Attitude control of the Delfi-n3Xt Satellite

This paper presents the attitude control algorithms of the Delfi-n3Xt, the second satellite of the Delft University of Technology. The Delfi-n3Xt, which is equipped with Sun sensors, magnetometers, magnetorquers and reaction wheels will be three-axis stabilized with its solar panels pointing towards the Sun in the nominal mode of operation. On demand, during ground station passes that are outside eclipses, the satellite will autonomously activate a Nadir pointing mode for high gain antenna data download, and a micro-thruster three-axis pointing mode for technology demonstration will be activated. The Detumble mode, which is also the safe mode, implements an improved version of the classical B- dot controller. This work introduces an efficient approach applied in the desaturation of the reaction wheels using magnetorquers. This issue is of crucial importance due to the combination of small angular momentum storage and relative high perturbation torques. In the novel approach, the magnetorquers are continuously activated according to a priority- based logic that handles the momentum exchange wheel by wheel, rather than with an average of their momentum. Another contribution of the present work consists of the design of the successive coordinates frames that are required during the various three-axis pointing modes, such that the shortest possible angular displacements from one pointing mode to the other are allowed. Extensive simulations over several orbits show that the novel unload algorithm almost completely avoids saturation of any wheel. Looking at the saturation duration time as a measure of performance, the novel algorithm outperforms the standard one by a factor of eight. As consequence, more control torque is available and thus pointing performances are dramatically improved. For all pointing modes, the controller implements a quaternion feedback regulator that has legacy from previous satellite missions. Estimation of the quaternion and of the angular velocity is performed using innovative Kalman filtering techniques. Extensive Monte-Carlo simulations illustrate the performances of the attitude estimation and control algorithms, and show that the pointing requirement are satisfied outside eclipses and in the presence of disturbance torques.