Invited speakers

Soon you will find here a complete list of the invited keynote speakers with abstract.

Danica Kragic

Danica Kragic is a Professor at the School of Computer Science and
Communication at The Royal Institute of Technology (KTH), Stockholm. She
received PhD in Computer Science from KTH in 2001. She has been a visiting
researcher at Columbia University, Johns Hopkins University and INRIA
Rennes. She is the Director of the Centre for Autonomous Systems at KTH
and the coordinator for the EC integrated project GRASP and TOMSY. Danica
received the 2007 IEEE Robotics and  Automation Society Early Academic
Career Award. She is a member of the Royal Swedish Academy of Sciences and
Young Academy of Sweden. She is charing the IEEE RAS Technical Committee
on Computer and Robot Vision and from 2009 serves as an IEEE RAS AdCom
member. Her research is in the area of robotics, computer vision and
machine learning.
Title:
Learning to grasp and manipulate objects
 
Abstract:
For humans and robots alike, objects in the environment provide context
for interactions, tools for executing tasks and means for grounding
semantics.  In robotics, an important open problem is to detect, recognize
and categorize objects given sensory data, both prior to and during
interaction with objects. Central to solving this problem is to represent
and parametrize sensory data so to provide fast, robust and scalable
solutions. In this talk, we focus on representations that not only
facilitate detection, recognition and categorization but also on-line
learning of object models. We present different scenarios
where structural representations and models are of key importance for
achieving fast and robust performance. One example is transferring
grasping strategies across objects that share similar parts. The agent
transfers grasps across objects by identifying, from examples provided by
a teacher, parts by which objects are often grasped in a similar fashion.
It then uses these parts to identify grasping points onto novel objects.
We also address the problem of grasp generation and grasp transfer between
objects using categorical knowledge.  Our system is built upon an
i)~active scene segmentation module, able of generating object hypotheses
and segmenting them from the background in real time, ii)~object
categorization system using integration of 2D and 3D cues, and
iii)~probabilistic grasp reasoning system.  Individual object hypotheses
are first generated, categorized and then used as the input to a grasp
generation and transfer system that encodes task, object and action
properties.

Giulio Sandini is Director of Research at the Italian Institute of Technology and full professor of bioengineering at the University of Genoa. After his graduation in Electronic Engineering (Bioengineering) at the University of Genova in 1976 he was research fellow and assistant professor at the Scuola Normale Superiore in Pisa until 1984. He has been Visiting Research Associate at the Department of Neurology of the Harvard Medical School in Boston where he developed diagnostic techniques based on brain electrical activity mapping. After his return to Genova in 1984 as associate professor, in 1990 he founded the LIRA-Lab (Laboratory for Integrated Advanced Robotics, http://www.liralab.it). In 1996 he was Visiting Scientist at the Artificial Intelligence Lab of MIT.
Since July 2006 Giulio Sandini is Director of Research at the Italian Institute of Technology where he has established and is currently directing the department of Robotics, Brain and Cognitive Sciences. RBCS department concentrates on a multidisciplinary approach to human centered technologies encompassing machine learning and artificial cognition, exploring the brain mechanisms at the basis of motor behavior, learning, multimodal interaction, and sensorimotor integration.
The department’s multidisciplinary research staff is composed of researchers with different backgrounds (engineers, biologists, psychologists, mathematicians, physicists, medical doctors) addressing three, strictly interconnected, streams of research: Humanoid Cognition, Human Behavior and Biomechanics, Brain Machine Interface. Giulio Sandini is author of more than 300 publications and five international patents.

Christian Ott

Christian Ott received his doctoral degree in Automatic Control from the University of Saarland, Germany, in 2005. He was a visiting researcher at the University of Twente, and worked as a project assistant professor at the University of Tokyo. Currently he is a group leader of a Helmholtz
Young Investigators Research Group in the Institute of Robotics and Mechatronics at the German Aerospace Center (DLR). He received several scientific awards including the “Conference Best Paper Award” at HUMANOIDS 2011, the “Dr.-Eduard-Martin”-Prize 2007, the Industrial Robot Outstanding Paper Award 2007, ICRA Best Video Award 2007, and the Best Paper Awards at the VDI-Conference “Mechatronik-2005” and from “at-Automatisierungstechnik” in 2005. His main research interests are the application of nonlinear control methods to robotic systems, force and impedance control, and control of bipedal humanoid robots.

Titel:
Control of bipedal humanoid robots with torque sensors

Abstract: This talk gives an overview of the state of the art on bipedal humanoid control and presents some recent results which utilize joint torque sensing. While torque sensing and control has shown advantages in manipulation tasks with arms and hands, it is still not widely used in legged robots. For compliant balancing an approach is presented, which uses ideas from robot grasping to distribute the necessary control forces for compliant balancing to the available contact points. In this mapping, a constraint optimization is employed in order to limit the interaction forces to the contact constraints. The resulting contact forces are then generated via joint torque control. This torque based approach allows for stable balancing on uncertain ground conditions. For gait stabilization, an extended capture point control law is presented which is not restricted to a simplified concentrated mass model and experimental results with the fully torque controlled robot TORO are shown. Finally, a bio-inspired control approach is presented in which the stability properties of nonlinear limit cycles are utilized for gait generation of fully actuated walking robots.

