A star sensor system based on liquid crystal polarization grating (LCPG)

Abstract

This invention discloses a star sensor system based on a liquid crystal polarization grating (LCPG), comprising an LCPG receiving optical path module, a main optical lens and detector module for detecting starlight, and an FPGA signal processing module. The LCPG receiving optical path module includes at least one polarizer and two stages of LCPG units. The first stage of the LCPG unit is used to adjust the incident light to two different diffraction directions. The two stages of the LCPG units form a two-dimensional adjustable field of view. The FPGA signal processing module controls the LCPG receiving optical path module in real time through a system status self-detection unit and a temperature control unit. Time-division multiplexing is used for image acquisition, and the image data processed by the FPGA is then subjected to image registration and fusion and output. This invention expands the field of view of star sensors and improves the problem of insufficient resolution through an adjustable liquid crystal polarization grating system. It is suitable for weight reduction and simplification of existing traditional optical star sensors, reducing the target surface size and resolution requirements of optical sensors.

Description

Technical Field

[0001] This invention relates to the field of spacecraft attitude determination technology, and in particular to a star sensor system based on a liquid crystal polarization grating (LCPG). Background Technology

[0002] Traditional geostationary orbit and low Earth orbit large satellite missions are the main application areas of star sensors. However, in recent years, with the rapid development of missions such as deep space exploration, high-precision Earth observation, and constellation networking, new mission scenarios have emerged and posed extreme challenges to their performance.

[0003] 1. Rapid autonomous attitude determination for satellites

[0004] SpaceX's Starlink, Planet Labs' Dove constellation, and China's future GW and G60, among tens of thousands of other low-Earth orbit (LEO) internet and remote sensing constellations, are all examples. These satellites, typically micro- and nano-satellites, are numerous and require a high degree of autonomous operation. After launch and orbit insertion, or after attitude instability due to the space environment, they must be able to autonomously complete the entire "acquisition-identification-attitude determination" process within a very short time, achieving autonomous attitude determination. This places extremely high demands on the star sensor's all-space-area identification capability and identification time. Thousands of LEO communication constellation satellites, after launch, need to complete attitude determination and commence operations within a very short time. If the field of view is too small, the satellite needs to rotate for a long time to collect enough star images to match the star map, significantly extending the operational preparation time. The paper "Fast Star Pattern Recognition for Sky Acquisition" points out that the field of view angle and initial acquisition time are inversely proportional; doubling the field of view can reduce the acquisition time by more than 60%.

[0005] Patent CN114705182A proposes a conical scanning field-of-view star sensor using a prism. This method places the prism in front of a lens and rotates it via a servo mechanism. Based on the light refraction effect of the prism, a motor drives the prism to rotate, achieving a conical scan of the star sensor's field of view and supplementing the star sensor's field of view. This system relies on motor-driven mechanical attitude determination, and scanning the entire field of view requires waiting for the motor to complete a 2-5 minute cycle, which is relatively slow.

[0006] 2. Miniaturization, integration, and low cost

[0007] This is a rigid constraint for micro and nano satellites. The field of view (FOV) of traditional star sensors is limited by a fixed optical structure, and they usually rely on mechanical rotation or multi-lens stitching to expand the field of view or determine the attitude, resulting in large size, high power consumption, low reliability, and easy damage.

[0008] The paper "Optical System Design of a Large-FOV High-Resolution StarSensor" designed a system with a 15°x15° field of view, but at the cost of using nine lenses, a total system length exceeding 150mm, and a weight exceeding 800g, making it unsuitable for micro / nano satellites. Reports from publications such as *Small Satellite Market Observations* indicate that nanosatellites (1-10kg) typically require an attitude determination and control system with a weight requirement of less than 500g and a power consumption requirement of less than 2W.

[0009] 3. High-resolution deep space and interplanetary exploration

[0010] In the deep space environment, without auxiliary references such as geomagnetism or GPS, star sensors are the absolute attitude reference. Their accuracy directly determines the success or failure of missions such as orbit insertion and landing site selection, and is of great significance in lunar / Mars bases and asteroid exploration. Higher resolution means being able to distinguish closer binary stars and detect fainter stars, thereby improving the accuracy of star barycenter positioning.

[0011] The model in *Fundamentals of Star Sensor Technology* suggests that detecting a +6 Mv star typically requires an angular resolution better than 20-30 arcsec / pixel. Higher resolution means each star point covers more pixels on the detector, allowing for more accurate centroid algorithms to distinguish it from noise. High resolution necessitates the smallest possible detector pixel size. However, small pixels reduce the full-well capacity per pixel and worsen the dynamic range, requiring a trade-off.

