Satellite antenna dynamic pointing calibration method and system based on satellite autonomous gradient optimization

Abstract

The invention discloses a satellite antenna dynamic pointing calibration method and system based on satellite autonomous gradient optimization, relates to the technical field of satellite communication, and realizes closed-loop calibration without ground intervention by processing signal to noise ratio (SNR) data uploaded by a ground station on a satellite in real time and combining a multi-physical field coupling model of a temperature and stress sensor. And the pointing error in a dynamic environment is reduced from more than 2.0 degrees of a traditional scheme to less than 0.3 degrees.

Description

Technical Field

[0001] This invention relates to the field of satellite communications technology, specifically to a method and system for dynamic pointing calibration of satellite antennas based on onboard autonomous gradient optimization. Key application scenarios include continuous beam alignment for geosynchronous orbit (GEO) communications satellites, enhanced stability of data transmission links from low-Earth orbit (LEO) remote sensing satellites, and autonomous pointing maintenance for deep space probes in long-latency communication environments. In the military, this method can enhance anti-interference communication capabilities. In civilian applications, it is suitable for multi-spot beam scheduling for high-throughput satellites (HTS). It is particularly well-suited for compensating for pointing deviations caused by periodic thermal deformation (typically up to ±80°C) of satellite platforms due to the solar cycle. Background Art

[0002] Existing satellite antenna pointing calibration technology has systemic shortcomings, primarily manifested in the contradiction between static calibration dependence and dynamic environmental mismatch. Taking the Extended Kalman Filter (EKF) solution used by Milstar satellites as an example, its calibration process relies on three dedicated ground stations distributed at different longitudes (such as those in California and Ohio). It collects 136 sets of power difference data over 48 hours using a dual concentric circle scanning pattern, and uses Fourier harmonic decomposition to establish a periodic error model. While this solution can reduce the static error to 0.03°, it suffers from three fundamental flaws:

[0003] First, the hardware cost is high. It requires the deployment of high-precision power measurement terminals (costing over $500,000 per station) and the use of optical auxiliary equipment such as prisms, which increases system complexity by more than 40%.

[0004] Secondly, the dynamic adaptability is insufficient. Its Fourier model can only compensate for regular thermal deformation within a 24-hour period (achieved through 21 harmonic coefficients) and cannot handle sudden mechanical stress deformation (such as the impact caused by the deployment of solar panels). strain);

[0005] Third, there are real-time limitations. The generation of calibration instructions is completely dependent on ground station calculations, and it fails in scenarios such as deep space exploration where communication delays exceed 20 minutes.

[0006] In addition, the traditional scheme adopts the double-circle scanning power difference method ( ,in, is the power difference, the core measurement quantity in the traditional two-circle scanning scheme; is the measured value of the antenna's transmission power on the outer scanning circular trajectory; The measured value of the antenna's transmission power on the inner scanning circle trajectory) can eliminate the common mode interference of atmospheric attenuation, but it requires strict synchronization of data collection and the synchronization accuracy of the ground station clock must reach the microsecond level (error tolerance ), in actual operation, ionospheric disturbances often lead to The signal-to-noise ratio deteriorates by more than 6dB. More seriously, existing technologies lack onboard autonomous decision-making capabilities. Each calibration requires re-uploading the entire set of Fourier coefficients (approximately 21 × 3 = 63 parameters), occupying a data transmission bandwidth of 150kbps. This makes real-time updates difficult during the short LEO satellite transit window (typically less than 10 minutes). Summary of the Invention

[0007] To address the degradation of pointing accuracy caused by thermal deformation, mechanical stress, and attitude drift during on-orbit operation of satellite-borne parabolic antennas, this paper presents an autonomous, highly dynamically adaptable satellite antenna dynamic pointing calibration method and system based on onboard autonomous gradient optimization. By processing signal-to-noise ratio (SNR) data uploaded by ground stations in real time onboard the satellite and integrating it with a multi-physics coupling model of temperature and stress sensors, a closed-loop calibration process is achieved without ground intervention, reducing pointing errors in dynamic environments from over 2.0° in traditional solutions to less than 0.3°.

[0008] The technical solution of the present invention is: a satellite antenna dynamic pointing calibration method based on on-board autonomous gradient optimization, the specific steps are as follows:

[0009] Step 1: Data Collection

[0010] (11) During the satellite's in-orbit operation, the ground station continuously receives satellite signals and records the received signal quality parameters at each transit, including but not limited to the signal-to-noise ratio (SNR). The ground station transmits the collected SNR information to the satellite via an uplink.

