Direct Control of an Active Tactile Sensor using Echo State Networks

Tactile sensors (antennae) play an important role in the animal kingdom. They are also very useful as sensors in robotic scenarios, where vision systems may fail. Active tactile movements increase the sampling performance. Here we directly control movements of the antenna of a simulated hexapod using an echo state network (ESN). ESNs can store multiple motor patterns as attractors in a single network and generate novel patterns by combining and blending already learned patterns using bifurcation inputs. Index Terms – Active Tactile Sensors, Motor Learning, Neuronal Networks André Frank Krause e-mail: [email protected] · Bettina Bläsing e-mail: [email protected] · Thomas Schack e-mail: thomas. [email protected] Neurocognition and Action Biomechanics Research Group, Bielefeld University, Germany Volker Dürr e-mail: [email protected] Dept. for Biological Cybernetics, University of Bielefeld, Faculty of Biology, PO Box 100131, 33501 Bielefeld, Germany all authors: Cognitive Interaction Technology, Center of Excellence, University of Bielefeld, 33615 Bielefeld, Germany