|dc.description.abstract||Mathematical models of various astrodynamical systems in the Solar system and identification of solutions within these models are crucial for the burgeoning space exploration efforts carried out by the government agencies and the private industry in the 21st century. Often
times, these mathematical models are highly nonlinear and coupled. Closed-form analytical solutions to these systems seldom exist, thus forcing the human experts to use numerical methods to identify solutions. However, random initial-conditions or guesses to these numerical
methods do not yield desired orbital motion solutions. And hence the identification of good initial-guesses became a problem in itself in this realm of engineering. There is a great need for novel techniques to identify good initial guesses to these powerful numerical methods. With
this dissertation work, I leverage the virtual reality platform to help human operators derive good initial guesses for trajectory design problems in the cislunar space. The real three dimensional visualization of orbital motion in a given dynamical model on the VR platform is quite powerful for human operators to leverage their natural intuitive abilities to design trajectories. Within this dissertation work, I present four new techniques on the VR platform to help the human operator identify potentially good initial guesses for the periodic orbit identification
problem in the circular restricted three-body model (CR3BP) of orbital motion for the trajectory design in the cislunar space. The four techniques are validated with periodic orbit identification of four types of Lagrange point periodic orbits, Lyapunov, halo, axial and vertical. All four orbit types are successfully identified with the initial guesses derived using the four techniques. I also implement an orbit-chain method application on the VR platform for low-thrust transfer trajectory identification in the cislunar circular restricted three-body model (CR3BP).
The low-thrust transfer trajectory identification problem is much more complex because of the size of the search space of the problem. This problem requires finding dynamically feasible paths that are also optimal given some criteria. The orbit-chain method allows for taking
advantage of naturally feasible paths of orbital motion in the initial guesses to help ease the computational cost of finding optimal transfer trajectories. This VR based orbit-chain method is successfully validated with two example transfer trajectory identification problems.
Together, these VR based initial-guess derivation methods prove to be a feasible option for preliminary trajectory design in the cislunar space.||en_US