[0011] (12) While receiving the signal-to-noise ratio information, the satellite records the antenna stress and temperature information through the mechanical and thermal sensors at its own antenna;

[0012] Step 2: Error analysis

[0013] (21) Using the signal quality parameters received by the ground station and the satellite attitude and orbit information, the antenna pointing error model is established:

[0014]

[0015] in, is the current signal-to-noise ratio measurement value; is the theoretical maximum signal-to-noise ratio, i.e., the optimal value when the antenna is aligned; is the pitch angle pointing error; is the yaw angle pointing error; is the comprehensive pointing error angle; is the pattern attenuation function;

[0016] (22) The satellite actively applies a small perturbation once in each of the four orthogonal directions along the antenna pointing axis, in the pitch and azimuth planes; based on the nonlinear mapping model Calculate the gradient vector of the current pointing deviation ;in, is the signal-to-noise ratio gradient vector; is the partial derivative of the signal-to-noise ratio with respect to the pitch angle; is the partial derivative of the signal-to-noise ratio with respect to the yaw angle;

[0017] This is achieved using the finite difference method:

[0018]

[0019] in, For the current moment The signal-to-noise ratio measurement value; The last measurement time signal-to-noise ratio; is the time interval; is the pitch angle disturbance actively applied;

[0020] Introducing a sliding time window mechanism Dynamically adjust to ,in, Is the basic sampling interval, in seconds , determined according to the satellite attitude control period; is the window width adjustment coefficient, unit ,control Sensitivity to gradient changes, which automatically shortens the sampling interval when the gradient is large to increase the convergence speed;

[0021] (23) By analyzing the relationship between signal quality parameters and satellite attitude and antenna thermal parameters, the magnitude and direction of the current antenna pointing error can be determined;

[0022] (24) When the satellite actively performs attitude disturbances, if the received signal quality parameters show a significant decrease, it indicates that the antenna pointing direction may have deviated from the direction of the ground station;

[0023] Step 3: Error Correction

[0024] (31) Based on the error analysis results, the antenna pointing error is gradually reduced by adjusting the satellite attitude, the angle of the antenna rotating arm, and the feed electronic scanning angle;

[0025] (32) After each adjustment, the ground station receiving signal quality parameters are collected again to evaluate the correction effect;

[0026] (33) Through iterative adjustment, the antenna pointing error is finally converged to the allowable range, achieving high-precision pointing;

[0027] Step 4: Calibration completed

[0028] (41) When the received signal quality parameters reach the expected indicators and stabilize at a high level, the antenna pointing calibration is considered completed;

[0029] (42) Record the final satellite attitude and antenna rotation arm angle parameters as a reference for subsequent antenna pointing control.

[0030] Furthermore, in step 3, error correction, adaptive gradient descent correction is adopted, and the iterative formula is designed as follows:

[0031]

[0032]

[0033] in, For the The pitch angle correction amount for the iteration; For the The yaw angle correction amount for the iteration; is the learning rate coefficient, ranging from 0.1 to 0.3; is the attenuation coefficient;

[0034] Exponential decay term The introduction of It increases and adaptively reduces, which not only ensures that the optimal solution is quickly approached in the early stage, but also avoids oscillation near the extreme point in the later stage.

[0035] Furthermore, in step 3, error correction, dynamic error compensation is used to fuse temperature sensor data in real time. and strain gauge output , establish a physically driven compensation:

[0036]

[0037] in, is the structural characteristic length, 、 are the elastic modulus and shear modulus, respectively. 、 are the thermal expansion coefficient and strain sensitivity coefficient of the material, is the total number of temperature sensors, is the number of measuring points of the strain gauge, For the The equivalent action area of ​​each strain point.

[0038] Furthermore, in adaptive gradient descent correction, the gradient vector is updated at a frequency of 10 Hz; in dynamic error compensation, the physical field parameters are refreshed at a frequency of 1 Hz; and a dual-rate mechanism is used to balance the computational load and response speed.

[0039] Furthermore, the temperature variation of the antenna structure Measured by the onboard PT1000 platinum resistance sensor; mechanical stress change Acquired by Wheatstone bridge strain gauges.

