Automated Maneuver Design and Checkout for the Lunar Reconnaissance Orbiter.

摘要

The Lunar Reconnaissance Orbiter (LRO) is a lunar surface mapping and data collection mission launched by NASA in 2009. As a mapping and imaging mission, frequent attitude maneuvering is required. The LRO currently follows a trial-and-error method to design maneuvers to prevent sensitive instruments from pointing at bright objects that may damage the equipment. Additionally, eigenaxis maneuvers are the primary method by which the attitude is controlled. In this thesis, optimal control theory is applied to provide automated maneuver design capabilities to support the LRO mission. The approach allows dynamic constraints, as well as other constraints such as occultation avoidance, to be easily incorporated into the maneuver design process. This aspect also simplifies maneuver checkout activities. The results of this thesis show that maneuvers can be designed to reorient the LRO in the presence of multiple occultation constraints. Moreover, maneuver times can be reduced up to 90 percent compared to the conventional approach. This increases the potential for efficient science data collection.