High-Performance Attitude Pointing Control for Multi-Target Acquisition in Agile LEO Scientific Satellites

This research compares the performance of Model Predictive Control (MPC) and conventional Proportional-Integral-Derivative (PID) – based controllers for pitch-axis attitude control of Low Earth Orbit (LEO) Scientific satellites. The proposed designs aim to achieve high agility for reorientation between multiple targets and stable pointing performance during science observations. To address each of the objectives, two modes of operation were designed – Rapid Slew mode and Science mode - featuring ’High-Performance’ Attitude and Orbital Control Systems (AOCS) for the observation part of the mission. The PID-based controllers were designed analytically, and for the MPC, the attitude control problem was formulated as a Convex Quadratic Optimisation Problem (QP). Similar performance was achieved with PD-based controllers and MPC, with the major difference being the controller demands. Specifically for the Science mode, the analysis shows that MPC requires significantly less torque to achieve the same performance as the PD-based one while also having the lowest Absolute Pointing Error. This study shows the potential of MPC, especially for achieving high-stability pointing in a fuel-efficient and, therefore, mass- and cost-optimal way, which would enable more complex operations for future scientific missions.