TECoSA Seminar – Semantics for Robotics

We aim to bring you a TECoSA Seminar at kl.15 on the first Thursday of each month. This Autumn they will once again be on-line, and all are welcome to join (members accept the Outlook invite, non-members please email “tecosa-admin@kth.se”). Each invited speaker will talk for about 40 minutes, followed by a panel discussion coordinated by TECoSA members.

We end  the seminar series for 2021 with a talk by Prof Amy Loutfi, Head of the AASS Machine Perception and Interaction Lab and Pro-Vice Chancellor at Örebro University.  Please see the abstract and short biography below.  To read more about the lab and Prof Loutfi’s research, please see: https://mpi.aass.oru.se/ and https://www.oru.se/english/employee/amy_loutfi

Panel: Iolanda Leite (Chair),  Rafia Inam (Ericsson), Alexis Linard (KTH)

Semantics for Robotics

ABSTRACT: As humans and robots start to cohabit the same environment and interact deeper with each other, the need for robots and other artificial agents to truly understand human instruction increases. This understanding not only involves the ability for robots to decode human speech, but also the ability to ground words that refer to physical objects, actions, and concepts –  to the perceptual information coming from the sensors embedded on the agent. This challenge in part deals with the symbol grounding problem, that is, how to ground symbols into something other than symbols.  Further with the advent of advances within representation learning, due to many of the advances within sub-symbolic architectures there has been an opportunity to examine how the grounding can be learned using large (and simulated) datasets. This talk will highlight the research at Örebro University where semantics are learned, integrated and used for robotic systems.

BIO: Amy Loutfi graduated as an electrical engineer from the University of New Brunswick in Canada in 2001.  She received her Ph.D. in Computer Science in 2006 with a research topic in machine perception. Specifically, she researched about how gas sensors could be integrated onto robotic platforms and how these robots can interact with humans in order to solve a range of problems that required sensing and perception. She has since broadened her research interests to include general research directions within machine perception, where AI methods like Machine learning are used for the interpretation of sensor data. She also has broadened her research in the area of Human Robot Interaction where she has studied HRI in various platforms that include fully autonomous robots, but also teleoperated robots. She has a long experience working with industry and the public sector on research projects dealing with AI, robotics and human-robot interaction.