|Ph.D Student||Pisarevsky Dmitry|
|Subject||Design of Low-Thurst Gravity-Assist Inter-Planetary|
|Department||Department of Aerospace Engineering||Supervisors||PROFESSOR EMERITUS Mauricio Guelman|
|DR. Alexander Kogan|
|Full Thesis text|
Gravity-Assisted Maneuvers constitute a technique of changing the velocity of a spacecraft due to the gravity of a passed-by massive celestial body. A wise uses of such maneuvers helps to considerably reduce fuel consumption and/or flight time. Generally, design of interplanetary trajectories that use multiple gravity-assisted maneuvers is extremely difficult. There is, however, a well-known type of trajectories, for which the design process is straightforward: the case of periodic trajectories (cyclers) in the four-body problem. A new technique is proposed to identify all candidates for cyclers in any restricted, coplanar, circular, two-planet system. In the general case where both planets enable gravity-assisted maneuvers, Hénon's diagram is extended to present all possible trajectories, which start and end near the same planet with the same relative velocity, and therefore can be used to predict all the cycler trajectories. Using cycler trajectory is restricted to those cases, where relative velocity is preserved. However, there are many space missions where changing the relative velocity is the main goal of the gravity-assisted maneuvers. A new method for building inter-planetary trajectories with relative velocity variation is proposed. It is based on the concatenation of repeated short sequences of Keplerian arcs, referred as blocks. These blocks differ from each other only in inclination. The method is applicable to any resonant planetary system counting an arbitrary number of planets in circular or elliptical orbits and allows the assembling of long trajectories otherwise not amenable to computation. In addition, a new algorithm is proposed for low-thrust orbit control. It enables to reduce the flight time and allows using the strategy in near-resonant systems as well. The proposed methods will be presented along with a number of representative examples.