Main research areas

The research of the lab aims at the development of behavior-oriented Machine Intelligence based on bioinspired information processing and learning techniques. One main focus is the evolution and organization of navigation and interaction behaviors in actively learning multi-modal Human-Robot systems. 

From methodological view, the following topics are of particular interest:

• Neural and probabilistic techniques for robot navigation and Human-Robot-Interaction (HRI)
• Real-time approaches to person detection, tracking, and re-identification in real-world video and laser data streams
• Deep learning approaches for image-based object recognition, semantic segmentation, pose recognition, grasp pose estimation, and action recognition
• Non-verbal human-machine communication by gestures and facial expression 
• Information-theoretic and Reinforcement Learning for problem decomposition, evolution and coordination of behaviors

Range of services

• Interactive mobile assistance and guidance robots for public environments (Shopping centers, airports, museums, authorities, etc.) 
• User-adaptive social robots for home assistance in context of demographic change 
• Intelligent and adaptive video surveillance systems
• Actively learning and self-optimizing control for large-scale industrial combustion processes (coal-fired power plants, cement, waste combustion)
• Automated deep-learning-based crack and surface distress detection in images of public roads
• Deep learning in industrial manufacturing scenarios for productivity optimization and pro-active learning of robotic actions to support human workers

Special equipment

• 1 mobile, interactive service robots (SCITOS G3) equipped with multimodal sensors (color and depth cameras, laser-scanner, sonar)
• 5 mobile, interactive service robots (SCITOS A5) equipped with multimodal sensors (color and depth cameras, laser-scanner, sonar)
• 1 mobile interaction robot PERSES (B21)
• 1 outdoor robot MILVA 
• 2 experimental robots HOROS (Pioneer II) equipped with multimodal sensors

Prof. Dr. Horst-Michael Groß TU Ilmenau Department of Computer Science and Automation Helmholtz-Platz 5 Zusebau, Z3060 98693 Ilmenau Phone +49 3677 69-2858 Fax +49 3677 69-1665 fg-nikr@tu-ilmenau.de www.tu-ilmenau.de/neurob