UAV Researcher / Embedded Systems Engineer
Student researcher at Chungbuk Science High School, South Korea.
I'm Muwon Lee, an independent student researcher and developer focused on autonomous aerial systems, embedded engineering, and robotics.
My work bridges physical simulation, control theory, hardware design, and software engineering. I enjoy building complete systems β from circuit boards and firmware to simulation environments, research pipelines, and user interfaces.
I'm particularly interested in making autonomous systems more reliable under real-world constraints, including uncertain environments, sensor limitations, and computational restrictions.
I believe that research is not only about achieving better numerical results, but also understanding when and why an approach fails.
Better predictions do not automatically make better controllers.
| Research | Description |
|---|---|
| Ground Effect Stabilizer (GES) | Investigating machine-learning-based ground-effect compensation in multirotor UAVs. Built a custom C physics simulator and evaluated prediction accuracy against closed-loop flight-control performance. |
| Drone Drop Finder (DDF) | Physics-based probabilistic prediction of drone crash impact locations using Monte Carlo simulations and an embedded C lookup-table library. |
| FPV LiPo Thermal Runaway Guard | Researching battery health estimation and thermal-risk prediction using equivalent circuit models, Extended Kalman Filters, and thermal reaction models. |
Current research interests
- Fault-tolerant flight control and UAV safety
- Autonomous navigation and sensor fusion
- Embedded machine learning and real-time control
- Physics-based simulation and reproducible experimentation
- RF-based environmental sensing and reconstruction
| Project | Description | Technologies |
|---|---|---|
| CanSat-childish | Two-stage CanSat system featuring reaction-wheel stabilization, deployable components, flight telemetry, and a custom ground station. | STM32, Arduino, C/C++, Python, KiCad |
| rpycrsf | Python library for controlling CRSF/ELRS-based drone systems through Raspberry Pi UART interfaces. | Python, Raspberry Pi, CRSF |
| Cutist | Adaptive Pomodoro companion robot with interactive hardware and a mobile application, developed for the IEEE SSCS Arduino Contest. | Arduino, C++, Flutter |
| CBFESTA | School festival management platform with QR check-in, live queues, team rankings, and administrator dashboards. | React, TypeScript, Supabase |
| GetYourEarth | An open-source software project from my development portfolio. | Software Engineering |
| IESBJ | Student-oriented research publishing initiative supporting technical writing and academic communication. | Web Development, Research |
|
Embedded Systems |
Programming |
CAD / EDA |
Development |
Additional tools and frameworks
My research workflow follows a simple principle:
Model β Simulate β Implement β Evaluate β Understand
I am interested in the gap between theoretical performance and real-world behavior.
A model with impressive offline accuracy is not necessarily a better controller. A sophisticated algorithm is not always better than a simple analytical solution. And negative results can be valuable when they reveal why an approach does not work.
I aim to develop engineering solutions that are physically grounded, computationally practical, and reproducible.
"Code is a work of art, and above all, it should be readable."
β Bjarne Stroustrup
Built with curiosity, tested against physics.


