Projects.

• Role: Undergraduate Research Assistant
• Summary: Developed a crowd density analysis model and built an MLflow-based pipeline for continuous on-device deployment and execution on Jetson Orin.

Developed an integrated system utilizing an ArUco marker on a ground robotic arm to guide drone flight and landing, controlled via a LangChain-based agent.

• Key Implementation: PPO-based drone tracking, ACT-based robotic arm guidance, and LangChain agent integration.
• Achievement: Korean Air Capstone Design Award

Developed a natural language-controlled item classification system by integrating an LLM agent with a custom-built Arduino-based wheel sorter.

• Developed a Raspberry Pi 4-based autonomous driving system, reducing track completion time by 58% through optimized image processing and YOLO-based perception.
• Achievement: Excellence Award at the Gachon University Autonomous Driving Competition

Developed an anomaly detection model for bearing vibration data, improving accuracy by 29.67%p through signal analysis and autoencoder architecture modifications.