ESA GNC Conference Papers Repository
High-precision pointing platform performance simulations for small spacecraft missions
As the growth of the small spacecraft market continues, so does the demand for high-quality data gathered from small spacecraft platforms, such as high-resolution images from Earth observation and astronomy missions or high-speed data transfer by laser communication. This demand, however, imposes stringent jitter requirements on the pointing performance in the range of arcseconds and better. While state-of-the-art Attitude Determination & Control Systems (ADCSs) of small spacecraft are reaching their physical limits, high-pointing accuracy and stability can be provided by dual-stage control system approaches with a payload-in-the-loop configuration in order to further improve the quality of the data. One concept of such a dual-stage control system approach is to combine a currently existing ADCS with a so-called High-Precision Pointing Platform (HPPP). The basic idea of this concept is to separate the low-frequency (< 10Hz) and high-frequency (> 10Hz) content of the pointing disturbances for control. While the spacecraft ADCS suppresses low-frequency disturbances, the HPPP deals with the high-frequency phenomena. This paper will describe two CubeSat scenarios that benefit from a dual-stage control system approach: (1) an Earth observation mission and (2) an astronomy mission. In both scenarios the spacecraft ADCS is equipped with a star tracker and a gyro to measure respectively the pointing error and pointing rate error, and reaction wheels to control the pointing error, as this configuration represents the majority of existing small spacecraft missions. For the Earth observation mission, a (fictive) 12U and a 24U CubeSat will be equipped with a Time-Delay Integration (TDI) line-scan camera, enforcing stringent requirements on the absolute and relative pointing error. In this scenario, different HPPP sensor configurations will be considered and analysed in terms of speed and pointing performance. One particular configuration, in which the TDI camera lines are merged, cut into small regions of interest (ROI), and fed to a correlation algorithm, is used as a reference. The disadvantage of the latter sensor output is the relative pointing error which can drift away over time. Therefore, additional information is required to have an absolute measurement, coming from star tracker data, ground control points (GCPs) within images, or a high-accurate gyroscope, This pointing performance analyses are performed within the framework of the Affordable QUALity Images from Space (AQUALIS) SBO 4-year project, funded by the Research Foundation Flanders (FWO). The consortium of this project exists of the Department of Mech. Eng. of the KU Leuven, VITO, and imec. For the astronomy mission, a 6U CubeSat is equipped with a spectroscopy instrument for which a stringent requirement on the absolute pointing error exist. In this scenario, the HPPP is composed of a fine-guidance sensor (FGS) which senses the deviation of the targeted star, and a fast-steering mirror (FSM) that deviates the light onto both the fine-guidance sensor and the payload via a dichroic beam splitter. In this configuration, the fast and accurate information of the fine-guidance sensor allows to control the fast-steering mirror. This analysis is performed within the PDR and CDR phase of the In-Orbit Demonstration (IOD) CubeSat mission called CubeSpec, funded by BELSPO, the Belgian federal science policy office. The goal of the mission is to demonstrate high-spectral-resolution astronomical spectroscopy from a 6-unit CubeSat. For both scenarios, simulations are performed first in the frequency domain (Pointing Error Engineering Tool PEET), allowing for a first estimate of the pointing performance, and time domain (Matlab Simulink), allowing for a more detailed study of the communication between the ADCS and HPPP control loops. This paper will describe both scenarios (context of the AQUALIS project and context of the CubeSpec mission), the related pointing performance requirements, the simulation models in PEET and Simulink, and their results.