Zafeirios Fountas

Research scientist @ Emotech, research associate @ UCL, visiting lecturer @ RCA







TimeStorm

comming soon...










Arizona speeding

comming soon...










My PhD

comming soon...










Emotech's Olly

comming soon...










Neurobot

NeuroBot is an A.I. system able to control avatars (bots) within the environment of the computer game Unreal Tournament 2004, in a human-like manner. To achieve this, NeuroBot is based around a large-scale modular spiking neural network, of around 20.000 neurons, divided into specialized regions. The high level coordination of this network is carried out by an architecture based on a cognitive theory of consciousness called global workspace theory.

an image an image

The first version of this system as well as the underlying theory was the outcome of my dissertation for the MSc course I attended in the Department of Computing of Imperial College London during the academic year 2010-11, under the supervision of Dr David Gamez and Dr Andreas Fidjeland who provided priceless support at every stage of this project.

The same version of NeuroBot achieved second place in the 2K BotPrize human-like bot competition at CIG 2011 in Seoul, South Korea.


Media
New Scientist cover New Scientist (magazine issue 2881)
Title: AI cyber-fighter: does it feel human, punk?
Date: 06 September 2012
by Celeste Biever
New Scientist online EDGE
Title: Unreal bots beat Turing test: AI players are officially more human than gamers
Date: 28 September 2012
by Keith Stuart


New Scientist online
Bonusweb.cz (In Czech)
Title: Pocitacem rizeny bojovnik chce porazit lidsky mozek. Souperi o 137 tisic
Date: 11 September 2012
by Andrej Brabec
Videos

The following video illustrates a part of the graphical user interface and the performance of the system during a human-versus-bot test game.

Downloads

The presentation slides of my MSc thesis provide a brief description of this system, while a complete description can be found on my final report. Also, the neural architecture of this system is described in this paper which was presented and published in the proceedings of the CIG 2011 (IEEE Conference on Computational Intelligence and Games) in Seoul, South Korea.

Finally, a running and stable version of this software can be downloaded from this link, while its sourcecode can be found on NeuroBot's sourceforge webpage and the most recent version can be provided under request.










PANDORA

P.A.N.D.O.R.A. robotics research team (Program for the Advancement of Non Directed Operating Robotic Agents) was found in 2007 by a number of engineering and science students (including me), who wanted to apply the knowledge they accumulated over their studies on a real project. It was found under the Electrical and Computer Engineering Department of Aristotle University of Thessaloniki and today, still at the same department, it consists of 5 faculty members and 22 undergraduate students. Over the years, PANDORA's main goal has been the development of robotic vehicles that can navigate areas having suffered a natural disaster – like an earthquake – and are able to locate human victims. Using the produced robotic vehicles, PANDORA team has participated in the RoboCup - Rescue competition four times: In 2008 in China, in 2009 in Austria, in 2011 in Istanbul and more recently in 2013 in Netherlands. For more information, visit here.

2007: Team's first year. Construction of a robotic Arm.
RoboArm1 RoboArm2 RoboArm3

2008: Team's second year. First robo-rescue platform.

2008: SFHMMY 2 (link)

SFHMMY1 SFHMMY2 SFHMMY3

Robocup World Championship 2008, Suzhou, China

RobocupChina1 RobocupChina2

2009: Robocup World Championship, Graz, Austria
RobocupAustria1 RobocupAustria2 RobocupAustria3 RobocupAustria4






© Zafeirios Fountas This is just a joke. As a kid, I always wanted to have copyright material with my name but you can obviously copy anything you like from the sourcecode of this website (only sourcecode not the content) without my permisison :-)
(2019)
Last Update on January 30 2019. Number of visits: 29132