[0040] The present invention also provides a satellite antenna dynamic pointing calibration system based on on-board autonomous gradient optimization for implementing the satellite antenna dynamic pointing calibration method based on on-board autonomous gradient optimization. A layered data processing architecture is constructed, and its hardware carrier is a multi-core processor module integrated into the satellite integrated electronic system. The multi-core processor module includes four parts: a data interface unit, a gradient calculation unit, a physical field compensation unit, and an instruction synthesis unit.

[0041] The data interface unit is connected to the satellite management system via the SpaceWire bus, and receives downlink signal-to-noise ratio (SNR) telemetry data packets fed back from the ground station in real time. Simultaneously, it collects analog signals from temperature sensors and strain gauges at key locations via the CAN bus, converting them into digital sequences via a 24-bit ADC. The satellite actively applies small perturbations in four directions in the elevation and azimuth planes, i.e., in four orthogonal directions about the satellite antenna axis, and calculates gradients based on changes in the received SNR values.

[0042] The physical field compensation unit has a built-in satellite structure parameter database to store the thermal expansion coefficients of each component. , elastic modulus , when receiving the temperature change and strain variation When the cantilever beam is deformed, the directional deviation caused by the deformation is calculated in real time based on the cantilever beam deformation model.

[0043] The beneficial effects of the present invention are as follows: firstly, in order to solve the high cost problem of ground calibration equipment, a spatial geometric mapping model based on downlink signal-to-noise ratio (SNR) inversion is proposed. The nonlinear relationship replaces the traditional power difference measurement, so that the ground station only needs to provide SNR raw data without having to undertake complex calculation tasks, reducing the cost of ground facilities by more than 90%; Among them, The signal-to-noise ratio (SNR) of the signal received by the ground station; is the antenna pointing error angle (unit: degree), which includes elevation and azimuth components; is the temperature change of the antenna structure (unit: °C), measured by the onboard PT1000 platinum resistance sensor; is the change in mechanical stress (unit: microstrain ), collected by Wheatstone bridge strain gauges.

[0044] Secondly, a multi-physics field joint compensation model is established to address the coupling interference of daily periodic thermal deformation and random mechanical stress. By fusion of real-time data of satellite-borne temperature and strain sensors, the frequency-domain-time-domain coordinated suppression of dynamic errors is achieved. The pointing deviation within the temperature range is stable within 0.3°; is the multi-physics field coupling compensation angle (unit: degree); is the thermal deformation compensation coefficient (unit: ), and the thermal expansion coefficient of the material Related; is the rate of change of temperature with time (unit: ); is the stress compensation coefficient (unit: ), which is determined by the structural modal parameters; is the rate of change of mechanical stress over time (unit: ).

[0045] Finally, to address the control instability problem in long-delay scenarios such as deep space exploration, a progressive gradient descent algorithm with historical data correlation capability is designed. , by introducing an exponential decay factor Balancing the convergence speed and overshoot risk, the convergence accuracy of 0.5° can be maintained under a 20-minute communication delay. is the angle correction value for the kth iteration (unit: degree); is the gradient descent step coefficient (dimensionless), and its value is related to the antenna beamwidth; is the signal-to-noise ratio gradient vector, , is the partial derivative of the signal-to-noise ratio with respect to the pitch angle, is the partial derivative of the signal-to-noise ratio with respect to the yaw angle; is the exponential decay factor; is a time-varying attenuation term, which realizes adaptive step size control.

[0046] Compared with the Milstar solution, this technology has three advantages: first, the on-board embedded processing unit replaces the ground computing center, which reduces the calibration instruction generation delay from hours to milliseconds; second, it adopts a lightweight parameter update mechanism, which only requires the transmission of The incremental data (about 10 bytes) is saved, which reduces the bandwidth requirement by 99% compared with traditional Fourier coefficient upload.

[0047] At the structural design level, this solution restructures the hardware architecture of traditional calibration systems through deep integration of onboard processing units. Traditional solutions rely on a measurement network consisting of three dedicated ground stations located at different longitudes, each equipped with a high-precision power meter and clock synchronization device, resulting in exponentially increased system complexity. This solution migrates the computing core to an onboard multi-core processor, leveraging the existing Attitude Determination System (ADS) interface resources to directly acquire gyroscopic and star sensor data via the SpaceWire bus. It also reuses the satellite's existing temperature and strain sensor network, reducing the cost of additional hardware by 92%. The key to this structural optimization lies in the design of a lightweight data pipeline. After onboard timestamp alignment, the raw SNR data stream (50 kbps) and the sensor data stream (10 kbps) are fed into a shared memory buffer. A dual-core processor then performs gradient calculations and physical field compensation, respectively. This reduces computational latency from minutes in traditional ground-based processing to milliseconds.

