7/28/2019 19107742

1/3

IN THIS FEATURE, we will try to describe the characteristics of current virtual reality (VR) research in Eu-rope. In particular, CyberEurope aims at describing the leading research groups and projects runningon the other side of the Ocean.

Within the research areas of the European Union, the Future and Emerging Technology (FET) programis the information and technology program nursery of novel and emerging scientific ideas:. Its mission is to promote research that is of a long-term nature or in-volves particularly high risks, compensated by the potential of a significant societal or industrial impact.

In this context, an emerging area is imitative robotics, a growing area of robotics whose main goal isthe development of living artificial prostheses that learn through imitation.

We know that the success of the human species critically depends on our extraordinary ability to en-gage in joint action. Our perceptions, decisions, and behavior are tuned to those of others with whom weshare beliefs, intentions, and goals, and thus form a group. These insights underlie the motivation of the

researchers working in this area to develop jointly acting autonomous systems that communicate andwork intelligently on mutual tasks in dynamic unstructured environmentsa goal that is far-reaching be-yond studying individual cognitive systems and that will expand the concept of group to human plusartificial agent(s). Below we discuss the outcomes of two Euroean projects working in this area.

PROJECT: ARTEFACT STRUCTURAL LEARNING THROUGH IMITATION (ARTESIMIT)

Contact: KNOLL, AloisOrganization: Technische Universitaet Muenchen, Lehrstuhl Informatik Vi Fakultaet Fuer Informatik,

Arcisstrasse 21, 80333 Muenchen, Germany. Tel: +89-28-918106. Fax: 89-28-918107. E-mail: knoll@in.tum.de.The last decade has seen enormous advances in the design and implementation of all kinds of robot

platforms, says Alois Knoll of Technische Universitt Mnchen and coordinator of the ArteSImit project.However, their abilities to learn how to solve even simple tasks are still very restricted.According to Knoll imitation is a powerful shortcut to successful task learning. If a robot learner is

shown a certain behavior necessary for solving a task by an instructor, the motion alternatives for produc-ing the goal-reaching behavior with the learners own devices and effectors can be dramatically reduced,says Knoll. The learner would then try to mimic the coarse motions of the instructor and adapt them toachieve the same goal.

Exploring the functional and neuropsychological mechanisms of imitation learning was the aim of theArteSImit project funded by the Future and Emerging Technologies initiative of the IST programme. Theoverarching goal was to reveal the neurophysiological structures for finger and hand movements in hu-mans and monkeys, and design a computer-operational dynamic model of imitation learning. The con-sortium constructed a fully functional visual-motor system for a limited domain to control a bio-analoguehand, which is thus capable of mimicking imitation behavior.

The system implements the full visual serving loop in that it observes the environment with a camera,recognizes the instructors gestures from a set of predefined gestures, makes the necessary decisions toachieve the goal, i.e., the appropriate sequence of actions of the arm, hand and finger movements. This isoptimized in such a way that it is achieved with the minimum amount of motion. The visual system ishighly innovative and unique: it recognizes finger positions in a precise way through a novel algorithmwith just one camera.

Our final objective was to implement imitational learning on this artificial hand and to suggest appli-cations of the new methods and models to other artifacts with many degrees of freedom that cooperatewith humans, says Knoll.

We believe that the methods we developed in the ArteSImit project could be directly applied, for ex-ample, to new generations of prostheses. These could react for example to finger movements, controlled

CYBERPSYCHOLOGY & BEHAVIORVolume 8, Number 6, 2005 Mary Ann Liebert, Inc.

CyberEurope

605

7/28/2019 19107742

2/3

606 CYBEREUROPE

by a camera applying the concept of imitation, and additionally, they could be linked directly to thenerves.

PROJECT: JOINT-ACTION SCIENCE AND TECHNOLOGY (JAST)

Contact: BEKKERING, HaroldOrganization: Stichting Katholieke Universiteit, Geert Grooteplein-Noord 9, Postbus 9102, 6525 Ez Nij-

megen, Netherlands. Tel: +31.24.361.2632. Fax: +31.24.361.6066. E-mail: H.Bekkering@nici.ru.nl. Website:http://www.jast-net.gr/.

Research will be continued within the follow-up project JAST, which took up some of the ArteSImitfindings. JAST aims to develop jointly-acting autonomous systems that communicate and work intelli-gently on mutual tasks in dynamic unstructured environments expanding the concept of group to humanplus artificial agents.

By combining a basic, gender-differentiated understanding of the cognition, neurobiology and dialoguestrategy of joint action JAST specifically aims at:

Building two autonomous agents, each endowed with a vision system, a gripper and a speech recogni-tion/production system that in cooperative configurations can carry out complex construction task

Implementing perceptual modules for object recognition and recognition of gestures and actions of thepartner (human or robot)

Implementing control schemes that generate motor behaviour on the basis of internal forward modelsfor the co-ordinated action of multiple cognitive systems

Implementing verbal and non-verbal communication structures, Developing autonomous systems with goal-directed and self-organizing learning processes Implementing an error monitoring system capable of reacting intelligently to self- and other-generated

errors.

Potential applications of imitative robotics are numerous. It could be applied to control instrumentsused for medical operations or even to capture and detect a persons movement within a crowd of people.This method could be interesting for example for security systems, Knoll says. Although some of the re-sults were mostly interesting for further basic research, there are some results, e.g. in the area of computer

vision and object tracking, that could be phased out and be developed into useful products in their ownright.

Compiled by Giuseppe Riva, Ph.D.Istituto Auxologico Italiano

Milan, Italy

7/28/2019 19107742

3/3