Research of Global Localization for Humanoid Robot Based on Vision

DOI

10.2991/csss-14.2014.7How to use a DOI?

Keywords

vision; evolutionary computation; humanoid robot; global localization

Abstract

Vision perception plays a key role in the research on humanoid robot. A new version of particle filters called coevolution based adaptive particle filters (CEAPF) is proposed for robot localization. Using vision and odometer, a robust perception model extracting from environmental features, which are unscented by Kalman filter, is established by effective fixed scale feature-transformation method. Dimensional feature points matching algorithms are used to match the feature points based on KD-Tree to merge a species of cooperative coevolution competition mechanisms into particle filters. Coevolution adaptive particle filters are proposed to track the different assumptions and the samples in inter-species evolution can be moved towards the larger desired regions by using the crossover and mutation operators in evolutionary computation. Finally, results of success and precision localization are shown by experiment.

Copyright

© 2014, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Qiubo Zhong
AU  - Chunya Tong
AU  - Yu Wang
PY  - 2014/06
DA  - 2014/06
TI  - Research of Global Localization for Humanoid Robot Based on Vision
BT  - Proceedings of the 3rd International Conference on Computer Science and Service System
PB  - Atlantis Press
SP  - 25
EP  - 28
SN  - 1951-6851
UR  - https://doi.org/10.2991/csss-14.2014.7
DO  - 10.2991/csss-14.2014.7
ID  - Zhong2014/06
ER  -