2020 Problem Statements: Trusted Autonomy


Space enterprise lags behind terrestrial and aerial robot development, in both hardware and software. Space qualified hardware is expensive and explains the lag there but software development is constrained by the lack of high speed space processors and general community trust of autonomous operations.

The national space community recognizes the need for highly autonomous, on-orbit operational spacecraft capable of an increasing number of complex tasks, but at the same time asks, how can we be certain that the platforms & algorithms being utilized for space operations can be relied upon? How can we trust autonomy?



Next-Level Autonomy in Remote Environments 

How might we achieve next-level autonomy in remote environments for servicing of modules or vehicles?

Improving Automatic Hazard Detection and Avoidance

How might we improve automatic hazard detection and avoidance to enable safer operations?


Adopting State-of-the-Art Machine Learning for Space Vehicles 

How might we leverage machine learning for autonomous vehicle control?
Untitled design (10)


Preventing and adapting to autonomy failures

How might autonomous systems detect and react to system/sensor failures?
Untitled design (11)


Enhancing Spacecraft Autonomy with Resilient Computing Solutions 

How might we develop the future of hardware solutions for autonomous software in harsh environments?


Subproblem: Automatic Hazard Detection for Space Weather 

How might we improve automatic hazard detection related specifically to space weather?