Papers
9
Total Citations
197
H-Index
8
About
Keunwoo Jang is a leading researcher in whole-body control and autonomous manipulation for high-degree-of-freedom robotic systems, including humanoids and mobile manipulators. His core contributions lie in developing robust, real-time motion generation frameworks, most notably through hierarchical quadratic programming (HQP). Jang’s pioneering work on continuous task transition and regularized HQP (cited over 60 times) enables robots to smoothly and safely execute multiple, strictly prioritized tasks simultaneously—a critical advancement for complex operations like door traversal and furniture assembly. He has also made significant strides in contact-rich manipulation, introducing a Gaussian mixture model for robust contact state estimation during peg-in-hole assembly (33 citations), and in reactive self-collision avoidance using a novel distance buffer border concept. Beyond theoretical advances, Jang’s impact is demonstrated through applied achievements, including leading the TEAM SNU avatar system to the finals of the ANA Avatar XPRIZE, showcasing intuitive tele-existence. With over 200 total citations, his work bridges optimal control, motion planning, and practical robotics, providing foundational tools for autonomous systems operating in unstructured, human-centric environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7Robotic furniture assembly: task abstraction, motion planning, and control13 citations · 2022
- 8
- 9