Gyeonghun Kim

I am an undergraduate student at Seoul National University majoring in Electrical and Computer Engineering (ECE). My research interests lie at the intersection of Robot Learning and Physical AI, specifically focusing on how agents can learn complex manipulation tasks in unstructured environments.

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn

profile photo

Research Focus

My goal is to build robots that can perceive and interact with the physical world as robustly as humans do. I am currently exploring Vision-Language Models (VLMs) for robotic control and Reinforcement Learning under physical constraints.

Publications

Learning Physical Interaction via Vision-Language Models
Gyeonghun Kim, Co-author 1, Co-author 2
Under Review / ICRA 2026
project page / arXiv / video

Integrating high-level semantic reasoning with low-level motor control for agile robotics.

Projects

Open-Source Robot Arm Controller
Personal Project, 2025
Github

Developed a custom PID-based controller for a 6-DOF robotic arm using Python and Arduino.