y -B y ; Z axis difference: Δ_z=A_z-B_z; the value range of each difference is [-2, 2] (obtained by subtracting two [-1, 1] components).

[0070] Combine Δ x , Δ y , Δ_z according to the orthogonal relationship (X, Y, Z axes are perpendicular to each other) of the satellite body coordinate system to form the sun direction vector difference Δ (Δ x , Δ y , Δ_z).

[0071] Elevation angle reference axis: select Z axis as the orbit normal axis (perpendicular to the orbital plane) as the reference for the calculation of the elevation angle; orbital plane reference plane: composed of the plane where X axis and Y axis are located (perpendicular to Z axis), as the reference plane for the calculation of the azimuth angle.

[0072] Spherical deviation angle calculation: project the sun direction vector difference Δ onto the orbital plane reference plane (X-Y plane) to obtain the plane projection vector Δ p (Δ x , Δ y , 0); during the projection process, the Z axis component is eliminated, and only the X and Y axis components are retained.

[0073] Spherical azimuth angle deviation calculation: select X axis as the reference direction in the orbital plane (such as the forward direction of the satellite); by measuring the included angle between the plane projection vector Δ p and the X axis, the azimuth angle deviation α is obtained.

[0074] Value range: 0°-360°, wherein 0° indicates that Δ p and the X axis are in the same direction, 90° indicates that it is in the same direction as the Y axis, 180° indicates that it is in the opposite direction of the X axis, and so on.

[0075] Spherical altitude angle deviation calculation: measure the angle between the sun direction vector difference Δ and the altitude angle reference axis (Z axis), get the altitude angle deviation β; value range: -90°~90°, where 0° represents Δ in the orbital plane (perpendicular to the Z axis), positive value represents Δ deviates to the positive direction of the Z axis (above the orbital plane), negative value represents Δ deviates to the negative direction of the Z axis (below the orbital plane).

[0076] Integrate the azimuth angle deviation α (0°~360°) and the altitude angle deviation β (-90°~90°) through spatial geometric relationship: if α is in 0°~180° and β is positive, the attitude deviation is represented as "positive X axis deflection + upward tilt"; if α is in 180°~360° and β is negative, the attitude deviation is represented as "negative X axis deflection + downward tilt".

[0077] Finally output the comprehensive attitude deviation angle, the value range covers the possible attitude deviation interval of the satellite (usually -180°~180°, depending on the satellite attitude control capability).

[0078] Model construction:

[0079] Construct the "sun vector difference-attitude deviation solving model" to realize the conversion from vector difference to spherical angle deviation, output the deviation amount which can be directly used for attitude control; input layer: receive 6 parameters of sun direction vector measurement (A x , A y , A_z) and reference vector (B x , B y , B_z).

[0080] Vector difference calculation module: perform component subtraction and vector reorganization; projection conversion module: project the vector difference to the orbital plane; angle solving module: calculate α and β based on geometric relationship, and integrate them into the final attitude deviation; output the attitude deviation angle (single value or α, β combination).

[0081] In the model, preset the satellite body coordinate system definition, the orbital plane equation (based on the orbital parameters), the reference direction axis pointing (such as the angle between the X axis and the orbital tangent), etc. as the operation reference.

[0082] Model training: through the satellite ground simulation system, generate 100,000 sets of sun vector measurements under different attitudes (including simulated noise), corresponding ephemeris reference vectors, and real attitude deviations measured by high-precision turntable (as labels); add extreme working condition data (such as measurement values when the sun is close to the horizon), expand the coverage range of the data set.

[0083] Training process:

[0084] The input data is substituted into the model to obtain an initial attitude deviation output; the model output is compared with the real label to count the azimuth angle deviation error (Δα) and the elevation angle deviation error (Δβ), and the mean absolute error (MAE) is used as an evaluation index.

[0085] The correction coefficient of the projection conversion module in the model (such as a correction value for compensating for a slight inclination of the orbit plane) is adjusted, so that Δα and Δβ are both less than 0.1° (satisfying the satellite attitude control accuracy requirement).

[0086] Iterative training: repeat the above process 500 times until the error of the model on the validation set (20% of the total data) stabilizes within the target range.

[0087] Introduce measurement noise (such as ±0.05° random error) and ephemeris error (such as ±0.1° orbit parameter deviation) to test the output stability of the model.

[0088] For special scenarios such as the Earth's shadow area and strong sunlight interference, separately verify the adaptability of the model to ensure that the error increase does not exceed 0.2°.

[0089] Model implementation: compile the model into binary code executable by the onboard computer, and store it in the FPGA chip of the satellite attitude control system to ensure that the time consumption of a single operation is ≤10 ms (satisfying the real-time control requirement).

[0090] By component difference reorganization and spherical coordinate mapping, the three-dimensional vector difference is converted into intuitive azimuth angle and elevation angle deviation, avoiding the multi-value problem of complex attitude angle calculation, and the clear definition of the value range ensures the uniqueness of the angle description; the azimuth angle deviation corresponds to the attitude deviation in the orbit plane, and the elevation angle deviation corresponds to the deviation in the orbit normal direction, which is consistent with the actual motion law of the satellite, facilitating the subsequent controller to generate correction torque. The model is based on geometric principles and does not require complex training, and the calculation process is simple. The accurate attitude deviation angle provides reliable input for the sliding mode controller, indirectly improving the accuracy of the entire control system.

