Hyunseo Jang

M.S. Student, Department of Mobility Science and Engineering
Field Robot Lab, Korea University
Advisor: Prof. Youngeun Song

hyunseo_jang.jpg
Field Robot Lab Korea University kelvin926@korea.ac.kr

I am an M.S. student in the Field Robot Lab at Korea University. My research focuses on simulation-based robot learning and control for mobile robotic systems, with an emphasis on mobile manipulation, wheeled-legged loco-manipulation, and stability-aware evaluation. Recent projects include Isaac Lab/RSL-RL policy training for KALO and shelf-transfer mobile manipulation, ROS/Autoware-based AMR integration, and CEBO-based simulation evaluation-loop automation.

Mobile manipulation and dynamic carrying control Object-conditioned carrying, shelf-transfer tasks, and bimanual/mobile manipulation.
Robot learning and simulation Isaac Lab, RSL-RL, PPO, MDP/reward design, curriculum design, and policy evaluation.
Stability-aware evaluation Object tilt, grasp stability, base slip, chassis leveling, and safety metrics.
AMR perception and navigation ROS/Autoware, LiDAR mapping, detection/tracking, and campus-scale validation.

current research direction

I am interested in object-conditioned robot learning and control for mobile manipulation. In particular, I aim to study how mobile manipulators can adjust arm motion, base acceleration, and safety margins depending on object properties such as mass, shape, grasp stability, and fragility.

This direction grew from my KALO, shelf-transfer mobile-manipulation, and AMR integration projects, where I worked on policy training, task formulation, and stability-oriented evaluation.

news

2026
KALO was accepted to IEEE/ASME AIM 2026.
2026
The lift-equipped mobile manipulator paper was accepted to IEEE/ASME AIM 2026.
2026
The local delivery robot system paper was accepted to IEEE CASE 2026.
2026
The CEBO manuscript was revised and resubmitted to Advanced Engineering Informatics after minor revision.
2025
Selected for the NRF Master's Student Research Grant (한국연구재단 이공분야 학술연구지원사업 석사과정생 연구장려금) as an individual research project. 🎉
CEBO-Based Stair-Climbing Wheel Design

selected publications

KALO: Kangaroo-LocoManipulation for Early Launch Acceleration in a Wheel-Legged Robot
IEEE/ASME AIM 2026, accepted.
A Lift-Equipped Mobile Manipulator Platform for Shelf Transfer in Multi-Level Rack Environments
IEEE/ASME AIM 2026, accepted.
Design and Implementation of a Local Delivery Robot System: A Reinforcement Learning Approach
IEEE CASE 2026, accepted.
Clustering-enhanced Bayesian optimization (CEBO): A case study on T-pedal wheel design of a stair-climbing robot
Advanced Engineering Informatics, revised manuscript resubmitted after minor revision.