Reconfigurable Spacecraft Controller Design via Location-Scheduled Control

Abstract

T HIS Note proposes a new strategy for the design of the attitude determination and control system (ADCS) of reconfigurable spacecraft, with a specific application example to so-called nanosatellites. These spacecraft adhere to the CubeSat standard [1], and the natural discretization of the overall volume into cube units makes them a popular platform for research into spacecraft reconfiguration [2]. The proposed design strategy, called location-scheduled control (LSC), seeks to reduce the development time required for the ADCS of a reconfigurable spacecraft conducting scientific research. In traditional ADCS design, mission science requirements levy pointing requirements onto the ADCS, which drive sensor and actuator selection. Once the ADCS hardware has been chosen, an appropriate control algorithm is selected and tested. In contrast, the location-scheduled approach to ADCS design is a two-phase process. In the first phase, a general purpose self-contained ADCS hardware unit is designed by selecting a suite of sensors and actuators that are appropriate for a class of science missions. Controller gains and ADCS location are then optimized in simulation for a number of generic science payloads, such as those depicted in Fig. 1, in which payload cube (P) layouts dictate possible locations for the ADCS (cubes A or B). In the second phase of the process, one of the previously tested configurations is selected based on the needs of the current science mission. Some amount offine gain tuningwill inevitably be required, but if a member of the catalog of spacecraft configurations satisfies the needs of the science payload, the developer need not return to the first phase of the design process. Although the LSC approach is applicable for many reconfigurable spacecraft, thiswork considers a nanosatellite ADCSdesign problem to demonstrate this methodology. The objective of the design problem is to find the optimal combination of ADCS unit location and control law gainvalues for a nanosatellite performing a detumble maneuver in which the spacecraft has high initial angular velocity and arbitrary initial orientation. Using the CubeSat standard, a 10 10 10 cm cube is designated as one unit of volume (abbreviated 1U), and complete spacecraft are created from these cubes. Traditional CubeSats are typically no larger than 3U in volume, and they have largely been passively controlled via permanent magnets [3,4], gravity gradient booms, or aerodynamic stabilization. This Note focuses on spacecraft that are 6U in volume (as in Fig. 1), with overall dimensions of 10 20 30 cm. This form factor creates greater volume for both complex science payloads and attitude control hardware and is currently under development at NASAAmes Research Center. A nonlinear control law introduced in [5] is used to control the nanosatellite using actuators for which the control authority is restricted by a known upper bound.