Selected Projects & Papers
A blog-style exploration of PPO for robust Earth→Mars low-thrust guidance, with interactive visuals, equations, and code. Read the post
Actor–critic scheduling for autonomous RSO imaging with Basilisk and BSK-RL. See page
RL-Driven Scheduling for Autonomous Space-to-Space Surveillance
This project introduces a reinforcement learning (RL)-based approach for autonomous inspection and imaging of Resident Space Objects (RSOs) from a spacecraft in orbit. The agent is trained in a high-fidelity Basilisk simulation and dynamically selects imaging targets while managing LOS, battery/wheel states, and data limits.
🎥 Project Video
📄 Abstract: Space-to-Space Surveillance
Reinforcement learning scheduler for autonomous RSO inspection with safety and lighting constraints in a high-fidelity Basilisk simulation. Integrates battery, data storage, wheel desaturation, and eclipse/LOS observability.
📄 Abstract: Autonomous Satellite Inspection
Genetic algorithm for optimizing complex interplanetary missions using gravity assists and impulsive maneuvers. Balances low ∆v and fast arrival time without pruning non-intuitive solutions.