Projects

LOST: Open-source Star Tracker is our star tracker software, extensively optimized to run on low-powered flight hardware. Using camera-based attitude determination, LOST identifies on-camera constellations to precisely determine a spacecraft's orientation in orbit.

LOST provides a free and publicly available alternative to expensive proprietary star trackers, making precision attitude determination accessible to university teams and small satellite operators worldwide. The software is designed to be hardware-agnostic, supporting multiple camera configurations and radiation environments.

On HuskySat-2, LOST will demonstrate GPS-independent optical navigation as part of our broader navigation system. For deeper space missions like HuskySat-3, LOST enables autonomous attitude determination in the cislunar environment where GPS is unavailable.

Learn more on the SmallSat paper and the LOST website.

Current Leads: Alice Fu, Wanhao Zhang

FOUND: Open-Source Universal Navigation Determiner calculates orbital trajectory using Earth limb detection from camera imagery. By analyzing pictures of the Earth against the void of space, FOUND precisely determines the spacecraft's position without relying on GPS signals.

This GPS-free navigation approach opens new possibilities for spacecraft operating in environments where GPS is weak, delayed, or entirely unavailable-- including cislunar space, lunar orbit, and beyond. The codebase follows the Christian Robinson algorithm to reconstruct orbital parameters from visual observations alone.

Combined with LOST, FOUND forms the core of our autonomous navigation stack, enabling deep space missions to determine both their attitude (which way they're pointing) and their ephemeris (where they are in orbit) using only optical sensors.

Current Leads: Senuka Liyanage, Josh Lando

Our orbital simulation project uses Basilisk, a modular spacecraft simulation framework, to model and validate our navigation algorithms, attitude control systems, and orbital mechanics before flight hardware is integrated.

Basilisk enables high-fidelity simulation of spacecraft dynamics, including orbital perturbations, attitude dynamics, sensor models, actuator performance, and environmental effects. This allows us to test our guidance, navigation, and control (GNC) algorithms in a realistic virtual environment that closely matches conditions in orbit.

Key simulation activities include validating LOST and FOUND algorithms under various lighting conditions and orbital regimes, testing attitude control performance for HS-2 and HS-3, and performing Monte Carlo analysis to characterize navigation accuracy and robustness.

We also use Basilisk for hardware-in-the-loop (HIL) testing to validate software against simulated avionics interfaces and timing.

Current People: Mahir Emran, the HS-3 Orbital team