[0048] In terms of functional performance, this solution achieves two breakthroughs through the synergy of gradient descent algorithm and multi-physics field model. First, the calibration accuracy reaches 0.3° in dynamic environment, which is 6 times higher than the traditional solution. This is due to the nonlinear mapping model. High-fidelity approximation of antenna pattern, and temperature compensation Accurate prediction of thermal deformation. Secondly, the ability to resist communication delay is significantly enhanced. In deep space exploration scenarios, even if the SNR data is delayed by 20 minutes, the historical data weighted algorithm can still It can still maintain a convergence accuracy of 0.5°, where the weight function Exponential decay based on temporal proximity. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Flowchart of the satellite antenna dynamic pointing calibration method;

[0050] Figure 2 This is a system connection diagram. DETAILED DESCRIPTION

[0051] The present invention will be further described below with reference to the accompanying drawings.

[0052] The satellite antenna dynamic pointing calibration method based on on-board autonomous gradient optimization has an implementation process that includes three main stages: satellite-ground coordinated data collection, on-board autonomous error analysis, and multi-physical field coupling compensation. Continuous optimization of the antenna pointing is achieved by building a closed-loop control architecture.

[0053] Step 1: Data Collection

[0054] While the satellite is in orbit, the ground station continuously receives satellite signals and records the received signal quality parameters, including but not limited to the signal-to-noise ratio (SNR), during each pass. The ground station then sends the collected SNR values ​​to the satellite via an uplink.

[0055] While receiving the signal-to-noise ratio information, the satellite records the antenna stress and temperature information through the mechanical and thermal sensors at its own antenna.

[0056] Step 2: Error Analysis

[0057] The antenna pointing error model is established using the signal quality parameters received by the ground station and the satellite attitude and orbit information.

[0058] The satellite actively applies small disturbances once in each of the four orthogonal directions along the antenna pointing axis, in the pitch and azimuth planes.

[0059] By analyzing the relationship between signal quality parameters and satellite attitude and antenna thermal parameters, the magnitude and direction of the current antenna pointing error can be determined.

[0060] When the satellite's attitude is actively disturbed, if the received signal quality parameters show a significant decrease, it indicates that the antenna pointing may have deviated from the direction of the ground station.

[0061] Step 3: Error Correction

[0062] According to the error analysis results, the antenna pointing error is gradually reduced by adjusting the satellite attitude / the angle of the antenna rotation arm / the feed electronic scanning angle.

[0063] After each adjustment, the ground station receiving signal quality parameters are collected again to evaluate the correction effect.

[0064] Through iterative adjustment, the antenna pointing error is eventually converged to the allowable range, achieving high-precision pointing.

[0065] Step 4: Calibration completed

[0066] When the received signal quality parameters reach the expected indicators and stabilize at a high level, the antenna pointing calibration is considered completed.

[0067] The final satellite attitude and antenna rotation arm angle parameters are recorded as the reference for subsequent antenna pointing control.

[0068] The specific implementation process is as follows Figure 1 、 2 As shown in the figure, the ground station first receives the satellite downlink signal and calculates the signal-to-noise ratio (SNR) parameter, which is then packaged with the CCSDS telemetry protocol and uploaded to the onboard processing unit, which simultaneously records the current attitude angle (pitch angle , yaw angle ) and temperature / stress sensor data to form a temporally and spatially correlated raw data set. Then, the error direction identification phase begins. The onboard embedded processor uses a nonlinear mapping model to identify the error direction. Calculate the gradient vector of the current pointing deviation ,in, is the current signal-to-noise ratio measurement (dimensionless, usually expressed in dB); is the theoretical maximum signal-to-noise ratio (optimal value when the antenna is aligned); is the elevation angle pointing error (unit: degree, in satellite coordinate system); is the yaw angle pointing error (unit: degree, in satellite coordinate system); is the comprehensive pointing error angle (equivalent space vector modulus); is the pattern attenuation function (derived from the cosine square approximation of the antenna pattern); is the signal-to-noise ratio gradient vector (unit: ); is the partial derivative of the signal-to-noise ratio with respect to the pitch angle; is the partial derivative of the signal-to-noise ratio with respect to the yaw angle.

