Star sensor is an attitude measurement device that uses stars as a reference source. Due to its advantages such as good autonomy, good passive measurement concealment, strong resistance to electromagnetic interference, high accuracy in attitude determination and orientation, and no accumulation of errors over time, it has become an indispensable attitude measurement device for platform attitude and orbit control systems such as satellites, missiles, ships, and aircraft.
The research on star tracker technology began in the 1950s, and early frame based star trackers based on photomultiplier tubes were successfully applied on platforms such as ships and missiles; The emergence of Charge Coupled Device (CCD) and the development of integrated circuits in the 1970s led to the rapid development of fixed probe star sensor technology; In the 1990s, with the increasing maturity of large-scale integrated circuit technology and CMOS (Complex Metal Oxide Semiconductor) processing technology, CMOS APS (Active Pixel Sensor) star sensors were successfully applied on platforms such as satellites. In short, star sensors are widely used in extraatmospheric space environments and platforms with clear night skies inside the atmosphere. However, in the atmosphere of near-Earth space, the development and application of all day star sensor technology are limited due to the interference of strong sky background light during the day.
The star sensor uses stars as a reference source, undergoes steps such as photoelectric imaging, star point extraction, centroid calculation, star map recognition, and attitude determination to ultimately obtain attitude information. The working principle of the all-time star sensor in near-Earth space is shown in Figure 1. Unlike onboard star sensors, all day star sensors in near-Earth space operate in the atmosphere. On the one hand, the atmosphere attenuates radiation from stars, leading to signal attenuation; On the other hand, the atmosphere scatters solar radiation, as well as thermal radiation from the atmosphere and surface, forming sky background radiation. Therefore, improving the contrast of stellar backgrounds and detecting dim stellar targets under strong daytime light is a key issue for all day star sensors.
In the 1950s, countries such as the United States began researching the technology of day and night star trackers. At present, countries such as the United States and Italy have developed various models of all day star sensors for applications on platforms such as ships, ground telescopes, airplanes, and high-altitude balloons. Below is an introduction to representative products and applications abroad.
In 2005, Microcosm Corporation of the United States developed MicroMak micro star sensors for spacecraft platforms; In March 2006, Microcosm announced that the system could capture stars of magnitude 7.1 at sea level during the day; During the observation period, almost all stars can be observed, and the attitude and navigation can be continuously updated.
The Italian National Institute of Astrophysics’s School of Radio Astronomy has developed an ST (StarTracker) star tracker, which is installed on a 32m radio telescope and aims to establish an accurate pointing model and dynamically track the antenna.
Since the 1960s, aircraft platforms have been using star assisted inertial navigation systems [4-8], such as the American ANS astronomical/inertial navigation system, NAS series navigation system, and LN-120G high-precision integrated navigation system.
1) NAS-26 Star Tracker in the United States
Northrop Grumman Company in the United States began the development of astronomical/inertial navigation systems in the 1950s, and successfully conducted flight tests on the fourth generation navigation system NAS-26 in 1977.
2) LN-120G Star Tracker in the United States
The research on astronomical inertial navigation systems by the American company Litton (acquired by Northrop Grumman in 2001) began in 1961 and developed a daytime star tracker in 1963. In 1975, the US Air Force began using the LN-20 system. In 2006, the system was upgraded to the LN-120G integrated navigation system. In May 2007, Northrop Grumman delivered the first LN-120G production system to the US Air Force, and later delivered 30 sets of the system for equipping RC-135 series aircraft. The LN-120G navigation system can track stars day and night, improve the position information of inertial navigation using star information, and provide a heading accuracy of up to 20 ″ for aircraft; The positioning accuracy of satellite inertial mode is 900m/h (CEP).
High altitude balloon platform is a low-cost and reliable scientific research platform. Both balloon platforms and payloads require high-precision attitude information throughout the day
1) HERO Star Camera in the United States
At the end of the 20th century, NASA’s Marshall Space Flight Center developed a star camera for the new generation of hard X-ray telescope HERO (High Energy Replicated Optics) system, which was used for balloon experiments at altitudes of 35-42 kilometers
2) BLAST Star Camera in the United States
In the early 21st century, several institutions, including the University of Pennsylvania in the United States and the University of Toronto in Canada, jointly developed the Balloon Platform Large Aperture Submillimeter Telescope BLAST (Balloon Bore Large Aperture Submillimeter Tele scope) with funding from NASA. It used a pair of redundant star cameras ISC and OSC for precise positioning.
