Papers
11
Total Citations
369
H-Index
8
About
Seungwoo Hong is a leading roboticist whose work is reshaping the capabilities of legged robots, from agile quadrupeds to reactive humanoids. His research centers on model predictive control (MPC), mechanical design, and state estimation, pushing robots to climb ferromagnetic surfaces, run efficiently, and recover from disturbances in real time. His most cited paper, “Agile and versatile climbing on ferromagnetic surfaces with a quadrupedal robot” (109 citations), demonstrates a robot that rapidly traverses floors, walls, and ceilings, vastly expanding operational workspaces. Hong’s contributions to real-time constrained NMPC on SO(3) (48 citations) and whole-body MPC for humanoids (39 citations) have enabled dynamic locomotion without pre-planned contact modes, as seen in his contact-implicit MPC framework (22 citations). He also co-designed KAIST HOUND, a quadruped optimized for fast, efficient running using mixed-integer nonlinear gear train optimization (30 citations), and developed a lightweight cycloidal reducer for legged robots (44 citations). His state estimation algorithm, leveraging dynamic contact events (45 citations), enhances robot awareness in complex terrains. With over 360 total citations and a growing portfolio of high-impact work, Hong is a key figure in advancing the autonomy and mechanical intelligence of legged systems.
Research Focus
Key Achievements
Top Papers
- 1Agile and versatile climbing on ferromagnetic surfaces with a quadrupedal robot109 citations · 2022
- 2
- 3Legged Robot State Estimation With Dynamic Contact Event Information45 citations · 2021
- 4
- 5
- 6
- 7
- 8
- 9
- 10