Star sensors have high measurement accuracy, large equivalent field of view, stronger resistance to stray light interference and direct sunlight, and stronger adaptability to high dynamic environments. At present, star sensors have become the most critical attitude measurement sensors for spacecraft, and are generally regarded as the attitude reference for satellites. If the polarity of the star sensor used as the reference is incorrect, it will lead to positive feedback in the control system, causing the satellite to quickly lose control or even disintegrate, posing serious risks to the satellite and space environment.
According to the closed-loop simulation test of the satellite control system’s semi physical (including whole satellite semi physical) star sensor in the loop, qualitative and quantitative verification of the satellite’s function, performance, software operation timing, etc. can be carried out, but it is not possible to test and verify the polarity of the satellite control system’s sensor and actuator. Therefore, traditional ground semi physical experiments also need to include polarity testing and verification. Polarity testing is a method of determining the polarity correctness of onboard software by observing the correctness of the trend of criterion changes through testing.
The coordinate system of star sensors generally does not coincide with the satellite coordinate system. Based on the characteristics of the star sensor system, such as multiple attitude calculation steps and long attitude determination algorithm paths, the traditional method that can be used for single machine polarity testing is no longer suitable for the polarity testing of the entire star sensor installation. In fact, the installation polarity of star sensors in satellite control system sensors and actuators is the only one that has not been directly tested.
Zhang Zhaodi et al. proposed a method for testing the polarity of star sensors through field observation. This method achieved single machine polarity testing using real star sky, but it cannot be applied to the installation polarity testing of star sensors on satellites. Ding Jianzhao et al. introduced a dynamic optical star simulator into the semi physical closed-loop simulation system of the Resource 3 satellite, achieving full optical path closed-loop testing of star sensors and improving the authenticity of star sensor testing, but did not involve star sensor installation polarity testing. In the public literature, no research has been found on the polarity testing method for star sensor installation. In the field of engineering, the polarity correctness of star sensors is generally ensured through traditional methods such as repeated theoretical calculations and multiple person reviews. Therefore, based on the hardware matching of traditional testing systems (optical star simulators), this article designs a highly versatile method for testing the polarity of star sensor installation, which verifies the engineering value of the correctness of the polarity of the star sensor installation matrix in onboard software. This article will discuss the method of using a light star simulator to test the polarity of star sensor installation, provide specific operating methods and criteria, and comprehensively analyze the feasibility, effectiveness, and coverage of this method.
In indoor environments, the optical path of a star sensor can generally be tested using a light star simulator. The light star simulator can simulate the star map of the star sensor at a three-axis 0 attitude (corresponding to the output of the star sensor in units of quaternion), which is called a standard star map. Install a star simulator equipped with a standard star map on the head of the star sensor in a nominal manner (i.e. the initial angular position of the star simulator). If the output of the star sensor is near the unit quaternion, it is said that the polarity of the star sensor is correct. In addition, the star map in the light star simulator rotates at a limited angle around the three-axis of the star sensor, which can visually confirm the correctness of the three-axis polarity of the star sensor.
The working principle of a light star simulator can be divided into two categories: the first type of light star simulator has a fixed parallel light source as the star point, and the position of the light source and the simulator is fixed. The relative attitude rotation is achieved by rotating the simulator. This type of optical star simulator has a simple structure, intuitive operation and criteria, but its disadvantage is that the inter star angular distance does not match the real star library, requiring star sensors to have star map learning function; The star point of the second type of optical star simulator is simulated by the bright spots on the high-resolution LCD screen, and the star map is driven by the computer input attitude. The advantage of this type of optical star simulator is its versatility, which can be used for polarity testing of star sensor installation and can also be driven by dynamic data to connect the star sensor optical path to semi physical closed-loop simulation tests. Its disadvantage is that there is no degree of freedom for mechanical rotation in the X/Y axis direction, and it is necessary to set a bias angle in the ground software to equivalent the rotation of the optical star simulator, which is not intuitive to operate.
Taking the first type of optical star simulator from Jena Company in Germany as an example, the main component names are shown in Figure 1.