[0012] The literature (Liu Xin, Wang Jingjin, et al. optical system design of largerelative aperture star sensor[J]. Journal of Beijing Information Science & Technology University, 2025) proposes a high-precision attitude sensing device, designing a traditional transmission optical system with an effective focal length of 100mm, a full field of view of 5.26°, and a spectral range of 480-760nm. The large relative aperture of 1.4 meets the requirements for detecting faint stars. However, the full field of view is relatively small and the system efficiency is relatively low.

[0013] In summary, given the current situation and shortcomings, star sensors need to simultaneously meet the requirements of being fast, small, having a large field of view, and high resolution. Existing traditional devices cannot simultaneously balance these factors.

[0014] Liquid crystal optical devices are photonic devices based on the electrically controlled birefringence (ECB) effect. Their phase, polarization, and optical path characteristics can be dynamically and programmably adjusted by an external electric field. Liquid crystal polarization gratings (LCPGs) are one such advanced device, characterized by their small size, easy integration, millisecond-level fast electronic switching, and absence of mechanical moving parts.

[0015] LCVR and LCPG, as novel self-assembly / self-phase modulation materials, combine polarization, phase, and amplitude modulation capabilities, and have the potential to achieve large field-of-view optical deflection functions on a single substrate. They can also be tightly coupled with the receiver detection unit, improving the overall system sensitivity and bandwidth. Applying LCPG to star sensors has significant potential value: its large-angle beam deflection capability without mechanical movement can be used to construct computational imaging star sensors. The proposed scheme involves deflecting the image formed by a large field-of-view optical system in a time-division and region-division manner using an LCPG array, guiding it onto a small, high-resolution image sensor. This replaces the complex optical design with time-scanning and electronic processing, potentially achieving a balance between large field of view and high resolution with a simple optical structure and low SWaP.

[0016] However, applying LCPGs to the harsh aerospace environment still faces significant challenges. Jones et al.'s article, "Performance of Liquid Crystal Polarization Gratings in the Space Environment: A Review," systematically reviews the challenges of LCPGs in space applications: temperature sensitivity and radiation-induced degradation. The article points out that without special design and protection, the diffraction efficiency of LCPGs undergoes irreversible degradation after experiencing wide temperature cycling and radiation doses, leading to a decrease in pointing accuracy and light energy utilization.

[0017] Temperature sensitivity: The refractive index, viscosity, response time, and other parameters of liquid crystal materials are strongly dependent on temperature. The temperature range of the space environment is extremely wide (-100°C to above +100°C). Dramatic temperature changes can cause the diffraction efficiency and switching rate of LCPGs to drift or even fail, resulting in loss of beam control precision and incorrect star point positioning.

[0018] LCPG Deformation and Failure: Temperature cycling and space radiation can cause aging and deformation of liquid crystal materials, changes in cell thickness, and displacement of support spacers, resulting in changes in device planarity, diffraction wavefront distortion, and the introduction of static aberrations that cannot be compensated for electronically. High-energy particles and radiation can ionize liquid crystal molecules, causing permanent degradation of their electro-optic properties, increased driving voltage, slower response speed, and ultimately, performance decline.

[0019] The above factors together cause the efficiency of LCPG to be unstable in actual on-orbit operation. The fluctuation of efficiency will directly translate into fluctuation of transmitted light intensity, which will be captured by image sensors and misjudged as changes in star brightness or introduce background noise, seriously affecting the imaging algorithm.

[0020] Existing solutions have failed to effectively address these engineering challenges, preventing the application of LCPGs in high-precision space missions from ever truly taking place at the theoretical level. In summary, this invention addresses the pain points of applying liquid crystal devices in space and designs the following star sensor system. Summary of the Invention

[0021] This invention provides a large field-of-view, high-resolution star sensor optical system based on LCPG. The core of this system lies in using high-aperture, programmable LCVR and LCPG elements to deflect light entering the star sensor at a large angle, while simultaneously combining this with a detector chip array at the receiver to achieve high-resolution imaging. This improves signal coverage and signal-to-noise ratio (SNR) under wide field-of-view conditions. This system not only leverages the large-angle electrically controlled beam scanning advantage of LCPG but also overcomes the negative impacts of the space environment on liquid crystal devices through an innovative system-level design, ensuring long-term, stable, and high-precision operation throughout the entire mission cycle.

[0022] The technical solution adopted in this invention is as follows: a star sensor system based on liquid crystal polarization grating (LCPG), including an LCPG receiving optical path module, a main optical lens and detector module, and an FPGA signal processing module;

[0023] The LCPG receiving optical path module includes at least one polarizer and two LCPG units. The first LCPG unit consists of a polymer grating and a liquid crystal variable delay device (LCVR) to adjust the incident light to two different diffraction directions. The polarizer and LCVR have the same polarization adjustment direction. The second LCPG unit is perpendicular to the polarization direction of the first LCPG unit, forming a two-dimensional adjustable field of view.