[0091] In a preferred embodiment of the present application, the attitude deviation angle is input into the sliding mode controller, and a three-axis attitude correction torque command is generated through a variable structure control algorithm, including:

[0092] Based on the attitude deviation angle and the historically stored attitude deviation angle, the sliding mode controller calculates the rate of change of the attitude deviation angle with time to generate an attitude deviation rate; the attitude deviation angle and the attitude deviation rate are weighted and combined according to the preset proportional coefficient inside the sliding mode controller to generate a sliding mode surface input value; the sliding mode surface input value is input into the sliding mode controller to calculate a sliding mode surface state value.

[0093] An equivalent control component is calculated based on the input value of the sliding surface and a preset equivalent control gain parameter; and a switching control component is calculated based on the sign of the state value of the sliding surface and a preset switching control gain parameter; the equivalent control component and the switching control component are subjected to scalar addition operation, and a sliding mode control signal is generated by fusion;

[0094] Based on the sliding mode control signal and a control distribution matrix pre-stored in the sliding mode controller, a three-axis control torque reference vector is calculated and generated, and the reference vector is distributed to each control axis according to the three orthogonal directions of the roll axis, the pitch axis and the yaw axis of the satellite body coordinate system, so as to generate a three-axis attitude correction torque instruction.

[0095] In the embodiment of the application, the sliding mode controller reads a current attitude deviation angle (denoted as θ t , with a value range of -180°-180°), and calls a historical stored attitude deviation angle at a previous time (θ t-1 ) and a time interval (Δt, usually 0.1-1 second, depending on the satellite control period); a change rate is calculated through the ratio of the difference between the two and the time interval: attitude deviation change rate=(θ t -θ t-1 ) ÷ Δt, with a value range determined by the satellite attitude dynamic characteristics, usually -10° / s-10° / s (a fault protection is triggered when the value exceeds the range).

[0096] Two preset proportional coefficients in the controller: an attitude deviation weight coefficient (k1, with a value of 0.5-2.0, adjusted according to the control accuracy requirement) and a change rate weight coefficient (k2, with a value of 0.1-1.0, affecting the response speed); the attitude deviation angle and the change rate are multiplied by the corresponding coefficients respectively, and then summed, to obtain the input value of the sliding surface: sliding surface input value=k1×θ t +k2×attitude deviation change rate, with a value range of usually -50-50 (unitless, determined by the combination of the coefficients and the angle dimension).

[0097] The input value is substituted into a preset sliding surface function (such as a linear function or a saturation function) in the sliding mode controller, to output a sliding surface state value (s); the value range thereof is associated with the input value, and is usually -100-100, used to represent the deviation degree of the current state of the system from the ideal sliding mode.

[0098] The controller calls a preset equivalent control gain parameter (k e , with a value of 0.1-5.0, set according to the inertia characteristics of the satellite), multiplies the sliding surface input value by the gain, to obtain an equivalent control component: equivalent control component=k e ×sliding surface input value, with a value range of -250-250 (corresponding to the input value and the gain range); the component is used to maintain the movement of the system on the ideal sliding mode.

[0099] The controller first determines the sign of the sliding surface state value (s): if s > 0, the sign is +1; if s < 0, the sign is -1; if s = 0, the sign is 0; then a preset switching control gain parameter (k s , taking a value of 1.0-10.0, greater than the upper limit of system disturbance) is called, and the sign is multiplied by the gain to obtain a switching control component: switching control component = k s x sign, taking a value range of -10-10. The component is used to force the system state to converge to the sliding surface.

[0100] The equivalent control component and the switching control component are subjected to scalar addition operation to obtain a sliding mode control signal (u): u = equivalent control component + switching control component, taking a value range of -260-260; the signal integrates the steady-state control and dynamic adjustment capability, and ensures that the system quickly converges to the target state.

[0101] The sliding mode controller calls an internally pre-stored control distribution matrix (a 3xn matrix, n being the number of actuators, the element value being determined by the actuator installation position and efficiency, taking a value of -1-1), multiplies the sliding mode control signal (u) by the matrix to obtain three-axis control torque reference vectors (T x , T y , T_z), the value range of each component being determined by the maximum torque of the satellite actuator, usually -5-5 N•m (Newton•meter).

[0102] According to the three orthogonal axes (roll axis X, pitch axis Y, yaw axis Z) of the satellite body coordinate system, the three components of the reference vector are directly distributed to the corresponding axes: the X-axis corresponds to the roll axis correction torque (T x ), the Y-axis corresponds to the pitch axis correction torque (T y ), and the Z-axis corresponds to the yaw axis correction torque (T_z). The finally generated three-axis attitude correction torque command needs to satisfy the physical limitations of each axis actuator (such as the minimum output torque ±0.1 N•m), and is automatically truncated to the limit value when exceeding the range.

