Enhanced Footsteps Generation Method for Walking Robots Based on Convolutional Neural Networks

View Sample PDF

Author(s): Sergei Savin (Southwest State University, Russia)and Aleksei Ivakhnenko (Applied Parallel Computing LLC, Russia)
Copyright: 2019
Pages: 24
Source title: Handbook of Research on Deep Learning Innovations and Trends
Source Author(s)/Editor(s): Aboul Ella Hassanien (Cairo University, Egypt), Ashraf Darwish (Helwan University, Egypt)and Chiranji Lal Chowdhary (VIT University, India)
DOI: 10.4018/978-1-5225-7862-8.ch002

Keywords: Artificial Intelligence / Computer Science & IT / Engineering Science Reference / Neural Networks

Purchase

View Enhanced Footsteps Generation Method for Walking Robots Based on Convolutional Neural Networks on the publisher's website for pricing and purchasing information.

Abstract

In this chapter, the problem of finding a suitable foothold for a bipedal walking robot is studied. There are a number of gait generation algorithms that rely on having a set of obstacle-free regions where the robot can step to and there are a number of algorithms for generating these regions. This study breaches the gap between these algorithms, providing a way to quickly check if a given obstacle free region is accessible for foot placement. The proposed approach is based on the use of a classifier, constructed as a convolutional neural network. The study discusses the training dataset generation, including datasets with uncertainty related to the shapes of the obstacle-free regions. Training results for a number of different datasets and different hyperparameter choices are presented and showed robustness of the proposed network design both to different hyperparameter choices as well as to the changes in the training dataset.

The IRMA Community

Research IRM

Enhanced Footsteps Generation Method for Walking Robots Based on Convolutional Neural Networks

Purchase

Abstract

Related Content

IRMA Sponsors