Satellite Augmented Lagrangian Trajectory Optimization (SALTRO)

SALTRO is a high-performance trajectory optimizer for satellite attitude control.

Using trajectory planning and advance knowledge of the magnetic field, it is for instance able to slew a MTQ-only satellite within one orbit, a task that usually takes multiple.

SALTRO’s predecessor flew on the MIT STAR Lab “BeaverCube 1” mission. SALTRO has a performance improvement of approximately 50x compared to the previous version and is intended for onboard computation.

Compute times for a 3 MTQ satellite over 5400 seconds, 10 second timestep

CPU	Algorithm	Compute Time
i9-13900K	BC2 SALTRO	0.782 s
i9-13900K	BC1 trajectory optimizer	50+ s

Furthermore, SALTRO supports hybrid magnetorquer + reaction wheel architectures, with interesting results like the below pointing maneuver that uses three magnetorquers and one reaction wheel:

Links

• The working principle behind SALTRO
• Installation
• OBC (FC) usage guide

Documentation

• limits
• constants
• math
• optimizer
• orbit_generation
• pybind
• validation