Spacecraft Dynamics (TA)
Led MATLAB recitations: rigid body dynamics, attitude kinematics, and control.
Attitude Determination & Control
ADCS
Where Physics Meets Imagination
ApexOrbital fuses programming, mathematics, and animation to bring the laws of physics to life, turning abstract principles into intuitive visual experiences.
About
I'm Mehdi Muzaffari, I investigate control and estimation systems for satellites and deep-space probes, fusing dynamics with real-time software to enable autonomous spaceflight.
Research Areas
Orbit Estimation & Navigation
ODHigh-fidelity propagation and measurement modeling for LEO-GEO; filters fusing GNSS, star-tracker and ground-based angles for robust navigation.
- Batch least squares, EKF
- SRP & J2 perturbations
- Angles-only navigation
Learning for Aero & Space
MLPhysics-informed networks and residual learning for aerodynamic modeling, fault detection, and on-board decision-making.
- PINNs & hybrid models
- Anomaly detection
- On-board inference
Projects & Demos
Open-Source CubeSat ADCS
cFS-integrated ADCS stack with EKF attitude estimation and quaternion control.
cFSC/C++Python
Explore →
Angles-Only Navigation
LEO navigation pipeline fusing ground angles and star-tracker centroids.
EKFAstrodynamicsNumPy
Explore →
PINNs for Aero Coefficients
Hybrid models blending CFD priors with residual learning for lift/drag estimates.
PyTorchCFDML
Explore →
Forthcoming Publications
- Muzaffari, M., et al. “Machine Learning for On-Orbit Anomaly Detection.” In progress.
- Muzaffari, M., et al. “Hybrid Navigation Filters for Deep-Space SmallSats.” In progress.
Teaching & Mentorship
-
2025
-
2024
Dynamic Systems (TA)
Led MATLAB/Simulink recitations: system modeling, transfer functions, and frequency response.