Kookmin University Team's Robotics Paper Accepted to IROS 2026

Kookmin University's research team, led by Professor Lee Sung-won of the Department of Electronic Engineering, has achieved a significant milestone by getting a paper accepted to the prestigious 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026). This marks another success for the university's robotics and AI research, following their earlier paper acceptance at ICRA 2026.

The achievement aligns with Kookmin University's "KMU VISION 2035: EDGE" initiative, which emphasizes strengthening future competitiveness through eight specialized fields. The IROS 2026 paper demonstrates the tangible results of the university's accumulated research prowess in robotics and AI+X.

Notably, the research involved undergraduate student Kim Jun-ho as the first author. Kim had also co-authored the ICRA 2026 paper, "VG3T: Visual Geometry Grounded Gaussian Transformer." His consecutive publications in world-class international conferences while still an undergraduate highlights his exceptional talent and the university's supportive research environment.

“

I am very pleased that the research I participated in as an undergraduate has been recognized at IROS, following ICRA.

The paper accepted by IROS 2026, titled "RayOcc: Occlusion-Agnostic Ray Occupancy Estimation via Gaussian Mixture Intensity," introduces a novel ray-based occupancy estimation technique robust to occlusions in complex 3D environments. The team designed a new framework using Gaussian Mixture Intensity to effectively estimate spatial occupancy even with limited visual information or partial occlusions. This research addresses a critical challenge in 3D spatial understanding for autonomous driving and robotics systems, offering a potential solution for handling observational uncertainties and occlusions in real-world scenarios.

Kim expressed his deep satisfaction with the recognition from IROS, following the earlier acceptance at ICRA. He stated his ambition to continue AI research that can contribute to real-world robotic and autonomous driving systems. The paper is scheduled to be presented at IROS 2026 in Pittsburgh, USA, in September 2026.

“

Based on my experience in research design, implementation, and experimentation, I want to continue AI research that can contribute to actual robot and autonomous driving systems.