[0103] The sliding mode control is insensitive to model errors and spatial environmental disturbances, and can still ensure control accuracy even if there is a parameter deviation, and the stability is better than that of traditional PID control. The fusion design of the equivalent control component and the switching control component can realize small steady-state error of attitude deviation angle, and meet the attitude stability of high-resolution remote sensing satellites. The control distribution matrix can flexibly adapt to different actuator configurations of different satellites (such as flywheel and thruster combination), and can be applied to low-orbit, high-orbit and other types of satellites through parameter adjustment, and has strong universality.

[0104] As Figure 2As shown, based on the preset orbit perturbation model, the gravity gradient interference torque and the solar pressure interference torque are calculated, compensation processing is performed to generate a compensation torque, and the compensation torque is superimposed on the three-axis attitude correction torque command to generate a three-axis combined control torque command, including:

[0105] Based on the satellite real-time orbit data and the preset orbit perturbation model, the orbit position parameters of the satellite are obtained; based on the orbit position parameters, the gravity gradient interference torque components in the orbit coordinate system are calculated and converted to the satellite body coordinate system; based on the solar direction vector measurement value, the solar pressure interference torque components in the satellite body coordinate system are calculated;

[0106] The gravity gradient interference torque components and the solar pressure interference torque components are subjected to vector addition operation in the satellite body coordinate system to generate a total interference torque vector;

[0107] The total interference torque vector is subjected to a negative operation to generate a compensation torque vector; the three-axis attitude correction torque command and the compensation torque vector are subjected to vector addition operation in the satellite body coordinate system to generate a three-axis combined control torque command.

[0108] In the embodiment of the application, based on the orbit data (such as orbit semi-major axis, eccentricity, orbit inclination, etc.) returned by the satellite in real time and the preset orbit perturbation model (including non-spherical items of the earth's gravitational field, lunar and solar gravitational perturbation factors), the position parameters of the satellite in the orbit coordinate system are calculated, including the geocentric distance (the distance from the satellite to the center of the earth, the value range is about 400-40000 km, which depends on the type of orbit), the direction angle of the orbit radius vector (the value is 0°-360°), etc.

[0109] According to the orbit position parameters and the inertia tensor parameters (reflecting the mass distribution characteristics of the satellite, preset in the system) of the satellite, the three components (X_g, Y_g, Z_g) of the gravity gradient interference torque in the orbit coordinate system are calculated, the value range is usually-0.01-0.01 N•m (small satellite) to-1-1 N•m (large satellite), which is positively related to the mass and size of the satellite. or b、Y or b、Z or b).

[0110] Through the conversion matrix (calculated from the satellite attitude angle) of the orbit coordinate system and the satellite body coordinate system, the components in the orbit coordinate system are converted to the gravity gradient interference torque components (X_g, Y_g, Z_g) in the satellite body coordinate system, and the value range remains unchanged.

[0111] Based on the solar direction vector measurement, satellite surface area parameters (preset, such as sailboard area, star body projection area) and surface optical properties (such as reflectivity, value 0-1), the solar pressure interference torque component (X_s, Y_s, Z_s) in the satellite body coordinate system is calculated; its value range is -0.001-0.001N•m (small satellite) to -0.1-0.1N•m (satellite with large sailboard), which is significantly affected by the distance to the sun and the satellite attitude (such as the maximum value when the sailboard faces the sun).

[0112] In the satellite body coordinate system, the gravity gradient interference torque component and the solar pressure interference torque component are subjected to vector addition operation: X-axis total interference torque: T_dx=X_g+X_s; Y-axis total interference torque: T_dy=Y_g+Y_s; Z-axis total interference torque: T_dz=Z_g+Z_s.

[0113] The total interference torque vector (T_dx, T_dy, T_dz) is generated, the value range of which is the superposition range of the sum of the components of each axis, usually -1.1-1.1N•m (the maximum possible value of the two interferences); the vector reflects the total interference effect of the space environment on the satellite attitude.

[0114] The total interference torque vector is subjected to negative operation to obtain the compensation torque vector (T_cx, T_cy, T_cz), wherein: T_cx=-T_dx; T_cy=-T_dy; T_cz=-T_dz; the value range thereof is symmetrical to the total interference torque vector, -1.1-1.1N•m, which is used to offset the interference effect of the space environment.

[0115] In the satellite body coordinate system, the generated three-axis attitude correction torque command (T_rx, T_ry, T_rz, value range -5-5N•m) is subjected to vector addition operation with the compensation torque vector:

[0116] X-axis combined torque: T_total_x=T_rx+T_cx;

[0117] Y-axis combined torque: T_total_y=T_ry+T_cy;

[0118] Z-axis combined torque: T_total_z=T_rz+T_cz;

[0119] The finally generated three-axis combined control torque command needs to meet the physical limitations of the actuator (such as maximum output torque ±5N•m), and when it exceeds the range, it is automatically truncated to the limit value, ensuring that the command is executable.

[0120] By compensating the main space interference torques such as gravity gradient and solar radiation pressure, the satellite attitude stabilization can be improved to meet the high-precision task requirements such as high-resolution imaging and laser communication. The environmental interference torques are counteracted in advance, so that the attitude correction torque does not need to resist the interference, which can reduce the output power of the actuator (such as reaction flywheel) and prolong its on-orbit life. The attitude drift caused by the interference torque is eliminated, and the system is prevented from shaking due to continuous resistance to interference, so that the attitude adjustment process is smoother. The compensation amount is automatically adjusted according to different orbital positions (such as perigee or apogee) and solar illumination conditions, so that the satellite can maintain stable control performance in the whole orbital period, which is especially suitable for elliptical orbit or sun-synchronous orbit satellites.

