Aerospace and Electronic Systems Magazine June 2017 - 16

Feature Article:

DOI. No. 10.1109/MAES.2017.150198

An Innovative Test Bed for Verification of Attitude
Control System
A. Tavakoli, Maleke Ashtar University of Technology, Karaj, Iran
A. Faghihinia, Sharif University of Technology, Tehran, Iran
A. Kalhor, University of Tehran, Tehran, Iran

INTRODUCTION
Experimental validation of the attitude determination and control
subsystem performance is an essential requirement for new satellites. It can provide a very low cost and low risk environment for
evaluating the proposed design before launch. One of the greatest
difficulties in developing an Attitude Determination and Control
Subsystem (ADCS) test bed is to perform the ground-based experiments in different environments to provide the orbital conditions
experienced by a real spacecraft. Simulating the space environment is not an easy task and also testing of the attitude determination and control is very complicated. The essential requirements
of functional tests in ADCS hardware are 1) rotational frictionless
free motions, 2) simulating orbital environment for sensors and
actuators, 3) software implementation of algorithms, and 4) finally
integrating of these parts. Three degree of freedom (DOF) air bearing based simulators are the most common method for simulation
of a frictionless media. Air bearing offers a nearly torque-free condition, perhaps as close as possible to that of space and hence it is
a preferred technology for ground-based research in space attitude
dynamics and control, although it certainly cannot provide the full
experience of microgravity. There are several simulators around
the world that are developed for many different education and
research purposes. One of these simulators is developed for testing the bifocal relay mirror spacecraft system. The platform hosts
several satellite subsystems, including rate gyros, reaction wheels,
thrusters, sun sensors, and an onboard computer for attitude control. An ordinary source of light simulates the sunlight for the sun
sensor, which provides a low accuracy attitude determination [1].
Another simulator is developed for the Adaptive Reconnaissance

Authors' addresses: A. Tavakoli, Space Research Center,
Maleke Ashtar University of Technology, 25th km of TehranKaraj Highway, Karaj, 3157693365 Iran. E-mail: (am_h_tavakoli@yahoo.com); A. Faghihinia, Sharif University of Technology, Tehran, Iran; A. Kalhor, Control and Intelligent Processing
Center of Excellence, School of Electrical and Computer
Engineering, University of Tehran, Tehran, Iran.
Manuscript received January 30, 2016, revised June 21, 2016,
and ready for publication July 20, 2016.
Review handled by A. Franchi.
0885/8985/17/$26.00 © 2017 IEEE
16

Golay-3 Optical Satellite (ARGOS) project, which is composed
of many sensors in a nonreal environment used for attitude determination [2]. A 3 DOF platform was constructed in the Georgia
Institute of Technology to perform new control strategies in an experimental framework. The reported tests are about the evaluation
of attitude controllers. Also in this simulator, there isn't any sensor
used by a real satellite for attitude determination [3]. The twin distributed spacecraft attitude control system simulators (DSACSS)
are designed to implement several distributed control laws for a
specific formation of satellites in Virginia Tech's Space Systems
Simulation Lab. The attitude sensors and algorithms in these simulators are not real as those that are used in the satellites [4]. Some
of these simulators include the real type or a prototype of satellite
attitude sensors. One of them is a small simulator developed in the
Universidad Nacional Autonoma de Mexico (UNAM center). The
platform includes a set of four static Earth sensors, inertial measurement unit (IMU), sun sensor, and magnetometer, which are
used as attitude sensors, but the attitude determination has many
serious differences in environmental conditions with a real satellite
[5]. Another simulator was developed in Changchun Institute of
Optics for spacecraft attitude control research and simulation. The
attitude determination system consists of three fiber optic gyros, an
inclinometer, and a magnetometer. The inclinometer is not utilized
as an attitude sensor and then it cannot be used in a satellite directly [6]. In [7] for attitude determination, the magnetometer is used
in conjunction with the accelerometer as the two sensor inputs to
the TRIAD algorithm. But the accelerometer is not an appropriate
choice for satellites in space. An adaptive unscented Kalman filter
is used to compensate for drift and noise of the only analog gyroscopic rate sensors for attitude determination in [8]. So, the real
condition for attitude determination with reference sensors is not
satisfied. Coarse sun sensors, cameras, and IMU are used for attitude determination of EyasSAT for HIL application [9]. The EKF
(extended Kalman filter) algorithm was the main attitude estimation algorithm used. The TRIAD and QUEST algorithms were also
implemented on the Microcontroller Unit (MCU). Nonetheless,
the reference vectors are stationary in the room and don't vary with
time. So the evaluation tests are performed in nonreal conditions. A
nanosatellite HIL setup uses the combination of sun sensors, IMU,
and metrology system for attitude determination of the platform
at the Spacecraft Robotics Laboratory of the Naval Postgraduate
School [10]. In fact the IMU and metrology system are not space

IEEE A&E SYSTEMS MAGAZINE

JUNE 2017

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine June 2017