Title: Intelligent Robots: Machines that Act, Learn, and Adapt by themselves

Speaker

Poramate Manoonpong


Associate Professor of Embodied Artificial Intelligence (AI) & robotics and a co-founder of Embodied AI & Neurorobotics Lab (http://ens-lab.sdu.dk/)
Center for BioRobotics (CBR), at the University of Southern Denmark.

Abstract

There is an increasing demand of using robots in our everyday life for a variety of applications, like service, medicine, healthcare, inspection, industry, household, etc. These robots will dynamically interact with the (unpredictable) real world. Dealing with this, they have to autonomously perform multiple functions (locomotion, manipulation, and navigation) as well as learn and adapt to new situations. There are many key challenges that need to be addressed to develop such intelligent robots. In this talk, I will discuss these challenges and present our bio-inspired approach which brings the goal of creating intelligent robots (machines that act, learn, and adapt by themselves) a little bit closer.


  Autoiography:

Poramate Manoonpong is an Associate Professor of Embodied Artificial Intelligence (AI) & robotics and a co-founder of Embodied AI & Neurorobotics Lab (http://ens-lab.sdu.dk/), part of Center for BioRobotics (CBR), at the University of Southern Denmark. He was the PI of the Emmy Noether research project for "Neural Control, Memory, and Learning for Complex Behaviors in Multi Sensori-Motor Robotic Systems" at Bernstein Center for Computational Neuroscience (BCCN) Goettingen in Germany. Currently he serves on an Associate Editor of Frontiers in Neuroscience (Neurorobotics) and the editorial board of International Journal of Advanced Robotic Systems (ARS), (Topic: Bioinspired Robotics) and Advances in Robotics Research, Techno press. He has published over 100 publications in major robotics journals and conferences.

The central goal of his research is to understand how brain-like mechanisms and biomechanics can be realized in artificial agents so they can become more like living creatures in their level of performance. According to this, his team has developed bio-inspired behaving systems with general neural architectures and could show that these agents can acquire complex behaviors with learning and adaptation (resulting in high level contribution to Nature Physics). In addition to this, his team also focuses on transferring biomechanical and neural developments of robots to other applications, like brain machine interfaces, prosthesis and orthosis control.

Joomla templates by a4joomla