Vision–force learning for dexterous bimanual control.

I am a Mechanical Engineering M.S. student at Sungkyunkwan University specializing in robotics. My work focuses on learning and control systems for manipulators that can acquire precise, transferable skills from human demonstrations.

I have built a VR-tracker teleoperation system and, separately, a force-feedback leader-arm teleoperation system designed specifically to collect high-quality demonstration data. I also developed a ROS 2 visual-servoing package during an internship with the Robot Control Team at Doosan Robotics.

For imitation learning, I have evaluated multiple policies and modified their internal architectures to improve learning performance. My current research extends vision–proprioception policy architectures to learn from force as an additional modality for dexterous, learning-based control of dual-arm robots with hands.