Cho Hyeon Joon

Cho Hyeon Joon

Undergraduate researcher · Mechanical Engineering (Feb. 2027); Double Major in Physics (Aug. 2027) · Sungkyunkwan University

I work on robot learning for physical manipulation under distribution shift. My projects cover imitation learning, vision-language manipulation, and recovery after safety interventions, with most evaluation carried out on a real RB10 manipulator.

Most recently I designed a recovery-aware policy bridge for states created by reactive safety interventions. The proposed method completed 146 of 160 real-robot trials (91.3%), compared with 64.4% reported for the original LPB under the same intervention setting. A manuscript is in preparation.

Recovery-aware LPB is evaluated across repeated RB10 inference rollouts; Cho Hyeon Joon monitors and resets the task beside the robot. Playback at 2× speed.

Research

  1. Recovery-Aware LPB for RMP-Induced OOD States2026 · Manuscript in preparation91.3% Recovery-aware LPB
  2. Force-Feedback Teleoperation for Real-Robot Imitation Learning2025 · Completed9/10 Integrated final trials
  3. Grounded Language-to-Motion Planning with Memory and Gripper Constraints2025 · Evaluated on 15 scenarios12/15 Improved system
  4. Zero-Shot Language-to-Motion Planning with Geometric Validation2024–2025 · Prototype · early robot trialsZero-shot Planning setting

All research, with figures and evaluation conditions →