%0 Journal Article
%T Neural Processing of Auditory Signals and Modular Neural Control for Sound Tropism of Walking Machines
%A Poramate Manoonpong
%A Frank Pasemann
%A Joern Fischer
%A Hubert Roth
%J International Journal of Advanced Robotic Systems
%D 2008
%I InTech
%X The specialized hairs and slit sensillae of spiders (Cupiennius salei) can sense the airflow and auditory signals in a low-frequency range. They provide the sensor information for reactive behavior, like e.g. capturing a prey. In analogy, in this paper a setup is described where two microphones and a neural preprocessing system together with a modular neural controller are used to generate a sound tropism of a four-legged walking machine. The neural preprocessing network is acting as a low-pass filter and it is followed by a network which discerns between signals coming from the left or the right. The parameters of these networks are optimized by an evolutionary algorithm. In addition, a simple modular neural controller then generates the desired different walking patterns such that the machine walks straight, then turns towards a switched-on sound source, and then stops near to it.
%K recurrent neural networks
%K neural control
%K auditory signal processing
%K autonomous robots
%K walking machines
%U http://www.intechopen.com/articles/show/title/neural_processing_of_auditory_signals_and_modular_neural_control_for_sound_tropism_of_walking_machin