Attitude determination and control for stabilization and pointing accuracy for 3u class nano-satellite
Loading...
Date
2016-04-01
Authors
Muhammad Fadly
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Research and development of 3U nano-satellite is increasing nowdays and have
been carrying various mission to space. InnoSAT is a 3U satellite developed by ATSB
and five universities in Malaysia. The satellite is developed with limited weight, power,
space and budget. The satellite mission is to capture images of Malaysia’s territory
from space. Existing system in the satellite produces unstable dynamic and unbalance
of weight. Three axis magnetorquers, sun and magnetic field sensor are installed on
satellite body axes. The satellite control torque is produced by perpendicular magnetic field from magnetorquer and Earth’s magnetic field vector which varies periodically in orbit. Three axis Micro-Electro Mechanical Systems (MEMS) gyro is installed as a primary attitude determination and optimal combination of sun and magnetic field sensor as a secondary. Accurate attitude, position and stability of the satellite in orbit while capturing the image is required. This thesis discuss about designing of accurate and fast computation of secondary attitude determination and also attitude stabilization, pointing accuracy, and disturbance rejection control for all satellite position in orbit with existing satellite configuration. The attitude is determined using five methods; TRIAD, QUEST, q-method, FOAM, ESOQ1 and a recursive EKF. The methods are chosen by considering the complexity and possibility to be able to implement in onboard micro-controller of InnoSAT. The point base methods are able to produce accurate and fast output when the satellite out of eclipse but the EKF is able to perform at all position. The q-method shows the fastest and accurate attitude determination
compare with other six methods with less than 5% error. The satellite is
controlled by inserting sliding mode into satellite model to design controller based on
linear Riccati equation; LQR, H2, H¥, m-synthesis and LPV based on Linear Matrix
Inequality (LMI). The closed loop is analyzed due to variation of weighting functions,
uncertainties, and number of feedback at any position in orbit. The controller
power consumption is highly considered to see the possibility to implement in existing
system. The controller is able to maintain the stability only in small radius position
from its design point but if the satellite go beyond the radius the new controller must
be recalculated. The continuous stability can be achieved by LPV controller, but the
controller unable to maintain stability, pointing accuracy, and disturbance rejection simultaneously
at the same time. The attitude is represented in Two Dimensions (2D)
and Three Dimensions (3D) simulation using Simulink MatlabTM.