Robotic Platforms and Experiments

To reach the challenging objectives of the SPARK project the research pathway followed different levels of abstraction: mathematical models, insect behavior study and bio-inspired solutions for robotics. We investigated biological principles as a source of inspiration and we finally applied the mechanisms provided by the complex dynamical system theory to realize mathematical models for cognitive systems. All these aspects deeply investigated in previous Chapters, are here applied to solve specific tasks. A series of wheeled and legged robots are described and applied as test beds for the proposed action-oriented perception algorithms. The cognitive architecture has been experimentally tested at multiple levels of complexity in different robots. Details on the robotic platforms are given together with a description of the experimental results that include multimedia materials collected in the project web page.

[1]  Luigi Fortuna,et al.  Visual homing: Experimental results on an autonomous robot , 2007, 2007 18th European Conference on Circuit Theory and Design.

[2]  Luigi Fortuna,et al.  Sensory Feedback in CNN-Based Central Pattern Generators , 2003, Int. J. Neural Syst..

[3]  Luigi Fortuna,et al.  Reactive navigation through multiscroll systems: from theory to real-time implementation , 2008, Auton. Robots.

[4]  P. Arena,et al.  An autonomous mini-hexapod robot controlled through a CNN-based CPG VLSI chip , 2006, 2006 10th International Workshop on Cellular Neural Networks and Their Applications.

[5]  Volker Dürr,et al.  Insect walking is based on a decentralized architecture revealing a simple and robust controller , 2007, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[6]  Paolo Arena,et al.  Walking capabilities of Gregor controlled through Walknet , 2007, SPIE Microtechnologies.