BIOLOGICALLY INSPIRED NEURO-NAVIGATION SYSTEMS FOR WALKING ROBOTS

BIOLOGICALLY INSPIRED NEURO-NAVIGATION SYSTEMS FOR WALKING ROBOTS

V.E. Pavlovsky
Doctor of Physical and Mathematical Sciences, Keldysh Institute of Applied Mathematics, Professor, Leading Research Scientist, 4, Miusskaya pl., Moscow, 125047, Russia, tel.: +7(916)146-39-87, This email address is being protected from spambots. You need JavaScript enabled to view it.


Abstract
The solution of localization problems for the walking robots in neural network basis is construct-ed. The analysis of applicability of a method and its sensitivity to changes of parameters of a task (characteristics of measurements errors, geometrical parameters) is carried out, the net-works determining navigation parameters for systems with different levels of sensors' errors are built. Tasks for biped and six-legged robots are considered. For the biped robot the model of a vestibular mechanism along with localization model is elaborated.

Key words
Walking robot, biped robot, localization, vestibular mechanism, artificial neural network.

Year 2016 Issue number 4 Consecutive issue number 13 Pages 43-51
UDC identifier: 004.896

 

Editorial office address: 21, Tikhoretsky pr., Saint-Petersburg, Russia, 194064, tel.: +7(812) 552-13-25 e-mail: zheleznyakov@rtc.ru