| Date: | 16 September 2026, 14:00 – 15:15 (CET) | Price: | Free (for eligible participants) | |
| Format: | AI:AT Webinar, ONLINE (Zoom) | Target audience: | Industry, public admin, academia | |
| Language: | English | Organizer: | AI:AT |
How can robots see, decide, and act? This 3-part webinar series shows how AI connects visual perception, world understanding, planning, and control, revealing how robotic systems move from raw sensor data to intelligent action in real environments.
This webinar series offers an accessible introduction to AI-based robotic vision and control, with a strong focus on how perception, planning, and action are integrated into intelligent robotic systems. Designed for a broad audience, the series will break down complex ideas to accessible explanations and examples, without requiring a deep technical background.
Building on the foundations established in the first session, this webinar dives into how modern AI systems encode objects, scenes, and spatial relationships to support robotic reasoning and planning. It introduces spatially aware perception technologies - including neural representation learning and pose-aware object detection - that enable robots to localize and interpret objects in 3D space for navigation and manipulation tasks. The session then turns to decision making, providing an intuitive explanation of emerging paradigms such as Vision-Language Models (VLMs) and Vision-Language-Action (VLA) systems, which connect perception, language understanding, and action generation. In contrast, neurosymbolic approaches are presented as a complementary direction combining data-driven learning with structured, rule-based reasoning. The session highlights the fundamental differences between these approaches - including their strengths in generalization, interpretability, and structure - as well as their respective limitations in real-world robotic applications.
This session builds on From Computer Vision to Robotic Understanding and prepares the ground for From Planning to Action in Robotic Systems.
Target audience
This webinar is open and free of charge for all participants from academia, industry, and public administration from EU and/or EuroHPC JU member countries.
Entry level & prerequisites
Basic - no previous experience required.
Agenda
- Neural representation learning for robotic vision (25 mins): How modern AI systems encode objects, scenes, and spatial relationships to support downstream reasoning and planning.
- Pose-aware object detection and task-oriented perception (20 mins): Overview of technologies for detecting, localizing, and interpreting objects in 3D space, with emphasis on how these capabilities support robotic navigation and manipulation.
- Decision making and task planning (30 mins): Introduction to VLMs, VLAs, and neurosymbolic approaches; comparison of their capabilities, limitations, and use cases for robotic decision making.
Speakers
Dr. Csaba Beleznai (AIT Austrian Institute of Technology)
Dr. Csaba Beleznai is a Senior Scientist at the AIT Austrian Institute of Technology, bringing over 26 years of expertise in Computer Vision, Machine Learning, and Deep Learning. His research focuses on solving complex 3D object pose estimation tasks essential for robotic navigation and object manipulation, as well as developing advanced vision-based perception models for broader robotic applications. Dr. Beleznai holds an M.S. from the Technical University of Ilmenau (Germany) and a Ph.D. in Physics from Claude Bernard University, Lyon (France). Committed to education, he regularly lectures at various summer schools and universities, including FH Wiener Neustadt and FH Technikum Wien. You can connect with him on Linkedin via: https://www.linkedin.com/in/csaba-beleznai-78b3821/
Dr. Patrik Zips (AIT Austrian Institute of Technology)
Dr. Patrik Zips is a Senior Scientist at the AIT Austrian Institute of Technology, working in the Center for Vision, Automation and Control. He has over 15 years of experience in robotic task and motion planning, with a strong focus on developing algorithms that are robust and feasible in real-world settings involving highly nonlinear dynamics. His research bridges theory and practice, with many of his methods deployed and validated in AIT’s large-scale robotics lab, an experimental environment for autonomous utility machines operating in challenging outdoor conditions. Dr. Zips holds a PhD in Automation and a Master’s degree in Electrical Engineering from TU Wien. LinkedIn: https://www.linkedin.com/in/patrik-zips-b95523286/
Language
English
Date, time, and location
16.09.2026, 14:00 – 15:15 (CET), LIVE ONLINE WEBINAR (Zoom)
Registration
Registration is required, the link to the registration form can be found on the top of this page in the menu on the left. Please register with your official institutional email address to prove your affiliation.
You will get the Zoom link in the automatic confirmation by email (subject starting with "[Indico] Registration"), please check your Spam/Junk folders). We will also send 1–2 reminders before the webinar.
Please do not hesitate to contact us at training@ai-at.eu if you have any questions.
Organizers
This webinar is organized by AI Factory Austria AI:AT.
Acknowledgements
AI Factory Austria AI:AT has received funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 101253078. The JU receives support from the Horizon Europe Programm of the European Union and Austria (BMIMI / FFG).