[0024] The main optical lens and detector module are used to detect starlight;

[0025] The FPGA signal processing module includes a system status self-detection unit, a temperature control unit, an LCPG imaging calibration unit, and an image registration and fusion unit. The temperature control unit is used to monitor and regulate the temperature, which is achieved through the temperature sensor and TEC of the LCPG imaging calibration unit. The FPGA signal processing module controls the LCPG receiving optical path module in real time through the system status self-detection unit and the temperature control unit. When acquiring images, the FPGA signal processing module uses time-division acquisition, and the image data processed by the FPGA is then performed by the image registration and fusion unit and output.

[0026] Furthermore, the liquid crystal variable delay device (LCVR) only adjusts the adjustable polarization direction to achieve the phase delay function.

[0027] Furthermore, the LCVR modulates the incident light in front, while the polymer grating behind provides different but symmetrical diffraction angles for left-handed and right-handed circularly polarized light.

[0028] Furthermore, the LCPG receiving optical path module can add more LCPG units with different diffraction angles in the two-dimensional direction to expand the field of view.

[0029] Furthermore, the specific implementation process of the LCPG imaging calibration unit is as follows:

[0030] Step 1: Initialize the LCVR voltage on the FPGA to zero;

[0031] Step 2: Start by iterating through the LCVR's driving voltage in small voltage steps, and acquire an image each time a different voltage is applied;

[0032] Step 3: Calculate the image sharpness curves under different voltages using the Tennengrad gradient direction;

[0033] Step 4: Index out the driving voltages corresponding to the two maxima of the sharpness curve, and use them as the field-of-view driving voltages corresponding to the two diffraction angles of the first-stage LCPG unit.

[0034] Furthermore, the specific implementation process of the temperature control unit is as follows:

[0035] Step 1: The system monitors the temperature. When the temperature exceeds the set range, the FPGA activates the temperature control unit.

[0036] Step 2: The PID controller calculates the error between the temperature setpoint and the monitored temperature feedback value, and outputs the current and voltage values, directions, and time parameters applied to the TEC;

[0037] Step 3: Apply temperature control signal to FPGA;

[0038] Step 4: Repeat steps 1-3 until the temperature is within the set range, then pause the temperature control.

[0039] Furthermore, the specific implementation process of the image registration and fusion unit is as follows:

[0040] Step 1: Switch between the two different fields of view and acquire images based on the voltages provided by the LCPG imaging calibration unit;

[0041] Step 2: Optimize aberrations based on camera intrinsic parameters and distortion parameters;

[0042] Step 3: Calculate the pixel positions on the image based on the diffraction angle of LCPG, and coarsely register the image;

[0043] Step 4: Sampling feature points in the overlapping areas of images from different fields of view, and performing grayscale curve matching;

[0044] Step 5: Repeat step 4 until the number of feature points exceeds the set threshold. Based on the feature point positions, finely register and fuse the images, and output the image.

[0045] Furthermore, the specific steps for the FPGA signal processing module to acquire images in a time-division manner are as follows:

[0046] Step 1: Initialize the LCPG drive voltage on the FPGA;

[0047] Step 2: The FPGA loads the first field-of-view driving voltage according to the driving voltage LUT table corresponding to different fields of view calibrated by the LCPG imaging calibration unit, waits for the liquid crystal response time, and at the same time, the FPGA acquires the first image data.

[0048] Step 3: The FPGA finds and applies the driving voltage corresponding to the next field of view in the horizontal direction. After waiting for the liquid crystal response time, it acquires the second image data.

[0049] Step 4: After traversing the horizontal field of view, the FPGA finds and applies the driving voltage corresponding to the next vertical field of view. After waiting for the liquid crystal response time, it acquires image data and repeats Step 3.

[0050] Step 5: After traversing all fields of view, the FPGA assigns image numbers to different fields of view based on the fields of view corresponding to the driving voltage in the LUT, and inputs them into the image registration and fusion unit.

[0051] Furthermore, the system also includes a radiation-resistant outer shell structure for shock protection, isolation of charged particles, and isolation of secondary particles. The radiation-resistant outer shell structure comprises a multi-layered polymer and heavy metal structure, with an outer layer for shock resistance and heat insulation, a middle layer for absorbing charged particles, and an inner layer for insulation and absorption of secondary particles.

[0052] The beneficial effects of this invention are:

[0053] 1. Successfully solved the space adaptability problem of LCPG: Through active temperature control and online calibration-feedback compensation mechanism, the influence of temperature and radiation effects on LCPG performance was effectively suppressed, enabling it to meet the high reliability and long life requirements of aerospace missions.