[0069] This process is achieved using the finite difference method:

[0070]

[0071] in, For the current moment The signal-to-noise ratio measurement value; The last measurement time signal-to-noise ratio; is the time interval (unit: seconds); is the pitch angle disturbance actively applied (unit: degree), with a typical value of 0.1°.

[0072] The improvement lies in the introduction of a sliding time window mechanism. Dynamically adjust to (in, The basic sampling interval, in seconds, is determined according to the satellite attitude control cycle. is the window width adjustment coefficient, unit ,control Sensitivity to gradient changes) makes it possible to automatically shorten the sampling interval when the gradient is large to increase the convergence speed.

[0073] The core step is adaptive gradient descent correction, and its iterative formula is designed as:

[0074]

[0075]

[0076] in, is the pitch angle correction value of the kth iteration (unit: degree); is the yaw angle correction value for the kth iteration (unit: degree); is the learning rate coefficient (dimensionless), ranging from 0.1 to 0.3 (matching the antenna beamwidth); is the attenuation coefficient (unit: 1 / time).

[0077] The essential difference from the traditional gradient method is the exponential decay term The introduction of this factor makes the iterative step size adaptively reduced as the number of times k increases, which not only ensures that the optimal solution is quickly approached in the early stage, but also avoids oscillation near the extreme point in the later stage. Experimental data show that when , the number of convergence steps is reduced by 40% and the overshoot is reduced to below 0.02°.

[0078] The dynamic error compensation module is another innovation that integrates temperature sensor data in real time. and strain gauge output , establish a physically driven compensation:

[0079]

[0080] in, is the structural characteristic length, 、 are the elastic modulus and shear modulus, respectively. 、 are the thermal expansion coefficient and strain sensitivity coefficient of the material, is the total number of temperature sensors, is the number of measuring points of the strain gauge, For the The model calculates the deformation influence in real time through the satellite structure finite element parameter library stored on board, and superimposes it with the gradient descent output to form the final attitude correction instruction.

[0081] Data exchange between each step is achieved through an onboard bus. The error identification module updates the gradient vector at a frequency of 10Hz, and the compensation module refreshes the physical field parameters at a frequency of 1Hz. A dual-rate mechanism balances computational load and response speed. Compared to the single batch processing mode of the Milstar solution, this design's streaming processing architecture reduces computational latency from minutes to sub-seconds, meeting the real-time requirements of the LEO satellite transit window.

[0082] The principle of this solution is as follows: Feedback substitution: After the satellite signal is received by the ground station next time, the ground station measures the signal-to-noise ratio (SNR) and transmits it back to the satellite. The satellite uses the SNR value as the sensor value and maximizes the SNR through an optimization algorithm to indirectly correct the pointing error; dynamic compensation: Real-time fusion of thermal, mechanical, and inertial sensor data to predict and offset the multi-physical field coupling effects.

[0083] The core concept of this solution lies in building an onboard autonomous closed-loop control system, dynamically eliminating antenna pointing errors through multi-source data fusion and a compensation model driven by physical mechanisms. Its theoretical framework comprises three core modules: error modeling, gradient optimization, and multi-physics field coupling, forming a complete error suppression chain.

[0084] The total pointing error can be decomposed into:

[0085] in:

[0086] Static assembly error: , obeys zero-mean Gaussian distribution;

[0087] Thermally induced deformation error: Modeling based on thermoelastic theory , where is the structural characteristic length, is the moment of inertia of the section, For the Temperature rise at each measuring point.

[0088] Mechanical stress error: expressed by modal superposition ,in For the order natural frequency, is the modal participation factor.

[0089] Time-varying drift error: , is a random walk process.

[0090] The error elimination mechanism is as follows:

[0091] 1. Gradient optimization module, using improved gradient descent algorithm:

[0092]

[0093] Where: exponential decay term :Achieve variable step size control and meet Lyapunov stability conditions (H is the Hessian matrix).

[0094] Symbolic function terms : Suppress local extreme value traps and improve global convergence.