Fig.1 Functional schematic diagram of ASTRO10star simulator
In Figure 1, D represents four parallel light sources used to simulate star points; A/B is the angle adjustment screw. Among them, A can be used to adjust the X-axis attitude, and B can be used to adjust the Y-axis attitude; C is a dial that is fixedly connected to a parallel light source. By rotating C, the attitude of the Z-axis can be adjusted; F is a mechanical interface with the star sensor light shield, used to block external light sources and excess objects from entering the interior of the star sensor light shield.
The combination of the light star simulator and ASTRO10 star sensor is shown in Figure 2.
Fig.2 Combination diagram of ASTRO10star sensor star simulator
The ASTRO10 star sensor has a star learning mode. Through the learning mode, the currently recognized star coordinates can be used as its standard star map library (the corresponding star sensor will output unit quaternions). After the end of the learning mode, the three-axis rotation of the light star simulator relative to the star sensor can be measured in real-time and used to verify the correctness of the star sensor polarity.
It should be noted that for star sensors without a learning mode, a light star simulator with a built-in real star library can be used, and different poses can be set and corresponding star maps can be called up to transmit to the star sensor for pose recognition. Simulating the rotation of the star sensor relative to the inertial space can also achieve the purpose of verifying the polarity of the star sensor.
(1) Design ideas
The polarity testing technology of star sensors is currently relatively mature, and the methods used include outfield star observation method, optical star simulator testing method, etc. The outdoor star observation method requires placing the star sensor outdoors, while the light star simulator testing method can be conducted indoors. The polarity testing of star sensor installation involves semi physical system testing and must be conducted indoors. Therefore, studying the polarity testing of star sensor installation based on optical star simulators has practical significance.
The design of the polarity testing method for star sensor installation should follow the following characteristics:
1) Easy to operate, based on the existing star sensor testing equipment and operating methods;
2) Accurate results and high certainty of polarity test results;
3) Intuitive and easy to understand, the correctness judgment of polarity testing should be intuitive and easy to understand, and easy to observe;
4) Strong universality, suitable for testing the installation polarity of satellite star sensors in various working modes.
In order to make the criteria more intuitive and easy to understand, it is necessary to correspond the initial angular position of the optical star simulator to the standard star map, and the corresponding satellite three-axis measurement attitude is all 0. Therefore, the polarity test of star sensor installation can be understood as follows: when the optical star simulator is in the initial angular position, the satellite body coordinate system coincides with the reference coordinate system. The rotation of the light star simulator along a certain axis of the star sensor can be equivalent to the satellite rotating in the opposite direction in the reference coordinate system. Ground data interpretation personnel can determine the correctness of star sensor installation polarity through the installation method of the star sensor on the entire satellite and the changes in satellite three-axis attitude.
(2) Testing Principles
The quaternion calculation algorithm for satellite attitude based on star sensors is:
The possible polarity errors include the following two aspects: star sensor polarity errors and star sensor installation polarity errors. When a star sensor polarity error occurs, and the original attitude quaternion output by the star sensor does not meet equation (2) when the optical star simulator is at the initial angular position, the star sensor polarity error can be immediately determined based on this. By comparing the changes in the original quaternion output of the star sensor with the rotation of the optical star simulator, it is possible to further locate the coordinate axis of the star sensor where polarity errors occur; When only the polarity error of star sensor installation occurs, when the optical star simulator is at the initial angular position, the three-axis attitude of the satellite is around 0. However, due to the mismatch between the quaternion of star sensor installation in the onboard software and the actual installation method of the star sensor, the three-axis attitude change trend output by the satellite will not match the expected one. Based on this, the coordinate axis with incorrect star sensor installation polarity can be located.
This article proposes a star sensor installation polarity testing method based on a light star simulator. Without adding new testing equipment, full path polarity testing can be performed on attitude determination including star sensor polarity and star sensor installation polarity. If there is an error in the polarity of the star sensor or any link in the polarity testing of the star sensor installation, it can further accurately determine the coordinate axis of the polarity error. The star sensor polarity testing method used has good operability, clear physical meaning of relevant criteria, and intuitive criteria. It can be applied to various spacecraft based on star sensor attitude determination, and has good engineering application value.
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.
