Aerospace and Electronic Systems Magazine June 2017 - 18

Test Bed for Verification of Attitude Control System

Figure 1.

Schematic diagram of the setup.

Figure 2.

Flowchart of the hardware in the loop procedure.

magnetic field simulator with a magnetometer and electronic units for
data acquisition, data communication, and algorithm implementation.
An air bearing based attitude simulator is equipped with reaction wheels and simulates weightlessness condition. This table is
employed for evaluation of the attitude control performance. An
AHRS (attitude and heading reference system) sensor was chosen as the attitude sensor unit. The AHRS provides attitude angles
18

and angular velocity of the table. A two-sided wireless communication link between the processor of the
air bearing simulator and the central processor is employed for data transmitting.
The attitude determination part of the setup
consists of two stands for simulating sun and magnetic vectors. These are simulated using attitude data
(AHRS), the orbital position of the satellite, and simulator models.
The operating condition of the sun sensor is provided by a sun simulator and a 2 DOF table, which is
controlled by two stepper motors. The motor drivers
receive commands (azimuth and elevation angle) via
a USB port for the central processor. The azimuth and
elevation commands are used to reorient the sun sensor against the sun simulator in the darkroom to create
the sun vector, which is experienced by a hypothetical
sensor in a typical satellite in a specific orbit along the
time. Similarly a 3-axis Helmholtz coils simulates the
environment orbital condition for the magnetometer.
Three coil drivers receive current commands from the
central processor via an RS485 interface to provide the
desired magnetic field. The commands are generated
so that the magnetometer inside the Helmholtz cage experiences the
magnetic vector, which is found by a hypothetical magnetometer
of a typical satellite in a specific orbit along the time. In view of
the fact that only the magnetometer is located inside the Helmholtz
cage, a very small cage is adequate for reproducing the controlled
magnetic field around the sensor. For typical tests two sensors have
been used: a 2-axis analog sun sensor and a 3-axis magnetoresistive
magnetometer. A 16 bits I/O card is used for data acquisition of the
sensors. The following flowchart (Figure 2) shows the main steps
of the HIL procedure. In fact, evaluation of different attitude determination and attitude control algorithms is one of the uses of this
test bed that can be executed in step 5 of the flowchart. We have two
independent measurements: sun vector and magnetic vector. So we
can employ most of the attitude determination algorithms, including deterministic and estimation methods.
We implemented a deterministic algorithm (QUEST) and an
estimation algorithm (EKF) for attitude determination. We just
used QUEST for initialization of the EKF algorithm. Therefore
this algorithm is activated in 15 s from the beginning of the test.
Also a linear controller based on the quaternion error feedback is
used for attitude control. These are routine algorithms in many satellites and here they are used as in the satellites in space. So, the
algorithms use sensor measurements, computed reference vectors,
and the simulator (satellite) dynamic model for attitude estimation
and control. Figure 3 shows the algorithms framework and the inq 
formation used in the algorithms. In the figure, m =   is the state
ω 
vector that includes a quaternion and an angular velocity vector.
The state vector is estimated by the EKF algorithm and used in the
control algorithm.
More details about the modeling and equations of the implemented determination and control algorithms can be found in most
of the ADCS references such as [17]-[18].

IEEE A&E SYSTEMS MAGAZINE

JUNE 2017

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