As space continues to become more congested, automated techniques for spacecraft maneuvering become increasingly attractive for tasks such as collision avoidance, rendezvous and proximity operations, and station keeping. This work uses hazard analysis to elicit requirements for an autonomous spacecraft controller. Spacecraft maneuvers today are planned by human operators and conducted days to hours in advance. This represents a risk averse climate that is hesitant to rely on automation. In the absence of regulations governing automated maneuvering, a risk-based approach is a promising technique. First, top-down accidents, hazards, and safety constraints are identified. Second, a functional control model for an automatic collision avoidance system on a spacecraft in the context of a theoretical Space Traffic Management system is constructed using System Theoretic Accident Models and Processes (STAMP). Third, unsafe control actions, scenarios, and mitigating requirements are identified using Systems Theoretic Process Analysis (STPA). These requirements form the foundation for the development of automatic control designs for spacecraft. Finally, the safety constraints are formally specified as high level requirements as a path towards formal analysis of the system.
|Original language||English (US)|
|Title of host publication||AIAA Scitech 2021 Forum|
|Publisher||American Institute of Aeronautics and Astronautics|
|Number of pages||25|
|State||Published - Jan 4 2021|
Bibliographical noteKAUST Repository Item: Exported on 2021-02-24
Acknowledgements: The authors would like to thank Dr. Chris "Chrispy" Petersen, Dr. Sean Phillips, Dr. R. Scott Erwin, Dr. Kendra Lang, Ms. Michelle Simon, and Dr. Daren McKnight for feedback and opportunities to present this information to others in the spacecraft community for feedback.