Advanced Optimal Spacecraft Guidance For Terminal Phase Rendezvous and Docking
Abstract
There have been growing interest in Rendezvous and Docking (RVD) missions for on-orbit refueling, on-orbit servicing, orbital debris removal, etc. A very important requirement in RVD missions is soft-docking at the terminal phase. Moreover, as satellite applications are increasing, space is getting clustered with more satellites and debris, and hence debris avoidance is necessary. There is also a critical limitation in the sensor's Line-of-Sight (LOS) angle with respect to the target satellite. In addition, limited availability of fuel and on-board computational capability are serious limitations. Due to these challenges, the terminal phase RVD guidance is a complex problem. Solutions are proposed in this thesis using state-of-the-art computationally efficient trajectory optimization based optimal guidance techniques.
First, a Model Predictive Control (MPC) trajectory optimization technique is applied to handle debris avoidance constraints, LOS constraints, and terminal soft docking constraints. Even though the results are satisfactory, MPC is computationally more expensive for on-board applications. Hence, a computationally efficient sub-optimal guidance technique, called Impulsive Model Predictive Static Programming is applied next. Attractive features of this technique are simplicity, closed-form control update capability and real-time applicability. To handle debris avoidance and LOS constraints, I-MPSP was extended to "Constrained I-MPSP" next and applied to the same RVD problem. This guidance solution turns out to be computationally very efficient. Further, the Pseudo Spectral (PS) based variable time impulse optimal terminal guidance scheme is successfully applied, which optimizes both control application timing as well as the magnitude of the pulses, thereby increasing the efficiency of the pulses.
Finally, the results from the I-MPSP guidance have been validated in a more realistic manner from Six Degree-of-Freedom (Six-DOF) simulation studies, accounting for the vehicle dynamics and inner-loop control synthesis details. This Six-DOF simulation accounts for the location, canting and pulsating nature of the thrusters as well as other related constraints such as minimum on/off time. The design has also been validated from randomized simulations by taking various initial conditions, number of debris, different sizes of debris, various transfer time, etc. Because of the encouraging simulation results from Six-DOF studies as well as its computational efficiency, the I-MPSP guidance is recommended for implementation in practical RVD missions.