I am a Ph.D. candidate in Mechanical Engineering at MIT, working in the Biomimetic Robotics Lab. My research focuses on robot learning and reinforcement learning for humanoid robots, with an emphasis on dexterous manipulation and whole-body control, and on combining model-based optimization with learning-based control. I develop MPC- and RL-based controllers for the MIT Humanoid.

In summer 2026, I was a Reinforcement Learning / Robotics Controls Intern at Tesla, where I developed and evaluated reinforcement learning and control algorithms for the Optimus humanoid robot.

Before MIT, I received my M.Sc. in Robotics, Systems and Control from ETH Zürich, where I worked at the Robotic Systems Lab on learning human-robot handover skills from human demonstrations with deep reinforcement learning. I received my B.S. in Mechanical and Aerospace Engineering (minor in Mathematics) from Seoul National University, where I worked on autonomous drone control and visual odometry for unmanned delivery.

News

• [05/2026] Started a summer internship at Tesla as a Reinforcement Learning / Robotics Controls Intern, working on the Optimus humanoid.
• [05/2026] Gave an invited talk at Holiday Robotics, South Korea.
• [05/2026] Our paper Learning Reactive Dexterous Grasping via Hierarchical Task-Space RL Planning and Joint-Space QP Control is on arXiv (under review at IEEE T-RO).
• [03/2026] Gave an invited talk at the LIDAR Lab, Georgia Institute of Technology.
• [2026] Point2Pose: Occlusion-Recovering 6D Pose Tracking and 3D Reconstruction for Multiple Unknown Objects via 2D Point Trackers was accepted to ECCV 2026.
• [2026] Residual MPC: Blending Reinforcement Learning with GPU-Parallelized Model Predictive Control was accepted to IEEE Transactions on Robotics.
• [2025] Learning Humanoid Arm Motion via Centroidal Momentum Regularized Multi-Agent Reinforcement Learning was accepted to IEEE RA-L.
• [2024] Integrating Model-Based Footstep Planning with Model-Free Reinforcement Learning for Dynamic Legged Locomotion was accepted to IROS 2024, and CusADi: A GPU Parallelization Framework for Symbolic Expressions and Optimal Control was accepted to IEEE RA-L.
• [09/2023] Started my Ph.D. at the Biomimetic Robotics Lab, MIT.
• [10/2022] Started as a visiting student researcher at the Biomimetic Robotics Lab, MIT.
• [02/2022] Started as a student researcher at the Robotic Systems Lab, ETH Zürich.
• [09/2021] Started my Master’s program in Robotics, Systems and Control at ETH Zürich.

Publications

For a complete and current record, please visit my Google Scholar profile.

Learning Reactive Dexterous Grasping via Hierarchical Task-Space RL Planning and Joint-Space QP Control Ho Jae Lee, Yonghyeon Lee, Alexander Alexiev, Tzu-Yuan Lin, Se Hwan Jeon, Sangbae Kim IEEE Transactions on Robotics (under review), 2026 Paper Video

Point2Pose: Occlusion-Recovering 6D Pose Tracking and 3D Reconstruction for Multiple Unknown Objects via 2D Point Trackers Tzu-Yuan Lin, Ho Jae Lee, Kevin Doherty, Yonghyeon Lee, Sangbae Kim European Conference on Computer Vision (ECCV), 2026 Website Paper Code Video

Residual MPC: Blending Reinforcement Learning with GPU-Parallelized Model Predictive Control Se Hwan Jeon, Ho Jae Lee, Seungwoo Hong, Sangbae Kim IEEE Transactions on Robotics (T-RO), 2026 Paper Video

Learning Humanoid Arm Motion via Centroidal Momentum Regularized Multi-Agent Reinforcement Learning Ho Jae Lee, Se Hwan Jeon, Sangbae Kim IEEE Robotics and Automation Letters (RA-L), 2025 Paper Video

CusADi: A GPU Parallelization Framework for Symbolic Expressions and Optimal Control Se Hwan Jeon, Seungwoo Hong, Ho Jae Lee, Charles Khazoom, Sangbae Kim IEEE Robotics and Automation Letters (RA-L), 2024 Paper Code Video

Integrating Model-Based Footstep Planning with Model-Free Reinforcement Learning for Dynamic Legged Locomotion Ho Jae Lee, Seungwoo Hong, Sangbae Kim IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024 Paper Code Video