This paper describes the development and validation process of a highly automated Guidance, Navigation, & Control (GN&C) subsystem for a small satellite on-orbit inspection application. The resulting GN&C subsystem performs proximity operations (ProxOps) without human-in-the-loop interaction. The paper focuses on the integration and testing of GN&C software and the development of decision logic to address the question of how such a system can be effectively implemented for full automation. This process is unique because a multitude of operational scenarios must be considered and a set of complex interactions between various GN&C components must be defined to achieve the automation goal. The GN&C subsystem for the Prox-1 satellite is currently under development within the Space Systems Design Laboratory at the Georgia Institute of Technology. The Prox-1 mission involves deploying the LightSail 3U CubeSat, entering into a leading or trailing orbit of LightSail using ground-in-the-loop commands, and then performing automated ProxOps through formation flight and natural motion circumnavigation maneuvers. Operations such as these may be utilized for many scenarios including on-orbit inspection, refueling, repair, construction, reconnaissance, docking, and debris mitigation activities. Prox-1 uses onboard sensors and imaging instruments to perform its GN&C operations during on-orbit inspection of LightSail. Navigation filters perform relative orbit determination based on images of the target spacecraft, and guidance algorithms conduct automated maneuver planning. A slew and tracking controller sends attitude actuation commands to a set of control moment gyroscopes, and other controllers manage desaturation, detumble, thruster firing, and target acquisition/recovery. All Prox-1 GN&C components are developed in a MATLAB/Simulink six degree-of-freedom simulation environment and are integrated using decision logic to autonomously determine when certain actions should be performed. The complexity of this decision logic is the main challenge of this process, and the Stateflow tool in Simulink is used to establish logical relationships and manage data flow between each of the individual GN&C hardware and software components. Once the integrated GN&C simulation is fully developed in MATLAB/Simulink, the algorithms are autocoded to C/C++ and integrated into flight software. The subsystem is tested using hardware-in-the-loop on the flight computers and other hardware.
展开▼
