Ph.D. Defense
Theresa Bender
(Advisor: Prof. Dimitri Mavris)
Methodological Improvements for the Integration of Spacecraft Trajectory Optimization into Conceptual Space Mission Design
Thursday, April 11th
10:00 a.m.
Collaborative Visualization Environment (CoVE)
Weber Space Science and Technology Building (SST II)
and
Microsoft Teams
Abstract
As humans continue to send spacecraft further into space and explore uncharted territories, the implementation of space mission design becomes of paramount importance. Trajectory design and optimization is a key element of space mission design. It provides information on the specific route a vehicle will take, as well as numerical estimates pertaining to fuel consumption and transfer time. Since such estimates are generally required for the analysis of other subsystems and the overall mission, some level of trajectory design must be performed during the conceptual design phase. Due to the complexity, high computational costs, and long runtimes of high-fidelity trajectory analyses, less accurate methods are typically used. Low-fidelity estimates provide sufficient accuracy for initial analyses; however, they often lack valuable information about the trajectory that is important to consider during the conceptual design phase. As a result, potential trajectory options that could impact mission concept of operations or objectives may not be considered.
Spacecraft trajectory design is often a manual process requiring humans-in-the-loop and intermediate steps, which hinders its integration into conceptual mission design studies. As new information about the mission becomes available, subject matter experts are needed to rerun the analysis under the new conditions. Furthermore, many factors and considerations external to the trajectory design problem but that influence its design space are analyzed outside of the trajectory design problem. The overall objective of this research is to develop methodological improvements for spacecraft trajectory design and optimization that integrate these factors, provide increased flexibility, and better enable trajectory considerations to be incorporated into conceptual mission design studies.
This research proposes a design space exploration-based approach to the integration of trajectory design and optimization into conceptual space mission design. It aims to provide a strong characterization of the design space and understanding of the problem behavior, as well as be better suited for early phase design studies that possess unknown or evolving mission requirements. The first phase of this research introduces a design of experiments and sensitivity analysis into the traditional trajectory design process in order to identify the behaviors, sensitivities, and trends of trajectory optimization problems. A regression-based approach for the selection of initial guesses is proposed in order to perform more efficient design studies and gain additional insight about the relationships between variables. The second phase of this research investigates the integration of additional evaluation criteria, namely robustness and stability analyses, that are often performed independent of the trajectory design problem. A methodology is proposed for their quantification and integration into design space exploration studies so that they may be analyzed and visualized alongside performance-based metrics. The third phase of this research integrates mission design considerations into this parametric environment through the superimposition of constraints onto the design space. This results in a set of feasible trajectories that meets performance, robustness, stability, and mission design requirements and constraints.
With the proposed methodology established and validated, a demonstration on a cislunar mission design problem is performed in order to showcase the benefits and insights gained from leveraging these methods. A characterization of the design space and identification of its behaviors, sensitivities and trends are performed, which provides insight into the characteristics of viable transfer trajectories and final lunar orbits. Six solution families are revealed, each with unique behaviors and trends with respect to the design variables, optimization variables, and objective function. The robustness and stability of these solutions are then quantified and integrated into the performance-based design space exploration, which enables comparisons among the various solution families. Mission design considerations are then evaluated to enable the superimposition of all constraints in a dynamic and parametric evaluation environment, which facilitates identification of solution characteristics more prone to violation of the selected mission design constraints. Overall, the demonstration illustrates how the use of the developed framework and methodologies results in trade studies between trajectory design and other mission design considerations that are more comprehensive and flexible than the traditional design approach allows.
Committee
· Prof. Dimitri Mavris – School of Aerospace Engineering (Advisor)
· Prof. Glenn Lightsey – School of Aerospace Engineering
· Prof. John Christian – School of Aerospace Engineering
· Prof. Mariel Borowitz – School of International Affairs
· Dr. Michael Steffens – Draper