[0095] 2. Multi-physics field compensation module to build a deformation predictor:

[0096]

[0097] This model achieves real-time deformation prediction by preloading satellite structure finite element parameters, and its compensation accuracy is improved compared with the traditional Fourier method:

[0098] This solution builds a layered data processing architecture on board the satellite. Its hardware carrier is a multi-core processor module integrated into the satellite's integrated electronic system, which includes four parts: data interface unit, gradient calculation unit, physical field compensation unit, and instruction synthesis unit. The data interface unit is connected to the satellite management system via the SpaceWire bus, and receives real-time downlink signal-to-noise ratio (SNR) telemetry data packets fed back from the ground station (the format complies with the CCSDS 132.0-B-2 standard). At the same time, it collects temperature data from temperature sensors (PT1000 platinum resistance, precision) distributed in key locations such as the antenna bracket and feed source via the CAN bus. ) and strain gauges (Wheatstone bridge, range The satellite actively applies small perturbations in four directions in the elevation and azimuth planes, i.e., in four orthogonal directions about the satellite antenna axis, and calculates the gradient based on the changes in the received SNR value.

[0099] The physical field compensation unit has a built-in satellite structure parameter database that stores the thermal expansion coefficients of each component. , elastic modulus When receiving the temperature change and strain variation When , the directional deviation caused by deformation is calculated in real time based on the cantilever beam deformation model:

[0100]

[0101] This formula is derived from the bending deformation theory of beams in material mechanics, where is the characteristic length, is the section inertia moment. For the dynamic stress disturbance caused by the vibration of the solar sail panel, the modal superposition method is used to perform frequency domain decomposition:

[0102]

[0103] In the formula is the kth order natural frequency, The outputs of the gradient calculation unit and the physical field compensation unit are vector-superimposed in the instruction synthesis unit to generate the total correction amount. , sent to the attitude control computer via the 1553B bus, driving the reaction flywheel or magnetic torquer to perform pointing adjustment.

[0104] The fundamental breakthroughs in the core steps involved in the improvement are reflected in two aspects: First, the exponential decay factor introduced in the gradient descent algorithm , its mathematical essence is to embed the time-varying gain in the iterative step, so that the algorithm can achieve the desired result in the early stage ( When it is small) keep a large step size to quickly approach the extreme point, and in the later stage ( When the value increases, the step size is automatically reduced to suppress overshoot. This design is based on the Lyapunov stability theory and constructs the energy function It can be proved that when ( is the Lyapunov coefficient). Secondly, the multi-physics compensation model transforms the traditional Fourier series pure mathematical fitting into a prediction driven by physical mechanism, directly linking temperature / stress changes with mechanical deformation through the material constitutive equation, making it possible to predict the deformation under non-periodic disturbances (such as those caused by micrometeoroid impacts). ), and experiments show that this model can reduce the pointing error by 67% under sudden interference.

[0105] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A satellite antenna dynamic pointing calibration method based on onboard autonomous gradient optimization, characterized in that: The specific steps are as follows: Step 1: Data Collection (11) During the satellite's in-orbit operation, the ground station continuously receives satellite signals and records the received signal quality parameters at each transit, including but not limited to the signal-to-noise ratio (SNR). The ground station transmits the collected SNR information to the satellite via an uplink. (12) While receiving the signal-to-noise ratio information, the satellite records the antenna stress and temperature information through the mechanical and thermal sensors at its own antenna; Step 2: Error analysis (21) Using the signal quality parameters received by the ground station and the satellite attitude and orbit information, the antenna pointing error model is established: in, is the current signal-to-noise ratio measurement value; is the theoretical maximum signal-to-noise ratio, i.e., the optimal value when the antenna is aligned; is the pitch angle pointing error; is the yaw angle pointing error; is the comprehensive pointing error angle; is the pattern attenuation function; (22) The satellite actively applies a small perturbation once in each of the four orthogonal directions along the antenna pointing axis, in the pitch and azimuth planes; based on the nonlinear mapping model Calculate the gradient vector of the current pointing deviation ;in, is the signal-to-noise ratio gradient vector; is the partial derivative of the signal-to-noise ratio with respect to the pitch angle; is the partial derivative of the signal-to-noise ratio with respect to the yaw angle; This is achieved using the finite difference method: in, For the current moment The signal-to-noise ratio measurement value; The last measurement time signal-to-noise ratio; is the time interval; is the pitch angle disturbance actively applied; Introducing a sliding time window mechanism Dynamically adjust to ,in, Is the basic sampling interval, in seconds , determined according to the satellite attitude control period; is the window width adjustment coefficient, unit ,control Sensitivity to gradient changes, which automatically shortens the sampling interval when the gradient is large to increase the convergence speed; (23) By analyzing the relationship between signal quality parameters and satellite attitude and antenna thermal parameters, the magnitude and direction of the current antenna pointing error can be determined; (24) When the satellite actively performs attitude disturbances, if the received signal quality parameters show a significant decrease, it indicates that the antenna pointing direction may have deviated from the direction of the ground station; Step 3: Error Correction (31) Based on the error analysis results, the antenna pointing error is gradually reduced by adjusting the satellite attitude, the angle of the antenna rotating arm, and the feed electronic scanning angle; (32) After each adjustment, the ground station receiving signal quality parameters are collected again to evaluate the correction effect; (33) Through iterative adjustment, the antenna pointing error is finally converged to the allowable range, achieving high-precision pointing; Step 4: Calibration completed (41) When the received signal quality parameters reach the expected indicators and stabilize at a high level, the antenna pointing calibration is considered completed; (42) Record the final satellite attitude and antenna rotation arm angle parameters as a reference for subsequent antenna pointing control.

