Design of low-thrust gravity-assist trajectories for cycler missions to Mars and for non-Newtonian physics experiments
Abstract
We design and optimize low-thrust gravity-assist trajectories for several applications. For Earth-Mars cycling trajectories, we approach the problem by designing a low-thrust version of the well-known Aldrin cycler, and by “patching” together a series of ballistic semi-cyclers. These new cycler trajectories allow significant savings in terms of taxi rendezvous propellant expenditure over the original (nearly ballistic) Aldrin cycler and the semi-cyclers. Trade studies show that even though the propellant requirement on the low-thrust cycling vehicle is not insignificant, the overall architecture-level propellant cost savings is worthwhile. For the investigation of new physics, we design low-thrust trajectories that employ Jupiter gravity assists to reach a distance of 1000 AU from the Sun. Trajectories that allow close flybys of the Sun are also presented. In addition, we analyze the unexpected gamma-ray detection readings from the MESSENGER spacecraft, and the results may suggest interactions between spacecraft-Sun distance and radioactive decay rates. Besides these new trajectories for various applications, we also present a new design tool for low-thrust gravity-assist trajectories. This new design technique relies on an extension of the Tisser and graph, which tracks the changes in orbital shapes and energies from gravity assist maneuvers. Several simple low-thrust control schemes and their corresponding Tisserand graph curves are investigated and used to bridge gaps on the ballistic Tisserand graph (which could only be used for ballistic trajectories in its original formulation). We use the new low-thrust Tisserand graph to generate candidate (both direct and multi-flyby) rendezvous trajectories and show that the new tool can predict the mass-optimized ΔV cost to within 15% in most cases.
Degree
Ph.D.
Advisors
Longuski, Purdue University.
Subject Area
Aerospace engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.