3) EBEX Star Camera in the United States
EBEX (The EndBExperience) is a balloon platform telescope used to detect polarized signals in the cosmic microwave background. The attitude control system of EBEX includes two redundant satellite cameras XSC0 and XSC1, which are used to achieve absolute and high-precision measurement of three-axis attitude. Several units, including Columbia University in the United States, JPL, and the Rutherford Laboratory in the United Kingdom, jointly designed this type of star camera.
4) WASP Platform Star Sensors in the United States
NASA’s Wallops Arc Second Pointer has developed a balloon borne angular second precision orientation platform called WASP (Wallops Arc Second Pointer). Between 2011 and 2014, a total of 5 test flight tests were conducted to verify the performance of the DayStar star sensor and the CARDS (Celestial Attention Reference and Determination System) star sensor on the WASP platform at an altitude of 32km.
1) OWLS star sensor
As early as the late 1980s, NorthropGrumman Company in the United States proposed an optical wide angle lens star tracker (OWLS) scheme, which uses holographic lenses for its optical wide angle lens [16-17]. OWLS uses holographic optical element elements (HOE) formed by multiple exposures in the holographic film to obtain multi field telescopes, each of which can receive signals at full aperture. This lens system can be processed through photolithography technology.
2) Mini OWLS
In 1993, NorthropGrumman Corporation of the United States proposed a conceptual design for a miniature optical wide-angle lens star tracker (Mini OWLS), as shown in Figure 13. The function of the Mini OWLS star tracker is to measure attitude drift on three vertical carrier axes. It consists of three wide field Schmidt telescopes, which share one multiplexing HOE and can simultaneously detect multiple starlight from different directions without a rotating mechanism.
3) DaytimeStellarImager, USA
In 2008, TrexEnterprises in the United States proposed an all day star camera for day and night navigation by observing the K or H-band near-infrared light of stars, and provided two preferred solutions, as shown in the figure [19]. One is a multi aperture scheme, which uses three relatively large aperture telescopes rigidly installed on the carrier platform, with a 45 ° angle to the vertical direction and a 120 ° angle to each other, and a 0.4 ° field of view × 0.5 °; Using 3 320 × The InGaAs detector camera with a 256 plane array achieves synchronous measurement. Stars with brightness greater than 6.4 (H-band) are usually available, and the positioning accuracy of the navigation system can reach 30m. The equipment in this scheme is relatively large and heavy, with a volume of about 1m3 and a weight of about 27kg. However, it does not have moving parts and has high reliability, making it suitable for platforms such as ships and airplanes. The second option is a small aperture tracking scheme, where the telescope aperture is usually less than 100mm. To solve the problem of reduced signal-to-noise ratio and weakened detection ability caused by reduced aperture, two technologies can be used for compensation: one is to use a two axis precision turntable to achieve a 15 ° × Bright star observations within a 15 ° range; The second option is to use detectors with low readout noise, which have a short integration time and become the main source of noise. This scheme is more complex in mechanical structure than the multi aperture scheme, but has a smaller volume and lighter weight, making it particularly suitable for navigation and guidance systems of aircraft and missiles.