[0054] 2. Achieved truly common-use large field-of-view, high-resolution imaging: By using a simple optical structure and a method without moving parts, it breaks through the limitations of traditional optical design and obtains a large field of view and high resolution.

[0055] 3. Improved system intelligence and reliability: Built-in calibration and management functions enable the system to have self-diagnosis and self-adaptation capabilities, reducing the dependence on ground calibration and the difficulty of on-orbit maintenance, and significantly improving overall reliability. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 : A simplified schematic diagram of the system of the present invention.

[0058] Figure 2 : Schematic diagram of the specific structure of the LCPG receiving optical path module.

[0059] Figure 3 Schematic diagram of a passive LCPG unit.

[0060] Figure 4 Schematic diagram of the combined structure of temperature sensor and LCPG unit.

[0061] Figure 5 : Schematic diagram of FPGA integrated system architecture.

[0062] Figure 6 : Schematic diagram of image registration and fusion algorithm process.

[0063] Figure 7 : Schematic diagram of the LCPG unit initialization calibration algorithm.

[0064] Figure 8 : Schematic diagram of temperature control algorithm.

[0065] Figure 9 : Schematic diagram of time-sharing image acquisition algorithm.

[0066] Figure 10 Schematic diagram of radiation-resistant outer shell structure.

[0067] Figure 11 : A schematic diagram of the specific structure of the finished product of the system of the present invention.

[0068] In the figure, LCPG receiving optical path module 1; LCPG unit status feedback system 2; FPGA signal processing module 3; steady-state temperature control system 4; image registration and fusion unit 5; first-stage LCVR 6; second-stage LCVR 7; first-stage polymer film LCPG 8; second-stage polymer film LCPG 9; photodetector 10; glass cover plate 11; ITO conductive glass 12; PI insulating layer 13; liquid crystal polymer layer 14; LCPG polymer layer 15; temperature sensor 16; LCPG unit 17; LCPG mounting substrate 18; aluminum alloy layer 19; polyester film layer 20; indium tin oxide conductive coating 21; polyethylene metal composite material 22; LCPG unit module 23; radiation-resistant shell 24; main optical imaging system and sensor module 25; FPGA 26; temperature sensing and TEC temperature control 27. Detailed Implementation

[0069] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are merely illustrative and not intended to limit the invention.

[0070] like Figure 1 As shown, this invention discloses a large field-of-view, high-resolution star sensor system based on a liquid crystal polarization grating (LCPG). The system includes an LCPG receiving optical path module 1 for receiving starlight and transmitting the acquired image data to an FPGA signal processing module 3 for signal processing. After data processing, the FPGA controls the LCPG receiving optical path module 1 in real time through two closed-loop systems—a steady-state temperature control system 4 and an LCPG unit state feedback system 2. Finally, the effective image data processed by the FPGA is processed by an image registration and fusion unit 5 and output.

[0071] The LCPG receiving optical path module 1 differs from traditional multi-lens complex optical systems. It includes a polarizer, an LCPG whose beam deflection direction is controlled by voltage, a main optical lens, and a detector module. The main optical lens and detector module consists of an optical lens and a high-resolution, high-sensitivity image sensor. The high-resolution, high-sensitivity image sensor supports rapid exposure and synchronous triggering, enabling real-time capture of multi-field star maps after LCPG deflection. The optical lens provides a small field of view and incorporates achromatic and thermal aberration reduction designs. LCPGs are divided into active and passive types. Active LCPGs achieve field-of-view deflection by changing the driving voltage when circularly polarized light is incident. Passive LCPGs utilize LCVRs (liquid crystal variable retarder) to adjust the polarization state of the incident light, changing the LCVR's driving voltage to match the circularly polarized light rotation direction required for entry into the LCPG, thus cooperating with the LCPG to achieve time-division multi-field imaging. Each passive LCPG unit includes two LCVRs and one LCPG. The passive LCPG is composed of a glass layer and a polymer film, forming a periodically varying polarization grating structure. Each LCVR unit consists of a transparent electrode, an alignment layer, and a liquid crystal polymer film, based on the electrically controlled birefringence effect of liquid crystal molecules. The alignment layer is an insulating layer with a special coating design, exhibiting a small tilt angle with consistent orientation to form a specifically oriented liquid crystal layer. This anchors the initial orientation of the liquid crystal molecules, typically forming a small angle with the LCVR glass substrate. When a specific AC voltage is applied, its effective optical axis direction changes, thereby achieving electrically controlled adjustment of the polarization state of the incident polarized light. Each active LCPG includes two LCVRs and one LCPG. The LCVR is used to adjust the polarization of the incident light to circularly polarized light. The LCPG consists of a transparent electrode, a PI layer, an alignment layer, and a liquid crystal layer. The transparent electrode applies a voltage to the liquid crystal molecules, the PI layer is insulating, the alignment layer anchors the orientation of the surface liquid crystal molecules, and the liquid crystal layer responds to different driving voltages, achieving different deflection directions of the incident circularly polarized light. In the LCPG unit, the LCVR dynamically adjusts the phase delay to eliminate imaging errors caused by the polarization sensitivity of the LCPG.

