Dynamic Synchronization of Spacecraft Modeling and Coordinated Control of Leader-Follower Spacecraft Formations

The work reported in this thesis proposes several solutions to coordination control of relative motion between spacecraft in a leader-follower formation. In this framework , detailed models of spacecraft relative translation and relative rotation in elliptical Earth orbits are derived, based on the principles of Newtonian mechanics and Euler's moment equation, and subsequently collected in one 6DOF motion model. Models and mathematical relations for several perturbing orbital forces are also presented, and dynamical effects of such perturbations on uncontrolled spacecraft formations are thoroughly discussed, and partially visualized through simulations. The relative 6DOF motion model is shown to resemble models of the general class of systems known as Euler-Lagrange systems, leading to a motivation for adopting established results within this particular class to solve the leader-follower formation control problem. The formulated control problems are solved separately for relative translational and rotational motion, and these results are further extended to solve the relative 6DOF motion control problem. The results comprise both state feedback and output feedback control solutions, and the state feedback approaches are especially motivated by well-known results for the general Euler-Lagrange class of systems. In particular, two passivity-based approaches are utilized, namely the passivity-based PD+ control solution of Paden & Panja, and the passivity-based sliding surface controller of Slotine & Li. In addition, a solution for the state feedback control problem is derived from the concept of integrator backstepping, and this solution is also extended to a PID control solution by inclusion of integral action in the control loop through augmentation of the system states. The output feedback control problem is solved by conceptually different solutions for the relative translational and relative rotational motion cases, whereas the relative 6DOF motion control problem uses a combination of both approaches. The joint designation for both solutions is the use of lead approximate differentiation filters for estimation of relative velocity and relative angular velocity; a favorable solution compared to the use of full-state model-based observers. The proposed output feedback control solution for relative translational motion is based on the PD+ control solution of Paden & Panja, using feed-forward from reference velocities and accelerations, and is presented for three different levels of knowledge with respect to leader and follower orbital parameters. The first case assumes that leader orbital parameters such as true anomaly rate and rate of change are known, together with follower orbital perturbations and a perfectly controlled leader spacecraft. These assumptions are …