Luc Steels

Luc Steels studied Computer Science and Electrical Engineering at MIT after a doctorate in computational linguistics from the University of Antwerp (UIA). He was a researcher at the Artificial Intelligence Laboratory at MIT working on message passing systems with Hewitt and Minsky. After his MIT studies he joined the Schlumberger research labs in the US where he became program leader for geological expert systems. Currently Steels is professor of Computer Science and Artificial Intelligence at the Free University of Brussels (VUB) and director of the AI-Lab which he founded in 1983. He is the President of the Belgian Association for Artificial Intelligence and active in various international organizations. He was Program Chairman of the largest European AI conference ECAI in 1986 and of the AISB conference in 1991. Steels is cofounder of Knowledge Technologies, a company specialised in intelligent systems technology, and Science Interface, a company which produces multi-media products for scientific education. He has been a consultant to a large number of IT companies in Europe, including Philips, Siemens and IBM. He has been visiting professor at the universities of Vienna, Beijing, Milan, Sophia-Antipolis (Theseus), and Hamburg, and lectures regularly for industrial audiences throughout Europe.

Steels is a member of the scientific board of Electrum (Stockholm) and past scientific board member of the German National AI institute DFKI (Germany). He has been heavily involved in the ESPRIT program as consultant on workplan formulation, reviewer and proposal evaluator. He has also been a consultant to MITI in Japan on the future of information technology.

Title: When will they start to speak? A roadmap for endowing robots with language capacities.

Artificial Intelligence methods and techniques have reached a high level of sophistication so that we can tackle difficult outstanding problems in science. In this talk, I will show how the question of the origins of language can be approached this way. This question has puzzled evolutionary biologists since Darwin and is still considered to be unsolved. I will outline a theory of language evolution by linguistic selection and then report a number of concrete experiments with humanoid robots that attempt to work out and validate this theory. The experiments all center around the notion of a language game, which is a routinized situated interaction that involves some form of language. Robots use linguistic strategies to evolve a communication system to deal with a particular class of language games. I will discuss examples of this and also address the question how new strategies can arise and how the robots can autonomously decide which strategies they will collectively use to bootstrap their language.

 

Noel Sharkey  (cancelled – will not come to HFR12)

Noel Sharkey BA PhD FIET, FBCS CITP FRIN FRSA is a Professor of AI and Robotics and  a Professor of Public Engagement at the University of Sheffield (Department of Computer Science) and a writer, broadcaster and journal editor (Editor in Chief of Connection Science and on the editorial board of 4 other journals). He has held a number of research and teaching positions in the UK (Essex, Exeter, Sheffield) and the USA (Yale, Stanford). Noel has moved freely across academic disciplines, lecturing in engineering, philosophy, psychology, cognitive science, linguistics, artificial intelligence, computer science and robotics. He holds a Doctorate in Experimental Psychology and an honorary Doctorate of Science (UU), is chartered Electrical Engineer and Chartered information technology professional. Noel is a Fellow of the Institute of Engineering and Technology, the British Computer Society, the Royal Institute of Navigation, the Royal Society of Arts and is a member of both the Experimental Psychology Society and Equity (the actors union).

Noel is probably best known for his work on many aspects of neural network learning. Noel appears regularly on TV and is interviewed regularly on radio and in the press.  As well as writing many academic articles, he also writes for the national newspapers and magazines and has been involved in thrilling robotics museum exhibitions and mechanical art installations. He was EPSRC Senior Media Fellow (2004-2010) and is holder of the Royal Academy of Engineering Rooke medal for the promotion of engineering. Noel’s core research interest is now in the ethical application of robotics and AI in areas such as the military, child care, elder care, policing, autonomous transport, robot crime and medicine/surgery. He is currently a Leverhulme Research Fellow on the ethical and technical appraisal of Robots on the Battlefield. http://www.dcs.shef.ac.uk/~noel

abstract: Would you let robots care for your children, mind your aging parents, perform surgery on you, protect your home, transport you and fight your wars? Since the turn of the century, sales of professional and personal service robots have risen sharply  and their numbers already far outstrip the 1.2 million operational industrial robots on the planet. Service robots are good at dull, dangerous, and dirty work, such as cleaning sewers and performing domestic duties. They harvest fruit, pump gasoline, assist doctors and surgeons, dispose of bombs, police us, entertain us, have sex with us and even kill us. This talk will briefly overview today’s service robots and their benefits and then focus on the ethical problems in the trade-offs  with human rights their application creates. The talk is an overview of some of the ethical issues in robot applications from the following papers written over the last 4 years. Several can be found with online search and some can be downloaded from http://bit.ly/mUu1BK.