Distributed Health Management for Resilient Multi-agent Collaborative Spacecraft Inspection

As the development of challenging missions like on-orbit construction and collaborative inspection that involve multi-agent systems increases, the requirements needed to improve post-failure safety to maintain the mission performance also increases, especially when operating under uncertain conditions. In particular, space missions that involve distributed spacecraft systems (DSS) in low Earth orbit for Earth observation and on-orbit servicing purposes are susceptible to threats, making DSS prone to failures that are detrimental to its overall system performance. This paper describes recent progress towards the development of a distributed health monitoring system that uses artificial intelligence mechanisms to detect for potential threats to a DSS. The approach is applied to the problem of providing resiliency to potential failures for a fleet of spacecraft that are performing an on-orbit inspection. Numerical simulations are used to demonstrate the performance and capabilities of this architecture when the spacecraft network is subjected to different threats such as bounded disturbances and actuator failures.