SmallSat Steward: Autonomous Online Learning and Planning for Safe Inspection of Cislunar Stations

This project will develop a sophisticated system called SmallSat Steward designed to enhance capabilities for small spacecraft, particularly those with limited computational resources. Based on the established Dyna framework, it is a reactive and integrated architecture capable of real-time model updates and trajectory planning, where the spacecraft's path can be determined on-orbit. Steward’s focus is on exterior inspection in close proximity operations, with a design reference mission of inspecting a future cislunar space station such as Lunar Gateway. Autonomous small spacecraft offer a cost effective and unique solution to cislunar station inspection and maintenance. A known challenge is the significant number of changes to the spacecraft models are likely to experience (subsystem failure, payload manipulation, or other unknowns). A prototype demonstration in a relevant environment at JPL's Small Satellite Dynamics Testbed facility will ready Steward for infusion into future space missions. This project could potentially apply the Dyna reinforcement learning architecture in the space domain for the first time.

The proposed Machine Learning technology focuses on improving the turnaround time for trajectory planning to enable on-orbit inspection of high value assets (such as the Gateway station) using SmallSats. Steward will enable autonomous free-flying small spacecraft to perform the complex mission of exterior inspection of a cislunar station with limited communication to Earth and within the constraints of small spacecraft computational capabilities. This technology is crucial for long-term presence in space and is applicable in inspection and maintenance services but also other orbital services such as space debris management, anomaly detection, and collision avoidance. A potential near-term infusion into Astrobee aboard the International Space Situation (ISS) – prior to infusion into free-flyer CubeSats – also offers an attractive development path for the proposed technology. This work is based on the already-established Dyna framework and could apply the Dyna architecture in space for the first time.

This is early/mid-stage (TRL 3) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.