[0072] like Figure 2The diagram shows the specific structure of the LCPG receiving optical path module 1. The incident light first reaches the first-stage LCPG unit, which includes a first-stage LCVR 6 and a first-stage polymer film LCPG 8. The first-stage LCVR 6 is used to adjust the polarization state of the incident light. The FPGA applies a voltage Vx to the first-stage LCVR 6, modulating the incident light into different polarizations. The first-stage polymer film LCPG 8 then diffracts at two symmetrical angles according to the adjusted polarization states. The FPGA switches the LCPG unit deflection voltage according to a preset timing sequence, causing the beam to sequentially cover adjacent field-of-view regions. Here, the FPGA applies a calibrated voltage to cause diffraction at a specific angle, completing the X-direction deflection. Subsequently, the light enters the second-stage LCPG unit. The FPGA applies a calibrated voltage Vy to the second-stage LCVR 7, causing the second-stage polymer film LCPG 9 to diffract at a specific angle in the Y direction. The vector synthesis of the two diffraction angles ultimately deflects the light precisely to a predetermined direction, forming the outgoing light that is received by the photodetector 10.

[0073] The LCPG receiving optical path module enables beam scanning without mechanical movement. It receives optical information from the entire field of view of the large field-of-view main optical lens and, according to instructions from the FPGA, precisely deflects and relays the starlight in the field of view to a stationary high-resolution image sensor in a time-division manner, achieving millisecond-level switching to acquire the field of view while maintaining both a large field of view and high resolution. Its operating state is precisely controlled by the drive control unit.

[0074] The LCPG is located before the main optical lens and detector module, and consists of at least two independent LCPG units. Each unit encompasses multiple directions, including at least the X-axis (horizontal) and Y-axis (vertical). Multiple units together constitute two-dimensional or higher-dimensional adjustment, and these units are cascaded through a precision optical adjustment frame. In this invention, the LCPG units are not limited to parallel placement; for additional system design requirements, they can be placed at an angle.

[0075] like Figure 3The diagram shows the structure of the LCPG unit 17, including the microstructures of the LCVR and LCPG. The LCVR is on top, and the LCPG is below. The LCVR is protected by two glass cover plates 11. Inside the glass cover plates is an ITO conductive glass 12, which also serves as the electrode driving the LCVR. A PI insulating layer 13 isolates the ITO conductive glass 12 and the liquid crystal polymer layer 14. An AC voltage supplied by the FPGA is applied to the ITO conductive glass 12. The electric field acts on the liquid crystal polymer layer 14, changing the arrangement of the liquid crystal molecules within it, thereby modulating its birefringence properties. The PI insulating layer 13 is manufactured with a small tilt angle and orientation, which determines the initial arrangement and deflection direction of the liquid crystal molecules, making it respond to a specific polarized light. After adjustment by the LCVR, the polarized light becomes circularly polarized light, passing through the glass cover plate 11 and the LCPG polymer layer 15, and is diffracted to two symmetrical angles depending on whether it is left- or right-handed. By changing the voltage amplitude, the phase modulation depth of the equivalent grating can be controlled, thereby achieving continuous electronic adjustment of the diffraction angle.

[0076] The steady-state temperature control system 4 is an active precision temperature control system. It includes an insulating cavity, a thermoelectric cooler, a temperature sensor, and a temperature control circuit. Through closed-loop control using a PID algorithm, it maintains the LCPG / LCVR operating temperature within ±0.5℃, strictly controlling it within a narrow range where its performance is optimal and stable. This fundamentally suppresses changes in the refractive index of the liquid crystal material, response speed drift, and diffraction efficiency fluctuations caused by drastic changes in ambient temperature. The insulating cavity encapsulates the LCPG beam receiving and control system, isolating it from the external thermal environment. The insulation structure utilizes multi-layer reflective foil and aerogel to isolate external thermal interference. The thermoelectric cooler is attached to the LCPG device mounting substrate for heating or cooling. The temperature sensor, with an adjustment accuracy within 0.5 degrees Celsius, is tightly fitted between the LCPG and the mounting substrate, monitoring its temperature in real time.

[0077] like Figure 4 The diagram shows the connection between the temperature sensor and the LCPG unit. The temperature sensor 16 is firmly attached to the LCPG mounting substrate 18 with thermally conductive adhesive, ensuring rapid and accurate sensing of temperature changes in the LCPG unit 17. The sensor converts the temperature value into a resistance change and transmits it to the analog-to-digital converter inside the FPGA. Based on this, the FPGA obtains the real-time temperature of the LCPG, which serves as a feedback signal for temperature closed-loop control.