The biggest challenge faced by all time star sensors in near-Earth space is how to suppress the sky light background and noise under limited platform resources, thereby achieving the detection of dim star targets. From the current research status at home and abroad, it can be seen that all day star sensors mostly use the following technical means:
1.Spectral filtering technology
According to Rayleigh scattering theory, the scattering intensity of air molecules is inversely proportional to the fourth power of wavelength. Therefore, the spectral characteristics of atmospheric scattering, solar radiation, and stellar target radiation can be utilized to filter the spectrum, select the optimal working band, suppress atmospheric scattering background light while retaining starlight, thereby improving the contrast of stellar background and signal-to-noise ratio of stellar detection. At present, the working wavelengths used by all day star sensors are mainly concentrated in two bands. Firstly, stars with peak wavelengths near 800nm, such as G-type, K-type, and M-type, are selected as navigation stars in the 600-1000nm band; Correspondingly, the detector is selected as a CCD or CMOS sensor that still has strong spectral response in the near-infrared band. The second is the 900~1700nm SWIR band, which further suppresses the background; Select the J-band (1.24) of the 2MASS (The Two Micron All Sky Survey) catalog μ m) Or H-band (1.65 μ m) A bright star; Select InGaAs detectors as sensors accordingly. In practical use, by combining the distribution law of sky background radiation brightness, atmospheric transmittance curve, detector spectral response curve, and navigation star spectral distribution curve, the above working bands can be further optimized. The vast majority of practical models of star sensor products abroad, such as ground-based, sea-based, airborne, and high-altitude balloon platforms, use the 600-1000nm band and have achieved good results in practical applications; Domestic all time star sensors have corresponding products available in both operating bands. In addition, domestic demand for services based on 2.0~2.5 μ Some research work has also been carried out on the optical lens design of m-band daytime star sensors.
2.Method of reducing the solid angle of pixels
The radiation angle of stars is generally less than 0.01 ″, and stars can be regarded as point light sources, and starlight can be approximated as parallel light; The sky background, on the other hand, radiates in all directions and can be regarded as a limited far-field light source. Therefore, the difference in radiation angle between stars and the sky background can be utilized to reduce the stereo angle of imaging pixels through optical means, achieving further suppression of the sky background on the image plane. Constrained by the pixel size and number of pixels in the detector, the stereo angle of the pixels can be reduced by using a small field of view and long focal length optical lens. When the sky light background is bright enough, a sufficient number of stars cannot be detected in a single small field of view for star map recognition. A “small field of view+mechanical scanning” method can be used to scan, capture, and track stars in a large airspace range, which is called a star tracker, such as NAS-26 and LN-120G astronomical navigation systems. The star tracker using the “small field of view+mechanical scanning” method can easily detect navigation stars, but due to the fact that there is only one star in the instantaneous field of view, it requires an inertial navigation system to guide it, resulting in poor autonomy and relatively low reliability.
3.Multi field of view detection technology
To eliminate the dependence of star trackers on inertial navigation systems, a multi field of view detection scheme can be adopted: while ensuring the detection ability of a single small field of view, multiple fields of view are used to increase the probability of detecting multiple stars simultaneously, in order to obtain the number of navigation stars that meet the matching requirements and achieve fully autonomous attitude measurement, such as the three field of view detection scheme of the DayStar star sensor in the United States. However, a multi field of view detection scheme requires each field of view to have a certain size, otherwise the probability of star detection cannot be guaranteed, and therefore the volume and weight are relatively large. Shared aperture optical and structural design schemes can be adopted to achieve the goal of reducing volume and weight, such as the optical design ideas of Micro Mak, OWLS, and Mini OWLS.
4.Spectral polarization imaging technology
Atmospheric scattered light mainly comes from the scattering of atmospheric aerosol particles when sunlight passes through the atmosphere. According to the Rayleigh molecular scattering theory, when the incident light is natural light, the scattered light has a high degree of polarization, especially when the scattering angle is around 90 °, the degree of polarization of the sky light is 100%; At other scattering angle positions, the polarization of the sky light varies between 0% and 100%; Atmospheric polarization is greater in areas with scattering angles greater than 40 °. The polarization of stellar light is relatively weak, generally much lower than the polarization of atmospheric scattered light. Therefore, during the day, the polarization image information of stars includes direct radiation component information of star targets and atmospheric scattering component information. By utilizing the difference in polarization characteristics of scattered light between stars and atmosphere, the contrast of star background can be improved, thereby increasing the probability of star detection. The limitation of this method is that when the angle between the sun and stars is small, the polarization of atmospheric scattering is small, which is not conducive to the extraction of target signals. In practical applications, navigation stars usually select stars with a solar angle greater than a certain angle (such as 30 °).