2. The satellite antenna dynamic pointing calibration method based on onboard autonomous gradient optimization according to claim 1, characterized in that: Step 3: In error correction, adaptive gradient descent correction is used, and the iterative formula is designed as follows: in, For the The pitch angle correction amount for the iteration; For the The yaw angle correction amount for the iteration; is the learning rate coefficient, ranging from 0.1 to 0.3; is the attenuation coefficient; Exponential decay term The introduction of It increases and adaptively reduces, which not only ensures that the optimal solution is quickly approached in the early stage, but also avoids oscillation near the extreme point in the later stage.

3. The satellite antenna dynamic pointing calibration method based on onboard autonomous gradient optimization according to claim 2, characterized in that: Step 3: In error correction, dynamic error compensation is used to integrate temperature sensor data in real time. and strain gauge output , establish a physically driven compensation: in, is the structural characteristic length, 、 are the elastic modulus and shear modulus, respectively. 、 are the thermal expansion coefficient and strain sensitivity coefficient of the material, is the total number of temperature sensors, is the number of measuring points of the strain gauge, For the The equivalent action area of ​​each strain point.

4. The satellite antenna dynamic pointing calibration method based on onboard autonomous gradient optimization according to claim 4 is characterized in that: In adaptive gradient descent correction, the gradient vector is updated at a frequency of 10 Hz; in dynamic error compensation, the physical field parameters are refreshed at a frequency of 1 Hz; and a dual-rate mechanism is used to balance the computational load and response speed.

5. The satellite antenna dynamic pointing calibration method based on onboard autonomous gradient optimization according to claim 1, characterized in that: The temperature change of the antenna structure Measured by the onboard PT1000 platinum resistance sensor; mechanical stress change Acquired by Wheatstone bridge strain gauges.

6. A satellite antenna dynamic pointing calibration system based on on-board autonomous gradient optimization that implements the satellite antenna dynamic pointing calibration method based on on-board autonomous gradient optimization according to any one of claims 1 to 5, characterized in that: A hierarchical data processing architecture was constructed, whose hardware carrier is a multi-core processor module integrated into the satellite's integrated electronic system. The multi-core processor module includes four parts: a data interface unit, a gradient calculation unit, a physical field compensation unit, and an instruction synthesis unit. The data interface unit is connected to the satellite management system via the SpaceWire bus, and receives the downlink signal-to-noise ratio (SNR) telemetry data packets fed back from the ground station in real time. At the same time, it collects analog signals from temperature sensors and strain gauges at key locations via the CAN bus, and converts them into digital sequences after 24-bit ADC conversion. The satellite actively applies small perturbations in four directions in the pitch and azimuth planes, that is, perturbations are applied in four orthogonal directions of the satellite antenna axis, and the gradient is calculated based on the changes in the received SNR value; The physical field compensation unit has a built-in satellite structure parameter database to store the thermal expansion coefficients of each component. , elastic modulus , when receiving the temperature change and strain variation When the cantilever beam is deformed, the directional deviation caused by the deformation is calculated in real time based on the cantilever beam deformation model.

Images