A Robotic Platform for Spiking Neural Control Architectures
-
1
Bernstein Center for Computational Neuroscience, Germany
-
2
Freie Universität Berlin, Institute of Biology, Germany
-
3
Freie Universität, Institute of Computer Science, Germany
Spiking neural networks (SNNs) have proven to be excellent control systems in biological organisms. Hence, they should have the potential of providing good control systems for autonomous robots. However, so far only few technical attempts to control artificial agents with SNNs have been made [2].
Here we present a robotic platform for testing biologically inspired SNNs as control architectures. The platform integrates real-time processing of sensory input and motor commands. The sensory data is fed into iqr [1], a software for the real-time simulation of SNNs. We extended its existing models of neurons and synapses to support adaptation and conditional plasticity. All communication between the robot and iqr is done via WiFi. As our platform we use a Rover based on the Arduino micro-controller [DFRobotShop Rover, RobotShop Distribution Inc.,Quebec, Canada]. A vision processing module [HaViMo2.0, fiveam robotics, Berlin] is mounted on the robot. The preprocessed image stream is transformed into spike trains which are fed into a SNN. The network processes this sensory input and generates output that is translated into motor commands. We test the robot on simple associative learning tasks, where we apply a conditional plasticity rule.
Acknowledgements
This project receives funding from the BMBF through grant 01GQ0941, within the Bernstein Focus Neuronal Basis of Learning (BFNL) - Insect Inspired Robots .
References
[1] Ulysses Bernardet and Paul F.M.J. Verschure. iqr: A tool for the construction of multi-
level simulations of brain and behaviour. Neuroinformatics, 8(2):113–134, 2010.
[2] Jeffery L. Krichmar and Hiroaki Wagatsuma, editors. Neuromorphic and Brain-Based
Robots. Cambridge University Press, 2011.
Keywords:
associative learning,
Autonomous Robots,
plasticity,
real-time sensory processing,
Robotics,
spiking neural networks
Conference:
Bernstein Conference 2012, Munich, Germany, 12 Sep - 14 Sep, 2012.
Presentation Type:
Poster
Topic:
Other
Citation:
Helgadottir
L,
Haenicke
J,
Landgraf
T and
Nawrot
MP
(2012). A Robotic Platform for Spiking Neural Control Architectures.
Front. Comput. Neurosci.
Conference Abstract:
Bernstein Conference 2012.
doi: 10.3389/conf.fncom.2012.55.00154
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
11 May 2012;
Published Online:
12 Sep 2012.
*
Correspondence:
Ms. Lovisa Irpa Helgadottir, Bernstein Center for Computational Neuroscience, Berlin, Germany, lovisa.helgadottir@bccn-berlin.de