Watch experiment ↗01Locomotion and loco-manipulation
Humanoid Control
- Locomotion
- Whole-body motion
- Physical AI
OPEN POSITIONSRCI Lab is recruiting motivated undergraduate researchers and graduate students.
Recruitment details →ROBOT CONTROL · PHYSICAL AI · INTELLIGENCE
Robots that move withpurpose.
We build control and intelligence that help robots move, collaborate, and make decisions in complex, human-centered environments.
- 2023
- Established
- 17
- Graduate researchers
- 5
- Research interns
WHAT WE DO
Control Everything.
In the Real World.
From optimal control to Physical AI, our work connects rigorous methods with experiments on real robotic systems.
Watch experiment ↗01
Watch experiment ↗02Dynamic coordination for complex robots
Whole-body Control
- Whole-body control
- Hierarchical QP
- Mobile manipulation
Watch experiment ↗03Real-time decisions under constraints
Optimal Control
- MPPI
- DDP
- Trajectory optimization
Watch experiment ↗04Planning on complex constraint manifolds
Constrained Motion Planning
- Motion planning
- Closed-chain
- RRT
Watch experiment ↗05Language-guided robot behavior
Visual Language Action
- VLA
- Robot learning
- Manipulation
06Semantic maps for embodied navigation
Visual Language Navigation
- VLN
- Semantic mapping
- Embodied AI
SELECTED WORK
Recent publications
Research across robot control, motion planning, navigation, and learning-enabled robotic systems.
Browse publications →GRACE: Gradient-Free Robot Action Generation via Combined Diffusion-MPPI Posterior Mean Estimation
RMRRT: Riemannian Barrier Metric RRT for Inequality-Aware Steering on Equality Manifolds
Task-Specific Acceleration-Domain Trajectory Optimization for Smooth and Energy-Efficient Industrial Manipulator Motion
LAB UPDATES
Latest news
Our GRACE paper by Leesai Park and Jiho Hong has been accepted to IEEE Robotics and Automation Letters.
Leesai Park*, Jiho Hong*, Sanghyun Kim†, “GRACE: Gradient-Free Robot Action Generation via Combined Diffusion-MPPI Posterior Mean Estimation,” IEEE Robotics and Automation Letters, 2026.
RCI Lab won a grant from Kyung Hee University's BRIDGE program.
과제명: 접촉 작업 자율화를 위한 힘 센싱 휴머노이드 상체 및 접촉 인지 모방학습 기술 · 과제형태: G-대학 창의적 자산 실용화 지원(BRIDGE) 사업 – Advanced Grant(기술이전형) · 지원기관: 경희대학교 · 수행기간: 2026–2027
Our RMRRT paper by Minhyeong Kang has been accepted to IEEE Transactions on Robotics.
Minhyeong Kang, Sanghyun Kim†, “RMRRT: Riemannian Barrier Metric RRT for Inequality-Aware Steering on Equality Manifolds,” IEEE Transactions on Robotics, 2026.
RCI Lab won a grant from Wonik Robotics.
과제명: 시뮬레이션 상 AGV(QD·SD 타입) 자율주행 환경 구축 및 SLAM·주행계획 개발 · 과제형태: 산학협력 · 지원기관: 원익로보틱스 · 수행기간: 2026
A paper with Prof. Sanghyun Kim as a co-corresponding author has been accepted to Engineering Science and Technology, an International Journal.
Dongjo Kim, Minhyeong Kang, Hojin Lee, Nakyoung Lee, Stephen Baek, Sanghyun Kim†, Sanga Lee†, Jeongsu Lee†, “Task-Specific Acceleration-Domain Trajectory Optimization for Smooth and Energy-Efficient Industrial Manipulator Motion,” Engineering Science and Technology, an International Journal, 2026.
RCI Lab won a grant from Konan Technology.
과제명: URDF 물리 파라미터 추정·보정 알고리즘 분석 및 프로토타입 적용·검증 · 과제형태: 산학협력 · 지원기관: 코난테크놀로지 · 수행기간: 2026
RCI Lab provided technical consultation for the drama “연애박사,” featuring robots from our lab.
RCI Lab이 드라마 「연애박사」의 자문에 참여했으며, 연구실의 여러 로봇이 작품에 출연합니다. 자문 참여: 김상현 교수, 강민형·박수환·박다움·조현호. 엔딩 크레딧은 Pictures에서 확인할 수 있습니다.
RCI Lab registered a domestic patent on autonomous chemical experiments.
김상현, 박보형, 조현호, “화학 실험 자율 수행 장치 및 방법,” 대한민국 특허 등록 · 등록번호: 10-3033149 · 출원일: 2026.02.09 · 출원인: 경희대학교 산학협력단.
Our MC-MPPI paper by Seulchan Lee has been accepted to International Journal of Control, Automation, and Systems.
Seulchan Lee, Sanghyun Kim†, “Manifold-Constrained MPPI: Real-Time Sampling-Based Control for Nonlinear Equality-Constrained Robotic Systems,” International Journal of Control, Automation, and Systems, 2026.
Our paper by Jaerak Son has been accepted to the Journal of Institute of Control, Robotics and Systems.
손재락, 심재훈†, 김상현†, “유계 참조 보상과 조향률 제약을 통한 이중 조향 이동로봇의 정밀 도킹 성능 개선,” 제어·로봇·시스템학회 논문지, 2026.