The Institute for Artificial Intelligence (IAI) investigates methods for cognition-enabled robot control. The research is at the intersection of robotics and Artificial Intelligence and includes methods for intelligent perception, dexterous object manipulation, plan-based robot control, and knowledge representation for robots.

As a researcher of the IAI, you actively research by applying and extending the elaborated methods and tools of the IAI (e.g., CRAM, KnowRoB, and openEASE) to the uncertainties of limited resources, computation, and sensor data of submarine robots. You develop a Digital Twin that simulates harsh continuously evolving environments and service tasks on real robots. You design a detailed representation of an environment to be inspected, including data coming from past inspections, both similar and different (e.g., transfer learning from an inspection of the interior of an installation). Your research will be especially the environment representation, semantic annotation, and sensor data fusion deal with noisy data and unreliable information in dynamic environments. You work with IAI researchers and engineers of the ROSEN Group, implementing a service for underwater pipeline inspection and maintenance.

Prerequisites: You have a strong background in some of the following fields:

• Digital Twins
• Knowledge Representation
• Data structures
• Data Stream Representation.

Hiring institution: University of Bremen

PhD Enrollment: PhD position in the Institute for Artificial Intelligence at University of Bremen. The project involves collaboration with ROSEN in Bremen and UiO in Oslo.

The PhD examination acceptance requires a “Certificate of Equivalence for Foreign Vocational Qualifications”. More information available at https://www.kmk.org/zab/central-office-for-foreign-education/certificate-of-equivalence.html and https://www.uni-bremen.de/fileadmin/user_upload/sites/international/Studieren_in_Bremen/AnerkennungPromotion.pdf . Obtaining such a certificate is not necessary for the job application, but it is required later in the recruitment process.

Duration of the project: 36 months

Main Academic Supervisor: Prof. Michael Beetz, contact: beetz@cs.uni-bremen.de

Co-supervisors: prof. Einar Broch Johnsen (UiO, Norway), dr. Peter Kampmann (ROSEN, Germany)

If you are looking for a bachelor/master thesis or a job as a student research assistant, you may find some interesting opportunities on this page.

Implementing physics-based grasping models in virtual environments, taking input from various devices such as Manus VR or Valve Index.

Requirements:

• Good C++ programming skills
• Familiar with skeletal animations
• Experience with simulators/physics-/game- engines
• Familiar with Unreal Engine API
• Familiar with version-control systems (git)
• Able to work independently with minimal supervision

Contact: Andrei Haidu

Developing and improving existing or new 3D (static/skeletal) models in Blender / Maya (or other). Further importing and testing the models against Unreal Engine.

Bonus: Working with state of the art 3D Scanners Go!SCAN for creating sketching new models.

Requirements:

• Experience with Blender / Maya (or other)
• Knowledge of Unreal Engine material / lightning development
• Familiar with version-control systems (git)
• Able to work independently with minimal supervision

Contact: Andrei Haidu

Entwicklung einer Mülltrennungs-App für Smartphones (Android) basierend auf Objekterkennung und Anbindung an Wissensgraphen. Dabei wird auf bestehende Objekterkennungs-Apps aufgebaut und eine Smartphone App entwickelt, die Objekte erkennt und auf einen Wissensgraphen zugreift, um weitere Informationen zu diesen Produkten zu erhalten. Eine Aufgabe der Arbeit ist es, den bestehenden Wissengraphen um Wissen über Produktverpackungen und deren Entsorgungsmöglichkeiten zu erweitern.

Aufgaben:

• App Entwicklung mit der Unity game engine
• Objekterkennung mittels Vuforia
• Arbeit mit Wissensrepräsentation und Wissensgraphen
• Erweiterung von .owl Dateien im Wissensgraphen

Contact: Michaela Kümpel

This is a knowledge representation topic including knowledge graphs. The idea is so link external Web-knowledge to an existing knowledge framework in order to include situational awareness so that a robot acting in a household environment can infer what an object is used for in a given situation.

A result would be that a spoon next to a bowl with cereal would be used for eating while a spoon on a stove next to a pot would be used for stirring.

requirements:

• Work with KnowRob knowledge processing framework
• Work with knowledge graphs and Linked Data to create a situational awareness knowledge graph that can be linked to KnowRob
• Implement reasoning about situations (based on perceived environment) and object use

Contact: Michaela Kümpel