[0121] In a preferred embodiment of the present application, the three-axis synthesized control torque command is input into the reaction flywheel controller, converted into a flywheel motor speed regulation command and driven to execute the reaction flywheel, comprising:

[0122] The three-axis synthesized control torque command is decomposed and processed to generate target control torque components of the roll, pitch and yaw axes respectively; based on the target control torque components of each axis, the target angular acceleration of each axis is calculated; and based on the preset flywheel configuration information, the axis responsible for each flywheel and its allocation ratio are determined;

[0123] Based on the target angular acceleration of each control axis, the flywheel configuration and the allocation ratio, the target torque required by each reaction flywheel is calculated; based on the target torque and the moment of inertia parameters of the flywheel, the target speed change amount required by the flywheel motor is calculated;

[0124] Based on the target speed change amount, the current real-time speed feedback value of each flywheel motor obtained in real time from the flywheel motor is combined to calculate and generate the flywheel motor speed regulation command value; the flywheel motor speed regulation command value is executed through the driving reaction flywheel to change the speed.

[0125] In the embodiment of the present application, the three-axis synthesized control torque command (T_total_x, T_total_y, T_total_z) is directly decomposed according to the three orthogonal axes of the satellite body coordinate system to obtain independent target control torque components of each axis:

[0126] The target control torque of the roll axis (X axis): T_x (value range -5~5N•m, consistent with the range of synthesized torque); the target control torque of the pitch axis (Y axis): T_y (value range -5~5N•m); the target control torque of the yaw axis (Z axis): T_z (value range -5~5N•m).

[0127] The moment of inertia parameters of each axis of the satellite (preset in the system, X axis moment of inertia J_x, Y axis J_y, Z axis J_z, unit: kg•m 2, the value range is usually 10~1000 kg•m 2 Target angular acceleration is calculated by the ratio of target control torque to moment of inertia:

[0128] Rolling axis target angular acceleration: α_x = T_x / J_x (value range -0.5~0.5 rad / s 2 , i.e. -28.6° / s 2 ~28.6° / s 2 ); pitch axis target angular acceleration: α_y = T_y / J_y (value range same as above); yaw axis target angular acceleration: α_z = T_z / J_z (value range same as above).

[0129] Based on the preset flywheel configuration information (such as the common 4-flywheel orthogonal + diagonal layout), the physical installation angle of each flywheel and the control axis responsible for are determined:

[0130] For example, flywheel 1 is mainly responsible for the rolling axis (allocation ratio 70%) and the yaw axis (30%), flywheel 2 is mainly responsible for the pitch axis (80%) and the rolling axis (20%), etc., and the sum of the allocation ratios of each axis is 100% (value 0~1).

[0131] The allocation ratio is set according to the moment output efficiency of the flywheel installation position, and the diagonal flywheel can participate in multi-axis control at the same time through vector decomposition.

[0132] According to the target angular acceleration of each control axis, flywheel configuration and allocation ratio, the shaft torque is distributed to each flywheel:

[0133] For example, the target torque of flywheel 1 = α_x × J_x × 70% + α_z × J_z × 30% (the product of angular acceleration and moment of inertia is used to restore the torque dimension); the target torque of a single flywheel is in the range of -3~3 N•m (determined by total torque distribution and maximum output capacity of the flywheel).

[0134] The moment of inertia parameter (J_f, value range 0.1~10 kg•m 2 , which is positively correlated with the size of the flywheel) of each flywheel is called, and the angular acceleration change rate is calculated by the ratio of the target torque to the flywheel moment of inertia, and then combined with the control period (Δt = 0.1~1 s) to obtain the target speed change amount: target speed change amount = (flywheel target torque / J_f) × Δt (value range -300~300 r / min, i.e. speed change per minute, limited by motor power).

[0135] The real-time speed feedback value (N_current, value range -6000~6000 r / min, negative sign indicating reverse rotation) of each flywheel is obtained from the speed sensor of the flywheel motor, combined with the target speed change amount, to calculate the speed command value that needs to be adjusted:

[0136] The rotational speed regulation instruction value = N_current + target rotational speed variation; the instruction value needs to be limited in the rated rotational speed range of the flywheel (such as -6000-6000 r / min), and when exceeding, it is automatically truncated to the maximum value or the minimum value.

[0137] The rotational speed regulation instruction value is converted into a motor driving signal (such as a pulse width modulation signal PWM), the motor rotating the reaction flywheel is driven through a power amplification circuit, and the actual rotational speed of the flywheel gradually approaches the instruction value. In the execution process, the rotational speed sensor feeds back the adjusted rotational speed in real time, forms a closed-loop control, and until the rotational speed error ≤5 r / min (satisfying the control precision requirement).