5.Image noise suppression algorithm
Although the use of spectral filtering technology and methods such as reducing pixel stereo angle have filtered out most of the background light, the star images still exhibit uneven background grayscale distribution, complex background changes, and interference stars due to the complex lighting conditions of the working environment of the star sensor throughout the day. Short wave infrared images also contain severe stripe non-uniformity noise and defective pixels, which are not conducive to star point extraction and centroid calculation accuracy. For star trackers, background subtraction is often used to suppress noise. Usually, the optical system is controlled to point in the direction near the target star (such as deviating by 1 °) to obtain an average background image; Afterwards, point to the target star in a short period of time, take a picture and obtain the star point image minus the average background; Subtract the average values of multiple pixels distributed around the field of view (such as the four corners of the image) to further suppress the residual background; Finally, star points in the field of view can be extracted and centroid calculated using conventional methods. For fixed probe star sensors, different detection sensitive units will bring different characteristics of noise. Multiple frame stacking methods are often used to suppress non correlated random noise. The noise suppression algorithm using one-dimensional feature point descriptors based on adjacent pixel information can adaptively and effectively suppress the influence of stripe non-uniformity and defective pixels in shortwave red extraterrestrial images, which is conducive to achieving higher star centroid positioning accuracy.
The Development Trend of All Time Star Sensor Technology
1.Astronomical/inertial deep integrated navigation technology.
The traditional combination method of star sensors and inertial navigation belongs to a simple combination method, where inertial navigation operates independently, and the output information of star sensors is used to correct inertial navigation errors. The near-Earth space all day star sensor can consider deep integrated navigation with inertial navigation information: inertial navigation provides initial pointing and high-precision incremental output information, assisting star sensors in star point extraction and recognition processing; At the same time, the original star map information and matched star image point coordinates of the star sensor before and after the time can be used to calibrate errors such as inertial navigation constant drift; Finally, the two types of output information are fused and processed to output high-precision attitude angle and angular velocity information at high frequencies, thereby achieving sensor level integrated navigation.
2.All day, all autonomous matching star sensor technology.
Compared with star trackers with rotating mechanisms, matched star sensors have advantages in autonomy, reliability, and measurement accuracy. The sensitive technology of light and small fully autonomous main satellites suitable for near-Earth space, especially airborne platforms, is currently a major challenge and important research direction. Introducing micro/nano optical design and processing technology into the field of star sensor technology, overcoming the shortcomings of traditional optical design methods, is a possible implementation method for lightweight and all day matching star sensors.
3.Environmental adaptability technology for all time star sensors in near-Earth space.
All day star sensors are one of the indispensable navigation methods for near-Earth space application platforms. However, different application platforms face different environmental characteristics and motion characteristics of different platforms. Therefore, relevant research work has been carried out on the suppression of aerodynamic optical effects, large dynamic and high and low temperature environmental adaptability technologies of star sensors; Choosing an aircraft to conduct space flight verification tests, collecting measured test data, and improving theoretical models are the basic prerequisites for the engineering application of all day star sensors.
Conclusion
The near-Earth space all time star sensors, represented by the United States, have been well applied on multiple platforms. In addition to using spectral filtering technology, these star sensors have different characteristics for products from different platforms:
1) Platforms such as sea-based and ground-based platforms are in the strongest sky background during the day, and star sensor systems are less constrained by volume and weight. They can suppress the background through optical systems with large aperture, long focal length, and small field of view.
2) The sky background of aircraft with an altitude of 30-40km is relatively dark, and a medium field of view (3 °) can be used × An optical system of around 3 °, combined with high-performance detectors, enables the detection of multiple stars in an instantaneous field of view.
3) Airborne platforms at an altitude of 8-10km are constrained by a bright sky background and limited platform resources. They often use small field of view star trackers with scanning mechanisms to achieve full-time detection of stars. Domestic research started relatively late and is currently in the testing and verification stage. Some airborne products have completed flight tests.
Send us a message,we will answer your email shortly!
TY-Space Technology (Beijing) Ltd. is professional focusing on advanced attitude optical sensors, especially Star Trackers for space industry.
