DistantNews
Support us
๐Ÿ‡ฐ๐Ÿ‡ท South Korea /Culture & Society

University of Seoul Team Wins RoboCup 2026 ARM Challenge for Third Consecutive Year

From Hankyoreh · () Korean

Translated from Korean, summarized and contextualized by DistantNews.

At a glance

News Sources not specified Outcome reported
  • The 'UOS Robotics' team from the Department of Mechanical Information Engineering at the University of Seoul won the 'RoboCup 2026 ARM Challenge' in Incheon, South Korea.
  • This marks the team's third consecutive victory in the Autonomous Robot Manipulation Challenge, solidifying their status as world-class in autonomous robot manipulation technology.
  • The competition, held for the first time in Korea, saw 364 teams from 45 countries participate, making it the largest RoboCup event to date.

The 'UOS Robotics' team, comprised of students from the Department of Mechanical Information Engineering at the University of Seoul, has achieved a remarkable third consecutive victory at the 'RoboCup 2026' competition. The team clinched the top spot in the Autonomous Robot Manipulation Challenge (ARM Challenge), held at the Songdo Convensia in Incheon, South Korea.

This latest win further cements UOS Robotics' reputation as a global leader in autonomous robot manipulation technology. Their consistent success, with previous victories in 2022 and 2023 in the same category, highlights their advanced research and development capabilities. The RoboCup, established in 1997, is one of the world's most prestigious robotics competitions, attracting leading researchers and aspiring engineers.

This year's RoboCup 2026 was particularly significant as it was the first time the event was hosted in Korea. It set a new record for participation, with 364 teams and a total of 2,879 participants from 45 countries competing. The ARM Challenge, co-hosted by MathWorks and Universal Robots, required teams to develop sophisticated AI-based algorithms for robots to recognize and manipulate objects like cans and bottles in dynamic environments, specifically performing a "Recycling task."

Our students' tireless research has led to advanced robot vision technology and manipulation algorithms, culminating in another world championship victory. Achieving this in Korea's first-ever, largest RoboCup makes it even more meaningful. I applaud the six students for their excellent teamwork and pledge to continue nurturing innovative robotic talent that can contribute to future industries.

โ€” Professor Hwang Myung-joongAdvisor to the UOS Robotics team, expressing pride in the students' achievement and the significance of the win.

The UOS Robotics team, consisting of six graduate and undergraduate students, utilized MATLAB and Simulink from MathWorks to create advanced AI-driven recognition and control algorithms. After excelling in the simulation-based preliminary rounds, they advanced to the finals, where they demonstrated superior grasping and manipulation skills using the "Universal UR5e" collaborative robot, ultimately securing a significant point lead over their competitors.

Professor Hwang Myung-joong, the team's advisor, expressed immense pride in the students' dedication and the resulting achievement. "Our students' tireless research has led to advanced robot vision technology and manipulation algorithms, culminating in another world championship victory," he stated. "Achieving this in Korea's first-ever, largest RoboCup makes it even more meaningful. I applaud the six students for their excellent teamwork and pledge to continue nurturing innovative robotic talent that can contribute to future industries."

The research results our team members are currently pursuing were actively utilized in performing the competition mission. After recognizing the target object, we generated various grasping poses using generative AI technology, and comprehensively considered the robot's posture and potential for collision to perform the grasp with the highest probability of success. In the finals, the score difference with the second-place team was not large, requiring a strategy adjustment, but we succeeded in the comeback through active communication with the team members.

โ€” Cho Sung-binTeam leader and master's student, crediting the team's collaborative efforts and innovative use of AI for their victory.
DistantNews Editorial

Originally published by Hankyoreh in Korean. Translated, summarized, and contextualized by our editorial team with added local perspective. Read our editorial standards.