[0138] Through rotational inertia matching and proportional distribution, the synthesized torque is accurately converted into flywheel action, ensuring the attitude control precision. The oblique flywheel layout and multi-axis distribution strategy are adopted, and a single flywheel can participate in the torque output of multiple axes at the same time, reducing the number of actuators, reducing the system weight and power consumption. The response time from the control torque instruction to the flywheel rotational speed adjustment is shortened, the space disturbance torque can be quickly offset, and the satellite attitude can be restored to stability in a short time. Through rotational speed range limitation and closed-loop feedback, the flywheel overspeed is avoided, the load of each flywheel can be balanced through torque distribution, and the service life is prolonged.

[0139] In a preferred embodiment of the present application, based on the actual attitude change after the reaction flywheel is executed, the continuously updated attitude deviation angle is adjusted and analyzed to obtain the updated sliding mode controller gain parameter, and the parameter of the updated sliding mode controller is fed back, including:

[0140] Based on the actual attitude change of the satellite after the reaction flywheel is executed, the updated angular velocity measurement value and the sun direction vector measurement value in the satellite body coordinate system are continuously obtained; the updated sun direction vector measurement value is input into the orbit attitude solver to generate the updated attitude deviation angle;

[0141] Based on the updated attitude deviation angle, the historical stored attitude deviation angle data is combined to establish time series data; based on the time series data, the statistical characteristics of the attitude deviation angle in the preset time window are calculated;

[0142] Based on the statistical characteristics of the attitude deviation angle, the preset sliding mode controller gain adjustment strategy rule library is analyzed; according to the analysis result, the gain adjustment factor of the equivalent control gain parameter and the switching control gain parameter of the sliding mode controller is generated;

[0143] The gain adjustment factor is multiplied by the equivalent control gain parameter and the switching control gain parameter stored in the sliding mode controller respectively, and the updated equivalent control gain parameter value and the updated switching control gain parameter value are calculated.

[0144] The equivalent control gain parameter value and the switching control gain parameter value are fed back to the internal parameter memory of the sliding mode controller to complete the online update of the sliding mode controller parameters.

[0145] After the reaction flywheel is driven to execute, the angular velocity sensor (such as a gyroscope) of the satellite continuously obtains updated angular velocity measurement values (ω x , ω y , ω_z) in the satellite body coordinate system, with a value range of -0.1-0.1 rad / s, i.e., -5.7° / s-5.7° / s, and simultaneously, the sun sensor obtains updated sun direction vector measurement values (A' x , A' y , A'_z) after normalization, with a value range of [-1, 1].

[0146] The updated sun direction vector measurement values are input into the orbit attitude solver to repeatedly calculate the logic (vector difference operation, spherical coordinate mapping, etc.) to generate updated attitude deviation angles (θ' t with a value range of -180°-180°, reflecting the actual attitude correction effect after the flywheel is executed.

[0147] The historically stored attitude deviation angle data (θ t-1 , θ t-2 ,..., θ t-100 of the past 100 control periods, each of which is 0.1-1 second, are recalled and combined with the updated θ' t to form a time sequence (length 101 points) containing the latest data.

[0148] The time window is set to the last 20 control periods (about 2-20 seconds, adjusted according to the control accuracy requirement), and the statistical characteristics of the attitude deviation angles in the window are calculated: the deviation mean value reflects the system steady-state deviation, with a value range of -1°-1°; the deviation maximum value reflects the dynamic overshoot, with a value range of 0-5°; the standard deviation of the deviation change rate reflects the attitude fluctuation degree, with a value range of 0-0.5° / s; the convergence time is the time from the peak value of the deviation to ≤0.1°, with a value range of 0-30 seconds.

[0149] The preset sliding mode controller gain adjustment strategy rule library is called, which contains multiple sets of "statistical characteristics-adjustment strategy" mapping relationships, for example:

[0150] If the deviation mean value is >0.5°, it indicates that the equivalent control is insufficient, and the equivalent control gain needs to be increased; if the deviation maximum value is >2° and the convergence time is >15 seconds, it indicates that the switching control is too weak, and the switching control gain needs to be increased; if the standard deviation of the deviation change rate is >0.3° / s, it indicates that the system oscillation is too strong, and the switching control gain needs to be reduced.

[0151] According to the matching results of statistical characteristics and rule base, a gain adjustment factor is generated:

[0152] Equivalent control gain adjustment factor (k ea , value range 0.9~1.1): 1.1 represents an increase of 10%, 0.9 represents a decrease of 10%; Switching control gain adjustment factor (k sa , value range 0.8~1.2): The adjustment amplitude is slightly larger than the equivalent control, in order to quickly suppress the oscillation or accelerate the convergence.

[0153] The equivalent control gain parameter (k e _current, value 0.1~5.0) currently stored by the sliding mode controller is retrieved and multiplied by the equivalent control gain adjustment factor:

[0154] The updated equivalent control gain k e _new=k e _current×k ea ; The result needs to be limited within the preset range (0.1~5.0), and the boundary value is taken when it is exceeded.

[0155] The current switching control gain parameter (k s _current, value 1.0~10.0) is retrieved and multiplied by the switching control gain adjustment factor: The updated switching control gain k s _new=k s _current×k sa ; The result needs to be limited within the preset range (1.0~10.0), to ensure that the gain will not be too small to cause the anti-interference ability to decrease, or too large to cause system oscillation.

