Universität Bonn

Institute of Computer Science

16. March 2026

Bonn’s domestic service robots defend their title Bonn’s domestic service robots defend their title

Prof. Dr. Sven Behnke and the NimbRo@Home team succeed once again at the RoboCup German Open

The NimbRo@Home team from the Institute of Computer Science confidently defended its championship title in the @Home League at the 2026 RoboCup German Open — finishing well ahead of the runner-up. From March 10 to 14, the assistive robots demonstrated their capabilities in everyday environments in Cologne. The household robots understood complex voice commands from people in need of assistance, navigated safely through a simulated apartment and an unfamiliar restaurant, delivered orders, and removed laundry from the washing machine.

Das Team NimbRo@Home der Universität Bonn
Das Team NimbRo@Home der Universität Bonn - hat beim German Open-Wettbewerb der Haushaltsroboter seinen Titel verteidigt. © Universität Bonn
Download all images in original size The impression in connection with the service is free, while the image specified author is mentioned.
Please fill out this field using the example format provided in the placeholder.
The phone number will be handled in accordance with GDPR.

The RoboCup German Open Championships are held annually. In her opening speech, Federal Minister of Research Dorothee Bär emphasized that the participants' infectious energy and the competition's innovative spirit send a clear signal about the future of German robotics. In the @Home League, service robots are tested for their ability to support people in need in everyday environments. These intelligent domestic robots exemplify AI-based, multipurpose robots, which the German government aims to promote through flagship projects in the AI Robotics Booster of the High-Tech Agenda Germany.

New Rules

This year, the competition rules underwent a major overhaul. In the preliminary round, the robots had to demonstrate their abilities in five complex tests in a realistic domestic environment and an unfamiliar restaurant setting. The robots were required to welcome visitors and introduce them to others, tidy the kitchen, set the breakfast table, remove laundry from the washing machine, and fold it. They also had to serve guests in the restaurant. The general-purpose service robot test was particularly challenging, as it required the robots to understand complex voice commands from people requiring assistance and to carry them out autonomously.

New Assistance Robot

In addition to last year's tried-and-tested domestic service robot, the NimbRo team from the University of Bonn deployed a new, lighter assistance robot for this competition. This robot has a human-like upper body with two articulated arms that end in three-fingered hands with tactile sensors. The robot perceives its surroundings using cameras, laser scanners, and a microphone. It navigates sefely on an omnidirectional base in confined indoor spaces and was used to remove laundry from the washing machine.

A clear victory

NimbRo took an early lead in the preliminary round and extended it in the final. There, the robot from Bonn demonstrated its ability to open the front door, close the dishwasher, and understand complex voice commands. NimbRo won the overall standings with 200 points, far ahead of Team ToBi from Bielefeld University, which scored 57 points.

Research at the Institute of Computer Science

The Bonn-based household robots are being developed at the Chair of Autonomous Intelligent Systems at the University of Bonn. Prof. Dr. Sven Behnke, Director of Department VI at the Institute of Computer Science, Intelligent Systems, and Robotics, as well as Head of the Embodied AI research area at the Lamarr Institute for Machine Learning and Artificial Intelligence, said after the competition: "Our research aims to enable people in need of assistance to live independently for longer in their familiar surroundings." Behnke is also a member of the Transdisciplinary Research Areas (TRA) "Modelling" and “Sustainable Futures” as well as the Cluster of Excellence “Phenorob” at the University of Bonn. Since everyday environments present numerous challenges, much research and development work, as well as transfer to practical applications, is still required. Methods of artificial intelligence and machine learning are crucial for this.

What can I get for you?
What can I get for you? - In the restaurant test, the domestic robot operated in an unfamiliar restaurant setting. It introduced itself to the waitstaff, recognized guests who waved, took orders, and served food and drinks. © Maximilian Waidhas | Universität Bonn
A helpful assistant
A helpful assistant - In the new “Laundry” task, Team NimbRo's domestic robot opened the washing machine and took out the laundry. © Maximilian Waidhas | Universität Bonn
How can I help you?
How can I help you? - In the test for general-purpose service robots, the robots had to understand naturally phrased user commands ranging from “fetch something from a room” to “find a person” and “answer questions about the apartment.” The computer science department’s domestic robot correctly understood three out of three commands, demonstrated a plan for their execution, and managed to partially execute two commands within the limited time. © Maximilian Waidhas | Universität Bonn

RoboCup German Open 2026:
https://robocup.de/germanopen-2026

Domestic robots NimbRo of University of Bonn:
http://www.ais.uni-bonn.de/nimbro/@Home

News release of Uni Bonn: https://www.uni-bonn.de/en/news/044-2026?set_language=en 

Prof. Dr. Sven Behnke
Institute of Computer Science, Department VI – Intelligent Systems and Robotics
Chair of Autonomous Intelligent Systems
University of Bonn
Phone: +49 228 73 4116
Mail: behnke@cs.uni-bonn.de

Rights
Universität Bonn
Wird geladen