Adaptive Sensor-Driven Neural Control for Learning in Walking Machines

Authors:
Poramate Manoonpong;Florentin Wörgötter
Affiliations:
Bernstein Center for Computational Neuroscience (BCCN), University of Göttingen, Göttingen, Germany D-37073;Bernstein Center for Computational Neuroscience (BCCN), University of Göttingen, Göttingen, Germany D-37073
Venue:
ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part II
Year:
2009

Citing 6
Cited 1

Strongly improved stability and faster convergence of temporal sequence learning by using input correlations only

Neural Computation
Strongly improved stability and faster convergence of temporal sequence learning by using input correlations only

Neural Computation
Neural Preprocessing and Control of Reactive Walking Machines: Towards Versatile Artificial Perception-Action Systems (Cognitive Technologies)

Neural Preprocessing and Control of Reactive Walking Machines: Towards Versatile Artificial Perception-Action Systems (Cognitive Technologies)
Sensor-driven neural control for omnidirectional locomotion and versatile reactive behaviors of walking machines

Robotics and Autonomous Systems
2008 Special Issue: Central pattern generators for locomotion control in animals and robots: A review

Neural Networks
SO(2)-networks as neural oscillators

IWANN'03 Proceedings of the Artificial and natural neural networks 7th international conference on Computational methods in neural modeling - Volume 1

The development of a biomechanical leg system and its neural control

ROBIO'09 Proceedings of the 2009 international conference on Robotics and biomimetics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Wild rodents learn the danger-predicting meaning of predator bird calls through the paring of cues which are an aversive stimulus (immediate danger signal or unconditioned stimulus, US) and the acoustic stimulus (predator signal or conditioned stimulus, CS). This learning is a form of pavlovian conditioning. In analogy, in this article a setup is described where adaptive sensor-driven neural control is used to simulate biologically-inspired acoustic predator-recognition learning for a safe escape on a six-legged walking machine. As a result, the controller allows the walking machine to learn the association of a predictive acoustic signal (predator signal, CS) and a reflex infrared signal (immediate danger signal, US). Such that after learning the machine performs fast walking behavior when "hearing" an approaching predator from behind leading to safely escape from the attack.