[0156] Write the calculated k e _new and k s _new to the internal parameter storage (such as non-volatile memory) of the sliding mode controller, overwrite the original parameter value; The update process adopts "double buffering" mechanism: The new parameter takes effect after the current control period ends, to avoid affecting the ongoing control operation during the update process; The parameter update frequency is consistent with the control period (0.1~1 second / time), to realize real-time adaptive adjustment of the sliding mode controller.

[0157] The steady-state error of the attitude deviation can be further reduced from ±0.05° to ±0.02° by adjusting the gain parameters through continuous feedback, so as to adapt to the performance drift caused by the change of mass distribution (such as fuel consumption) or the aging of equipment during the on-orbit operation of the satellite. In different orbit stages (such as the light area or the shadow area) or task modes (such as imaging / maneuvering), the gain is automatically adjusted to balance the response speed and stability, for example, the switching gain is reduced to reduce the oscillation during imaging, and the equivalent gain is increased to accelerate the convergence during maneuvering. When the space disturbance torque suddenly changes (such as the light angle of the sailboard suddenly changes), the disturbance influence can be suppressed within 3-5 control periods by rapidly increasing the switching control gain, so as to avoid attitude instability. By optimizing the gain parameters, unnecessary torque output is reduced, the mechanical loss and energy consumption of the reaction flywheel are reduced, and the on-orbit life of the actuator is indirectly prolonged.

[0158] In a preferred embodiment of the present application, based on the updated parameters of the sliding mode controller, the newly output sun direction vector measurement is verified for direction consistency, and the actual included angle is verified. If the actual included angle is less than the preset precision threshold, a new data acquisition cycle is triggered to realize closed-loop control, including:

[0159] Based on the updated parameters of the sliding mode controller, a complete control chain is executed to drive the satellite attitude adjustment, and a newly output sun direction vector measurement under the control of the new parameters is obtained.

[0160] The newly output sun direction vector measurement is dot multiplied with the sun reference direction vector predicted by the ephemeris to calculate the actual directional included angle therebetween. The actual directional included angle is compared with the preset threshold to determine whether the actual directional included angle is less than the threshold.

[0161] According to the determination result, if the directional included angle is less than the threshold, a new data acquisition cycle is triggered; and if the directional included angle is not less than the threshold, the control flow is maintained to continue the control and verification.

[0162] The updated parameters (the equivalent control gain k e _new, the switching control gain k s _new) of the sliding mode controller are used to re-execute a complete control flow, from sun direction vector measurement, attitude deviation calculation, sliding mode control torque generation, to disturbance torque compensation, flywheel driving and parameter updating, to finally complete the satellite attitude adjustment.

[0163] Under the control of the new parameters, the adjusted sun direction vector measurement (denoted as A'' x , A'' y , A''_z) is collected by the sun sensor. The vector has been normalized, and each component has a value range of [-1, 1], reflecting the sun pointing state after the satellite attitude adjustment.

[0164] Acquire the sun reference direction vector (B x , B y , B_z, component value [-1, 1]) of ephemeris prediction; Perform dot product operation on the newly output sun direction vector measurement (A'' x , A'' y , A''_z) and the reference vector to obtain a scalar value reflecting the direction consistency (value range [-1, 1], 1 indicates complete direction consistency, -1 indicates complete opposite direction); Calculate the actual direction angle (θ_actual) of the two based on the dot product result, value range 0°-180°: The closer the dot product is to 1, the smaller the angle (such as dot product = 0.9998 corresponding to angle ≈1°); Dot product = 0 corresponds to angle 90°.

[0165] The preset precision threshold (θ_threshold) is set according to the satellite task requirement, usually 0.1°-0.5° (such as 0.1° for high-resolution remote sensing satellites, and 0.5° for general communication satellites); If the actual direction angle θ_actual<θ_threshold, it means that the current attitude precision meets the requirement; If θ_actual≥θ_threshold, it means that the attitude still needs to be further adjusted.

[0166] When θ_actual<θ_threshold, the system automatically triggers a new round of data acquisition period: the sun sensor reacquires the sun direction vector (as the initial measurement value for the next round of control), the ephemeris module updates the reference vector, and the attitude solver, sliding mode controller, etc. in turn enter the next round of operation to form a closed loop control cycle; The data acquisition period is consistent with the control period (0.1-1 seconds to ensure real-time performance).

[0167] When θ_actual≥θ_threshold, the system does not trigger a new period and continues to maintain the current control process: the sliding mode controller continuously outputs the control torque based on the existing parameters, the flywheel maintains the speed adjustment, and the sun sensor still acquires data at the original frequency (updated every 0.1-1 seconds), until the next angle verification meets θ_actual<θ_threshold and enters a new period.

[0168] The direction angle verification ensures that the attitude control always converges within the preset precision threshold, meeting the core task requirements of the satellite energy system (such as sailboard sun tracking), attitude reference establishment, etc. Only when the attitude meets the precision, a new period is triggered to avoid unnecessary high-frequency adjustment, reducing the on-board computer operation load and flywheel energy consumption. Through the "measurement-verification-trigger" mechanism, negative feedback is formed to effectively suppress cumulative errors (such as sensor drift and control delay caused by deviation), so that the satellite attitude remains stable in long-term operation. The preset threshold can be dynamically adjusted according to the task stage, taking into account the precision and efficiency requirements in different scenarios.

