Road Following in an Unstructured Desert Environment Based on the EM(Expectation-Maximization) Algorithm

Author

Lee, Jaesang ; Crane, Carl D., III

Author_Institution

Center for Intelligent Machines & Robotics, Florida Univ., Gainesville, FL

fYear

2006

Firstpage

2969

Lastpage

2974

Abstract

This paper describes the development and performance of a vision system, named PFSS (path finder smart sensor) for autonomous navigation of an unmanned ground vehicle. A monocular camera and vision processing algorithms were used as the sensor system to identify traversable terrain. Unlike the Bayesian based method which was used by Team CIMAR in the 2005 DARPA Grand Challenge, the expectation-maximization (EM) algorithm is applied. The implementation and performance of this approach are reported here

Keywords

cameras; expectation-maximisation algorithm; image segmentation; remotely operated vehicles; robot vision; sensors; Bayesian method; autonomous navigation; expectation-maximization algorithm; monocular camera; path finder smart sensor; road following; traversable terrain identification; unmanned ground vehicle; unstructured desert environment; vision processing algorithms; Cameras; Intelligent robots; Intelligent sensors; Intelligent vehicles; Machine vision; Navigation; Remotely operated vehicles; Road vehicles; Robot sensing systems; Robot vision systems; autonomous vehicle; expectation-maximization (EM) algorithm; image segmentation; navigation; vision system;

fLanguage

English

Publisher

ieee

Conference_Titel

SICE-ICASE, 2006. International Joint Conference

Conference_Location

Busan

Print_ISBN

89-950038-4-7

Electronic_ISBN

89-950038-5-5

Type

conf

DOI

10.1109/SICE.2006.314963

Filename

4108147