<aside> <img src="/icons/promoted_yellow.svg" alt="/icons/promoted_yellow.svg" width="40px" /> Me, toward Research

I distinctly recall the moment our modest TurtleBot began to move for the first time. This moment was a culmination of collaboration among the five members of our club in a compact room, studying Linux and ROS, and assembling robots. In 2020, during the first summer break at university, I dedicated two whole months to this autonomous robot project, successfully implementing navigation for our robot to maneuver around the club room corridors. Starting from this, much like in a Markov Decision Process, my passion for robotics and AI evolved through such decisive moments over my four years at university.

</aside>

<aside> <img src="/icons/thought-dialogue_yellow.svg" alt="/icons/thought-dialogue_yellow.svg" width="40px" /> Research Interests

• Perception for Robotics
• Robot learning methods
• Perception and Decision-Making in Autonomous Vehicles
• Questions that I am digging into these days.
  1. How can an agent acquire skills with minimal (or without) human-engineered intervention?
  2. How can an agent acquire general-purpose skills to efficiently solve practical, long-horizon tasks?
  3. What are the key differences that distinguish AI agents from humans in the process of learning new skills?
  4. How can autonomous vehicles optimize motion planning to handle dynamic environments and unpredictable scenarios? </aside>

<aside> 💡 Projects

1. Reinforcement Learning for Manipulator - Target object approach algorithm (2024.01 ~ 2024.06)

• Designing a custom simulated environment that enables the robot to identify target objects
• Implementing RL algorithms to guide the manipulator towards the target efficiently
• Utilizing IsaacSIM, ORBIT RL framework, and Franka Emika - Panda
• Exploring various methodologies to process and convey 3D representation information to the robot for improved training

1. A Bedridden Patient Monitoring System Utilizing Computer Vision (2022.03 ~ 2022.12)

• Initiated this project after directly witnessing the lack of care workers in nursing homes and aiming to address these challenges with AI.
• Developed an algorithm to detect situations of high risk (falling, bedsores, delirium) for bedridden patients by analyzing postures.
• Implemented the OpenCV library and MediaPipe model with PyTorch
• The project resulted in publications.
  • Award

1. Chain of Thought Prompt Engineering for Pro vs. Con Discourse Newletters (2023.09 ~ 2023.12)

• Conducted prompt engineering experiments and LLM fine-tuning experiments.
• Analyzed sentiment on the comment data using BERT-based model
• Github Repository Link
• Presentation & Award </aside>