Papers
11
Total Citations
177
H-Index
8
About
Nosan Kwak is a leading researcher in mobile robotics, specializing in simultaneous localization and mapping (SLAM) for humanoid robots. His major contributions address the critical "particle depletion problem" in FastSLAM algorithms, where he developed novel compensation techniques and resampling analyses that significantly improve robot pose estimation accuracy. With over 175 citations across his most-cited works, Kwak's research has advanced both theoretical SLAM frameworks and practical robotic applications. He pioneered vision-based 3D motion estimation for humanoid robots, notably through his work on the Roboray platform, which enabled dynamic walking with human-like heel-toe motion. Kwak also made notable contributions to autonomous door manipulation, creating integrated visual recognition systems for doors and knobs, combined with systematic touch schemes for grasping—a foundational achievement in humanoid-environment interaction. His work on 3D grid and particle-based SLAM using stereo vision addressed the challenge of noisy sensor data in home environments, while his adaptive prior boosting techniques optimized computational efficiency in particle filtering. Kwak's research continues to influence modern SLAM approaches and humanoid robot autonomy.
Research Focus
Key Achievements
Top Papers
- 1
- 23D grid and particle based SLAM for a humanoid robot25 citations · 2009
- 3On-board odometry estimation for 3D vision-based SLAM of humanoid robot22 citations · 2012
- 4
- 5Adaptive prior boosting technique for the efficient sample size in fastSLAM16 citations · 2007
- 6Visual recognition of a door and its knob for a humanoid robot13 citations · 2011
- 7Result representation of Rao-Blackwellized particle filtering for SLAM10 citations · 2008
- 8
- 9Systematic touch scheme for a humanoid robot to grasp a door knob8 citations · 2011
- 10