[0078] FPGA Signal Processing Module 3 is an integrated FPGA system that manages all operating modes, including power-on, initialization, calibration, observation, and sleep. It comprises an LCPG drive signal generation system, an image sensor timing control and data acquisition system, an image fusion system, a temperature closed-loop control system, a calibration system, and a communication system. The LCPG drive signal relies on a high-precision, programmable voltage waveform to control the LCPG deflection state. A voltage-deflection angle lookup table (LUT) is pre-stored within the FPGA. Based on the current sub-field coordinates (i, j) to be scanned, the amplitude code of the high-voltage analog drive signal required by the corresponding multi-level LCPG unit is read from the LUT and sent to the high-precision digital-to-analog converter chip via the SPI / I2C interface to generate the required analog voltage. The image sensor driver generates the required pixel clock, horizontal and vertical synchronization signals, and controls the exposure time. It generates drive signals that meet the timing requirements of the image sensor. It receives the image data stream from the image sensor, deserializes it, and stores it in the image buffer. Image data acquisition and preprocessing are performed, receiving the raw image output data and performing real-time dark current, flat field correction, and non-uniformity correction. The temperature closed-loop control system executes a PID algorithm to dynamically adjust the thermoelectric cooler. It reads temperature sensor data converted by a high-precision ADC via an SPI interface. Internally, a digital PID controller compares the measured temperature with the setpoint, calculates the control input, and controls the current magnitude and direction of the TEC driver chip via the SPI interface. The on-orbit calibration system acquires on-board images and executes the calibration process. Upon receiving a calibration command or during periodic automatic startup, it traverses all LCPG deflection states, acquires calibration images, calculates the actual diffraction efficiency and beam pointing deviation for each state, and updates the LUT in real time to achieve adaptive calibration. The communication system enables communication with the satellite's main control computer. It implements a communication protocol stack with the satellite platform, receives remote control commands, and sends telemetry images and attitude data.

[0079] like Figure 5 The diagram shows a schematic of an FPGA integrated system. The FPGA master state machine transmits and receives global state signals. The LCPG unit drive control module generates instructions for the external DAC to drive the LCPG unit based on the current scan state. The image acquisition and processing module receives CMOS data and performs preprocessing. The temperature control module calculates the TEC control quantity based on temperature sensor data using a digital PID controller. The calibration function and process control module initiates and executes the calibration algorithm when needed. The satellite information communication module interacts with the external system. The self-detection and closed-loop feedback module determines the module's operating status based on the feedback data from all modules. All modules work in parallel and exchange data via an internal bus.

[0080] Image registration and fusion unit 5 includes image registration algorithms and image stitching algorithms. Based on the driving state of the LCPG, it automatically registers multiple high-resolution sub-images acquired in time-division format and fuses them into a complete, seamless, ultra-large field-of-view high-resolution synthetic star map. Using the system point spread function (PSF) and distortion model obtained from the calibration module, it performs digital deconvolution and centroid correction on the stars in the synthetic star map, eliminating fixed aberrations introduced by the optical system and LCPG, ultimately achieving sub-pixel-level star centroid positioning accuracy.

[0081] like Figure 6 The diagram illustrates the image registration and fusion algorithm. Upon receiving the acquisition command, the FPGA initiates the calibration algorithm and receives the image acquired by the FPGA, including the sub-image sequence and corresponding LCPG status codes. First, each sub-image is corrected using a pre-calibrated geometric distortion correction model on the ground to eliminate distortions caused by optics and LCPG. Then, an image registration algorithm calculates theoretical coordinates based on the LCPG theoretical diffraction angle and combines them with image features to perform coarse registration between the sub-images. Next, sampling is performed in the overlapping registration areas for feature extraction. If a sub-image of the same scene successfully matches grayscale near a feature point, the feature point count is incremented, and feature points in other areas are searched. If the number of feature points is ≥3, the loop exits. Finally, an image stitching algorithm seamlessly fuses all registered sub-images into a complete, large-field-of-view global fused image and outputs it.

[0082] like Figure 7 The diagram shows the initialization calibration algorithm for the LCPG unit. Upon receiving the calibration command, the FPGA first initiates calibration, initializing the voltage signal applied to the LCPG unit (setting it to 0). Then, the LCPG unit drive controller begins traversing all drive voltages in a certain step (). For each voltage state, the CMOS sensor acquires a calibration image until all drive voltages have been traversed (exceeding the maximum drive voltage V). max The Tennengrad gradient method is used to calculate the sharpness of each image under different voltages, and the precise centroid coordinates and sharpness curve of the reference spot are calculated. The curve theoretically has two maxima, and the driving voltage corresponding to the maxima is indexed as the optimal voltage for the two fields of view. These measured values ​​are mapped to theoretical values ​​and driving voltages to generate or update a voltage lookup table (LUT). This LUT will be used for feedforward compensation during normal operation.