[0169] Embodiments of the present application also provide a computing device, comprising a processor, a memory storing a computer program, the computer program being executed by the processor to implement the system as described above. All implementation manners in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0170] Embodiments of the present application also provide a computer readable storage medium storing instructions, which, when executed on a computer, cause the computer to implement the system as described above. All implementation manners in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0171] The above describes the preferred embodiments of the present application. It should be noted that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. An attitude control system for shortwave satellites, characterized in that, include: The acquisition module is used to obtain angular velocity measurements and solar direction vector measurements in the satellite's body coordinate system. The calculation module is used to perform vector difference calculations between the measured solar direction vector value and the solar reference direction predicted by the ephemeris, through the orbital attitude solver. The vector difference is then converted into spherical azimuth and elevation angle deviations using a spherical coordinate mapping algorithm to obtain the attitude deviation angles. This includes: in the orbital attitude solver, obtaining the measured solar direction vector value and the solar reference direction vector predicted by the ephemeris in the satellite body coordinate system; extracting the component values ​​of both on the X, Y, and Z axes; performing subtraction operations on the corresponding coordinate axes based on the component values ​​to generate the component difference values ​​for each axis; and then calculating the difference values ​​according to the satellite body coordinate system. The difference in the recombined components of the intersection direction is used to generate the solar direction vector difference. Based on the satellite body coordinate system, the orbital normal axis is selected as the elevation angle reference axis, and an orbital plane reference plane is established. The solar direction vector difference is projected onto the orbital plane reference plane to obtain the planar projection vector. The azimuth angle of the planar projection vector relative to the preset reference direction axis in the orbital plane reference plane is calculated to obtain the spherical azimuth angle deviation. The angle between the solar direction vector difference and the elevation angle reference axis is calculated to obtain the spherical elevation angle deviation. The spherical azimuth angle deviation and the spherical elevation angle deviation are fused to finally obtain the attitude deviation angle. The control module is used to input the attitude deviation angle into the sliding mode controller and generate a three-axis attitude correction torque command through a variable structure control algorithm. The compensation module is used to calculate the gravity gradient disturbance torque and the solar radiation pressure disturbance torque based on the preset orbital perturbation model, perform compensation processing to generate a compensation torque, and then simultaneously superimpose the compensation torque onto the three-axis attitude correction torque command to generate a three-axis synthetic control torque command. The execution module is used to input the three-axis composite control torque command into the reaction flywheel controller, convert it into a flywheel motor speed adjustment command, and drive the reaction flywheel to execute it; The analysis module is used to adjust and analyze the continuously updated attitude deviation angle based on the actual attitude change after the drive reaction flywheel is executed, to obtain the updated sliding mode controller gain parameters, and to feed back the updated sliding mode controller parameters. The verification module is used to control based on the updated parameters of the sliding mode controller, verify the direction consistency of the newly output solar direction vector measurement value, and verify the actual angle. If the actual angle is less than the preset accuracy threshold, a new data acquisition cycle is triggered to achieve closed-loop control.

2. The attitude control system for shortwave satellites according to claim 1, characterized in that, The attitude deviation angle is input into the sliding mode controller, and a three-axis attitude correction torque command is generated through a variable structure control algorithm, including: Based on the attitude deviation angle and the historically stored attitude deviation angle, the sliding mode controller calculates the rate of change of the attitude deviation angle over time and generates the attitude deviation rate of change. The attitude deviation angle and the rate of change of attitude deviation are weighted and combined according to the preset proportional coefficient inside the sliding mode controller to generate the sliding surface input value. The sliding surface input value is then input into the sliding mode controller to calculate the sliding surface state value. The equivalent control component is calculated based on the sliding surface input value and the preset equivalent control gain parameter; the switching control component is calculated based on the sign of the sliding surface state value and the preset switching control gain parameter; the equivalent control component and the switching control component are scalarly added and fused to generate the sliding control signal. Based on the sliding mode control signal and the control allocation matrix pre-stored inside the sliding mode controller, a three-axis control torque reference vector is calculated and generated. According to the three orthogonal directions of the satellite body coordinate system, namely the roll axis, pitch axis and yaw axis, the reference vector is allocated to each control axis to generate a three-axis attitude correction torque command.

3. The attitude control system for shortwave satellites according to claim 2, characterized in that, Based on a pre-defined orbital perturbation model, the gravity gradient disturbance torque and solar radiation pressure disturbance torque are calculated, compensated, and then a compensation torque is generated. This compensation torque is then synchronously superimposed on the three-axis attitude correction torque command to generate a three-axis composite control torque command, including: Based on real-time satellite orbit data and a pre-set orbit perturbation model, the satellite's orbital position parameters are obtained; the gravity gradient disturbance torque components in the orbital coordinate system are calculated based on the orbital position parameters and transformed to the satellite body coordinate system; the solar radiation pressure disturbance torque components in the satellite body coordinate system are calculated based on the solar direction vector measurement values. The gravity gradient disturbance torque component and the solar radiation pressure disturbance torque component are vector-added in the satellite body coordinate system to generate the total disturbance torque vector. The total disturbance torque vector is negativeed to generate the compensation torque vector; the three-axis attitude correction torque command and the compensation torque vector are vector-added in the satellite body coordinate system to generate the three-axis composite control torque command.

