Master Thesis Physically Informed Machine Learning Based System Identification in MEMS Gyroscopes
Core
Develop physically informed machine learning algorithms for system identification and performance modeling of MEMS gyroscopes using real-world sensor data.
Role type
Master's thesis researcher (physically informed ML for sensor systems)
Builds
ML models for system identification and performance prediction of MEMS gyroscopes
Domain
MEMS sensors, physical modeling, machine learning
Deliverable
research
Required skills
data-driven parameter identification, Python, PyTorch, Pandas, probabilistic modeling, hands-on data handling, pipeline construction
Preferred skills
deep physical understanding of MEMS gyroscopes
Technologies
Python, PyTorch, Pandas
Responsibilities
Construct ML algorithms for system identification and performance prediction; examine MEMS gyroscope data through detailed analysis; assess physically informed ML against other architectures; work with real-world sensor data to validate findings
Seniority
Master's student researcher