[0083] like Figure 8 The diagram shown is a schematic of a temperature closed-loop control algorithm. The FPGA receives a temperature control command and starts the temperature control program. The PID controller continuously calculates the temperature setpoint T based on the temperature feedback data it reads. set With temperature feedback value T feedbackThe error between them determines whether it meets the error threshold T. Tol If the condition is met, the temperature control program is paused. A temperature control parameter, or control quantity u, is calculated in real time based on a proportional (P), integral (I), and derivative (D) algorithm. This control quantity u is converted into a temperature control signal applied to the TEC via the TEC drive circuit, which specifies the magnitude, direction, and duration of the current. This signal heats or cools the controlled object to counteract disturbances and stabilize its temperature near the set value.

[0084] like Figure 9 The diagram illustrates the time-division multiplexing image acquisition method of this system. After receiving the command to acquire the full field of view image, the FPGA initiates time-division multiplexing. The FPGA first initializes the LCPG unit drive voltage. Based on the calibrated field-drive voltage LUT table, it applies the drive voltage to deflect the horizontal and vertical fields of view to their outermost positions, acquiring the first image. Before each drive voltage is applied to the acquired image, a wait for liquid crystal response is required; this wait time is typically 10ms at room temperature. After receiving the feedback data indicating the completion of one sub-field of view image acquisition, the FPGA searches for the corresponding drive voltage for the next horizontal angle in the LUT table, applies it, waits for liquid crystal response, and then acquires the second sub-field of view image. This process is repeated until the horizontal sub-field of view is traversed and acquired. Next, the FPGA searches for the corresponding drive voltage for the next vertical angle in the LUT table and applies it. The horizontal field-drive voltage traversal and image acquisition process is repeated until the vertical sub-field of view is traversed. The FPGA assigns numbers to all sub-field of view images included in this full field of view and stores them for later retrieval.

[0085] like Figure 10The diagram shows a schematic of the radiation-resistant shell structure. This is primarily achieved through material and structural design. A liquid crystal mixture with high resistance to total dose effects, verified through irradiation experiments, is selected. High-density materials such as tungsten and tantalum are incorporated into the packaging substrate of the LCPG device, or a local shield is designed on its exterior to attenuate high-energy particle radiation. The driving circuit is designed with voltage redundancy backup; when radiation causes the liquid crystal threshold voltage to drift, it can automatically switch to a higher voltage output level to compensate for performance degradation. The radiation-resistant shell structure is a form of local shielding. The LCPG unit is encapsulated in a special substrate containing a three-layer structure. The outermost aluminum alloy layer 19 and polyester film layer 20 serve as impact-resistant material layers. Simultaneously, an indium tin oxide conductive coating 21 serves as a metallic shield around or above the LCPG unit. The innermost layer is a polyethylene metal composite material 22, serving as a secondary particle protection layer. When high-energy radiation particles attempt to bombard the LCPG unit, they must first penetrate these three layers. During this process, charged particles are attracted to and collide with the metal layer, losing energy; secondary particles collide with the heavy metal layer, also losing energy. Ultimately, the energy and number of particles reaching the LCPG liquid crystal material are greatly reduced, mitigating the damage of ionizing radiation to the liquid crystal material and delaying performance degradation.

[0086] like Figure 11 This is the overall design structure, showcasing the invention—a large field-of-view, high-resolution star sensor device based on LCPG. Starlight sequentially passes through the LCPG unit module 23 and the main optical imaging system and sensor module 25. The FPGA 26 controls the LCPG unit module 23 to acquire images from different fields of view, imaging different regions of the large field of view onto the main optical imaging system and sensor module 25 in a time-division manner. Simultaneously, the FPGA 26 reads data from the temperature sensor and the TEC temperature control 27, runs an algorithm to control the TEC operation, and maintains a constant temperature for the LCPG unit module 23 within the insulated cavity. After processing all the data, the FPGA 26 outputs image information through the satellite interface. The radiation-resistant outer shell 24 provides protection for the entire module.