4. The attitude control system for shortwave satellites according to claim 3, characterized in that, The three-axis composite control torque command is input to the reaction flywheel controller, converted into a flywheel motor speed regulation command, and driven to execute by the reaction flywheel, including: The three-axis composite control torque command is decomposed to generate target control torque components for each of the three axes: roll, pitch, and yaw. Based on the target control torque components of each axis and the moment of inertia parameters, the target angular acceleration of each axis is calculated. Based on the preset flywheel configuration information, the axis to which each flywheel is responsible and its allocation ratio are determined. Based on the target angular acceleration of each control axis, the flywheel configuration and distribution ratio, the target torque required to be generated by each reaction flywheel is calculated; based on the target torque and the rotational inertia parameters of the flywheel, the target speed change required by the flywheel motor is calculated. Based on the target speed change, and combined with the real-time speed feedback value of each flywheel motor obtained from the flywheel motor in real time, the speed adjustment command value of each flywheel motor is calculated and generated; the speed adjustment command value of each flywheel motor is then used to drive the reaction flywheel to change the speed.

5. The attitude control system for shortwave satellites according to claim 4, characterized in that, Based on the actual attitude changes after the drive reaction flywheel is executed, the continuously updated attitude deviation angle is adjusted and analyzed to obtain the updated sliding mode controller gain parameters. These parameters are then fed back to update the sliding mode controller's parameters, including: Based on the actual attitude change of the satellite after the execution of the driving reaction flywheel, the angular velocity measurement value and solar direction vector measurement value in the satellite body coordinate system are continuously updated; the updated solar direction vector measurement value is input into the orbital attitude solver to generate the updated attitude deviation angle. Based on the updated attitude deviation angles and combined with historically stored attitude deviation angle data, time series data is established; based on the time series data, the statistical characteristics of the attitude deviation angles within a preset time window are calculated. Based on the statistical characteristics of the attitude deviation angle, an analysis is performed using a pre-defined sliding mode controller gain adjustment strategy rule base. Based on the analysis results, gain adjustment factors are generated for the equivalent control gain parameter and the switching control gain parameter of the sliding mode controller. The gain adjustment factor is multiplied by the equivalent control gain parameter and the switching control gain parameter stored in the sliding mode controller, respectively, to calculate the updated equivalent control gain parameter value and the updated switching control gain parameter value. The equivalent control gain parameter value and the switching control gain parameter value are fed back to the internal parameter memory of the sliding mode controller to complete the online update of the sliding mode controller parameters.

6. The attitude control system for shortwave satellites according to claim 5, characterized in that, Control is performed based on the updated parameters of the sliding mode controller, and the newly output solar direction vector measurement value is used to verify the direction consistency and verify the actual angle. If the actual included angle is less than the preset accuracy threshold, a new data acquisition cycle is triggered to achieve closed-loop control, including: Based on the updated sliding mode controller parameters, the complete control chain is executed to drive the satellite attitude adjustment and obtain the new output solar direction vector measurement value under the control of the new parameters; The newly output solar direction vector measurement value is multiplied by the solar reference direction vector predicted by the ephemeris to calculate the actual direction angle between the two; the actual direction angle is compared with a preset threshold to determine whether the actual direction angle is less than the threshold. Based on the judgment result, if the directional angle is less than the threshold, a new round of data acquisition cycle is triggered; if it is not less than the threshold, the control process is maintained, and control and verification continue.

7. The attitude control system for shortwave satellites according to claim 6, characterized in that, Based on real-time satellite orbit data and a pre-set orbit perturbation model, the satellite's orbital position parameters are obtained; based on the orbital position parameters, the gravity gradient disturbance torque components in the orbital coordinate system are calculated and transformed to the satellite body coordinate system; The components of the solar radiation pressure interference torque in the satellite body coordinate system are calculated based on the solar direction vector measurement values, including: Based on the orbital coordinate system position coordinates extracted from real-time satellite orbit data, the geocentric modulus scalar value is calculated, and the gravitational field strength scalar value is generated by combining it with the pre-stored Earth gravitational constant. The three-axis gravity gradient torque components are generated by calling the main diagonal element value of the satellite's moment of inertia. Based on the three-axis gravity gradient torque components, a rotation transformation matrix is ​​constructed using the satellite's real-time attitude quaternion. The vector coordinate system transformation operation is then performed through the matrix to obtain the three-axis gravity gradient disturbance torque component values. Based on the unit vector of the solar incident direction, the effective projected area and reflection coefficient are retrieved from the satellite surface optical parameter database. The scalar value of the solar radiation pressure is calculated by combining the solar radiation pressure intensity constant. The solar radiation pressure interference torque component is generated by the torque fusion operation between the scalar value of the solar radiation pressure and the pre-calibrated position deviation vector.

8. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the system as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the system as described in any one of claims 1 to 7.

Images