Inha University Presents Four Papers at World's Top Robotics Conference

Society|
|
By Ahn Jae-kyun
|
Inha University presents 4 papers at world's top robotics conference - Seoul Economic Daily Society News from South Korea
Inha University presents 4 papers at world's top robotics conference

Inha University has achieved the distinction of presenting four regular papers at IEEE ICRA 2026, the world's most prestigious robotics academic conference.

According to Inha University on the 26th, the Spatial Intelligence and Robotics Laboratory led by Professor Cho Young-geun of the Department of Electrical and Electronic Engineering will present four papers on autonomous driving and recognition technologies at the conference. Two papers were accepted for presentation at ICRA 2026, while the remaining two earned publication in IEEE RAL (Robotics and Automation Letters).

The first study, "KISS-IMU," proposed a technology enabling IMU (Inertial Measurement Unit) sensors to learn autonomously without separate ground truth data. The system uses LiDAR sensors to precisely calibrate movement paths and reflects motion balance and computational reliability to enable stable position estimation across various environments.

The "GSAT" study developed technology allowing robots to independently determine safe movement ranges based on driving experience in unstructured environments. The system effectively distinguishes driving difficulty levels without large-scale pre-trained models, maintaining performance even in new environments.

The "Freeze-Frame with StaticNeRF" study addressed the problem of dynamic objects such as people and vehicles causing interference in map generation. The research presented technology that processes moving elements as uncertainty information to stably reconstruct only static spatial information.

The "MSG-Loc" study focused on improving robot localization accuracy in complex environments. While conventional methods predict only a single answer, this design considers multiple possibilities simultaneously, enabling stable position estimation even in semantically ambiguous environments. The research was conducted jointly with Professor Shin Young-sik of Kyungpook National University.

"We were able to achieve these results thanks to collaboration with partner institutions and the active participation of our researchers," Professor Cho Young-geun said. "We will continue our research to contribute to the advancement of robotics and autonomous driving technology."

The research was supported by the Phase 4 BK21 project, the Institute of Information and Communications Technology Planning and Evaluation's Human-Centered AI Core Technology Development Project, and the National Research Foundation of Korea's Global Basic Research Laboratory and Excellent Young Researcher Support programs.

Related Video

AI-translated from Korean. Quotes from foreign sources are based on Korean-language reports and may not reflect exact original wording.