[0087] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A star sensor system based on a liquid crystal polarization grating (LCPG), characterized in that, This includes the LCPG receiving optical path module, main optical lens, radiation-resistant housing structure and detector module, and FPGA signal processing module; The LCPG receiving optical path module includes at least one polarizer and two LCPG units. The first LCPG unit consists of a polymer grating and a liquid crystal variable delay device (LCVR) to adjust the incident light to two different diffraction directions. The polarizer and LCVR have the same polarization adjustment direction. The second LCPG unit is perpendicular to the polarization direction of the first LCPG unit, forming a two-dimensional adjustable field of view. The main optical lens and detector module are used to detect starlight; The radiation-resistant shell structure is used to protect against impact, isolate charged particles, and isolate secondary particles. The radiation-resistant shell structure includes a multi-layer polymer and heavy metal structure, with an outer layer that is impact-resistant and heat-insulating, a middle layer that absorbs charged particles, and an inner layer that is insulating and absorbs secondary particles. The FPGA signal processing module includes a system status self-detection unit, a temperature control unit, an LCPG imaging calibration unit, and an image registration and fusion unit. The temperature control unit is used to monitor and regulate temperature, which is achieved through the temperature sensor and TEC of the LCPG imaging calibration unit. The FPGA signal processing module controls the LCPG receiving optical path module in real time through the system status self-detection unit and the temperature control unit. The FPGA signal processing module uses time-division acquisition when acquiring images, and the image data processed by the FPGA is then performed by the image registration and fusion unit and output. The specific implementation process of the LCPG imaging calibration unit is as follows: Step 1: Initialize the LCVR voltage on the FPGA to zero; Step 2: Start by iterating through the LCVR's driving voltage in small voltage steps, and acquire an image each time a different voltage is applied; Step 3: Calculate the image sharpness curves under different voltages using the Tennengrad gradient direction; Step 4: Index out the driving voltages corresponding to the two maxima of the sharpness curve, and use them as the field-of-view driving voltages corresponding to the two diffraction angles of the first-stage LCPG unit; match the measured values ​​with the theoretical values ​​and driving voltages to generate or update a "voltage-deflection angle" lookup table (LUT), which is stored in the FPGA; the process of updating the lookup table LUT is as follows: when a calibration command is received or when it is automatically started periodically, traverse all LCPG deflection states, acquire calibration images, calculate the actual diffraction efficiency and beam pointing deviation of each state, and update the LUT in real time to achieve adaptive calibration; The specific implementation process of the temperature control unit is as follows: Step 1: The system monitors the temperature. When the temperature exceeds the set range, the FPGA activates the temperature control unit. Step 2: The PID controller calculates the error between the temperature setpoint and the monitored temperature feedback value, and outputs the current and voltage values, directions, and time parameters applied to the TEC; Step 3: Apply temperature control signal to FPGA; Step 4: Repeat steps 1-3 until the temperature is within the set range, then pause temperature control; The specific implementation process of the image registration and fusion unit is as follows: Step 1: Switch between the two different fields of view and acquire images based on the voltages provided by the LCPG imaging calibration unit; Step 2: Optimize aberrations based on camera intrinsic parameters and distortion parameters; Step 3: Calculate the pixel positions on the image based on the diffraction angle of the LCPG, and coarsely register the image; Step 4: Sampling feature points in the overlapping areas of images from different fields of view, and performing grayscale curve matching; Step 5: Repeat step 4 until the number of feature points exceeds the set threshold. Based on the feature point positions, finely register and fuse the images, and output the image. The specific steps for the FPGA signal processing module to acquire images in a time-division manner are as follows: Step 1: Initialize the LCPG drive voltage on the FPGA; Step 2: The FPGA loads the first field-of-view driving voltage according to the driving voltage LUT table corresponding to different fields of view calibrated by the LCPG imaging calibration unit, waits for the liquid crystal response time, and at the same time, the FPGA acquires the first image data. Step 3: The FPGA finds and applies the driving voltage corresponding to the next field of view in the horizontal direction. After waiting for the liquid crystal response time, it acquires the second image data. Step 4: After traversing the horizontal field of view, the FPGA finds and applies the driving voltage corresponding to the next vertical field of view. After waiting for the liquid crystal response time, it acquires image data and repeats Step 3. Step 5: After traversing all fields of view, the FPGA assigns image numbers to different fields of view based on the fields of view corresponding to the driving voltage in the LUT, and inputs them into the image registration and fusion unit.

2. The star sensor system based on a liquid crystal polarization grating (LCPG) according to claim 1, characterized in that, The liquid crystal variable delay device (LCVR) adjusts only the adjustable polarization direction to achieve the phase delay function.

3. The star sensor system based on a liquid crystal polarization grating (LCPG) according to claim 1, characterized in that, The LCVR acts as a front element to modulate the incident light; the polymer grating acts as a rear element, providing different but symmetrical diffraction angles for left-handed and right-handed circularly polarized light.

4. The star sensor system based on a liquid crystal polarization grating (LCPG) according to claim 1, characterized in that, The LCPG receiving optical path module can add more LCPG units with different diffraction angles in the two-dimensional direction to expand the field of view.

Images