Seung‐Joon Yi

About

Seung-Joon Yi is a robotics researcher whose work spans humanoid locomotion, push recovery control, quadruped robotics, and autonomous manipulation — areas where robust real-world performance demands tight integration of hardware design, machine learning, and control theory. Yi's most influential contributions center on making bipedal robots resilient to real-world instability. His 2011 series of papers on push recovery control — collectively drawing over 110 citations — introduced practical hierarchical strategies and online learning methods enabling humanoid robots to maintain balance on uneven terrain under unexpected collisions, without requiring precise environmental models or specialized hardware. These foundational ideas were later extended in his 2016 whole-body balancing walk controller, which moved beyond conventional ZMP-based approaches toward more human-like dynamic responses. Yi's work on competitive robotics is equally notable. His contributions to Team THOR's entries in the 2013 and 2015 DARPA Robotics Challenge demonstrated modular, deployable systems under severe real-world constraints, earning broad recognition in the disaster robotics community. More recently, he has expanded into open-source quadruped platforms, high-speed robotic assembly, and intelligent service robots, reflecting a career-long commitment to accessible, capable autonomous systems. With over 300 cumulative citations, Yi's research has meaningfully shaped modern humanoid and mobile manipulation robotics.

Research Focus

Key Achievements

Top Papers

1. 1
   Online learning of a full body push recovery controller for omnidirectional walking

   52 citations · 2011
2. 2
   Team THOR's Entry in the DARPA Robotics Challenge Trials 2013

   46 citations · 2015
3. 3
   Learning full body push recovery control for small humanoid robots

   35 citations · 2011
4. 4
   Practical bipedal walking control on uneven terrain using surface learning and push recovery

   31 citations · 2011
5. 5
   Design and Control of a Open-Source, Low Cost, 3D Printed Dynamic Quadruped Robot

   30 citations · 2021
6. 6
   High-Speed Autonomous Robotic Assembly Using In-Hand Manipulation and Re-Grasping

   25 citations · 2020
7. 7
   Design of a Sliding Mode Controller with Fuzzy Rules for a 4-DoF Service Robot

   23 citations · 2021
8. 8
   Mobile Manipulation for the HSR Intelligent Home Service Robot

   22 citations · 2019
9. 9
   Whole-Body Balancing Walk Controller for Position Controlled Humanoid Robots

   21 citations · 2016
10. 10
    Team THOR's adaptive autonomy for disaster response humanoids

    17 citations · 2015

Key Collaborators

Contact & Links