Learning Agent for Autonomous Spacecraft Activity Planning and Orbit Selection

This proposal aims to study how artificial intelligence can be used to allow small body observing spacecraft to assume control over the mission, science, and activity planning processes that typically occur on the ground. This research will investigate intelligent learning agents that will use both machine learning techniques to predict resource usage onboard spacecraft and co-optimization strategies to generate spacecraft activity plans and select orbital geometries about small bodies like asteroids, comets, and Kuiper belt objects. Specific challenges that this research will address are operational burden and cost, responsiveness to anomalies and science opportunities, and uncertainty in spacecraft activity resource usage and execution time. To achieve these objectives, the research will be broken down into four distinct phases. In phase one of the proposed research, new tools will be developed to simulate the power and data generated and consumed onboard spacecraft during operations. In phase two, different optimization strategies will be explored to solve a traveling purchaser problem (TPP) formulation of the small body observing spacecraft activity planning problem. In phase three, the activity planning optimization strategies will be combined with machine learning techniques to predict activity resource usage and execution time onboard spacecraft, improving the accuracy of the plans developed in phase two. Finally, in phase four, the optimization strategies and machine learning techniques developed in phases two and three will be simulated in realistic scenarios with hundreds of activities and orbits to choose from. The techniques will be evaluated on computational efficiency, quality of their solutions, and the likelihood of violating resource constraints.

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.