Welcome to the Institute for Artificial Intelligence!
The Institute for Artificial Intelligence (IAI) directed by Prof. Michael Beetz is part of the Faculty of Computer Science and member of the Center for Computing and Communication Technologies (TZI) at the University of Bremen.
Nowadays humans will teach robots the necessary knowledge directly to make them competent in a specific task. Enabling robots to teach themselves by exchanging knowledge with other robots is the goal of our software platform openEASE.
The IAI will present openEASE at the CeBIT 2017, which is the biggest IT expo in the world. You can find us in Hall 8, booth E01 from the 22nd to 24th of March 2017.
- English: Press release openEASE at CeBIT
Our FP7 project, SHERPA, came to an end with a final review between 1st and 3rd March in Davos, Switzerland. As IAI team, we had the chance to demonstrate our HMI-enabled high-level planing and reasoning in a wide-area rescue scenario. We also developed an Unreal game engine-based simulation with ROS interface and RoboSherlock perception for this rescue mission.
Our demonstration has received very positive feedback from the reviewers and partners. We thank all of the contributors for this successful project. It was a great experience!
We are thrilled to announce that The Institute for Artificial Intelligence has been selected as mentoring organization for the Google Summer of Code 2017 as one out of 201 organizations world-wide. The initiative, started in 2005, sponsors students during the summer months for contributing code to selected open source projects. This is our third year in the Google-sponsored program and we're looking forward to mentoring interested students and introducing them to our software tools and research endeavors.
You can find more information about the proposed projects as well as contact information of the designated mentors at our Google Summer of Code page.
Twelve pupils from Arequipa, Peru interested in studying Artificial Intelligence and Robotics at University of Bremen visited the robots from our lab. Next year is their final year of studies before continuing to university and knowing the international reputation of University of Bremen in Robotics and Artificial Intelligence they chose to visit the research laboratories of our university. It was our pleasure to have them as guests and hence confirm the international reputation of our university. In our lab they interact with our Pepper robot and explored his social capabilities. Furthermore they saw the PR2 robot making popcorn and Boxy robot baking pizza for them. We look forward to having them back as students.
Probabilistic Action Cores (PRAC) system has a new web interface: PRACWeb which allows users to test out on smart devices the capabilities of the system. PRAC system is the implementation of a novel approach for probabilistic high-performance natural-language interpreter. The interpreter has a knowledge-driven approach. It is inspired by concepts of cognitive psychology and models actions as abstract event types. PRACWeb allows the user to generate robot-executable plans from natural-language instructions. PRAC relies on pracmln, which is a toolbox for statistical relational learning and reasoning. It includes tools for standard graphical models (in particular, MLNs). The web interface of pracmln is WebMLN. It allows the users to try out the capabilities of pracmln system.
Prof. Michael Beetz PhD
Head of Institute
Dr. habil. Hagen Langer
Prof. em. Dr. Otthein Herzog
Open Knowledge for