Adversarial Machine Learning Threats to Spacecraft
This addresses security risks for spacecraft systems that rely on autonomy, but it is incremental as it applies known AML concepts to a new domain.
The paper tackles the vulnerability of spacecraft to adversarial machine learning (AML) attacks by introducing a threat taxonomy and demonstrating these attacks through experimental simulations using NASA's Core Flight System and OnAIR Platform, showing that such attacks can disrupt autonomous processes.
Spacecraft are among the earliest autonomous systems. Their ability to function without a human in the loop have afforded some of humanity's grandest achievements. As reliance on autonomy grows, space vehicles will become increasingly vulnerable to attacks designed to disrupt autonomous processes-especially probabilistic ones based on machine learning. This paper aims to elucidate and demonstrate the threats that adversarial machine learning (AML) capabilities pose to spacecraft. First, an AML threat taxonomy for spacecraft is introduced. Next, we demonstrate the execution of AML attacks against spacecraft through experimental simulations using NASA's Core Flight System (cFS) and NASA's On-board Artificial Intelligence Research (OnAIR) Platform. Our findings highlight the imperative for incorporating AML-focused security measures in spacecraft that engage autonomy.