Title :
A Visual-Perception Layer Applied to Reactive Navigation
Author :
Neto, A. Miranda ; Victorino, A. Corrêa ; Ferreira, J.V.
Author_Institution :
Heudiasyc Lab., UTC, France
Abstract :
Environment perception is a major research issue which is very important in the field of robotic system. In order to identify the horizon line and the drivable region, we have proposed a visual-perception system based on an automatic image discarding method as a simple solution to improve the system performance. In this paper, all these previous methods are organized in a visual-perception layer which also includes a method for estimating the risk-of-collision based on Pearson´s Correlation Coefficient and an evolution of the Threshold and Horizon Finder (TH Finder). These methods were successfully evaluated from real data.
Keywords :
collision avoidance; mobile robots; robot vision; statistical analysis; visual perception; Pearson´s correlation coefficient; TH finder; automatic image discarding method; drivable region identification; environment perception; horizon line identification; reactive navigation; risk-of-collision estimation; robotic system; system performance improvement; threshold-and-horizon finder; visual-perception layer; visual-perception system; Equations; Image segmentation; Navigation; Roads; Robots; Sensors; Vehicles; Autonomous Vehicles; Computer Vision; Mobile Robots; Perception;
Conference_Titel :
Robotics Symposium and Latin American Robotics Symposium (SBR-LARS), 2012 Brazilian
Conference_Location :
Fortaleza
Print_ISBN :
978-1-4673-4650-4
DOI :
10.1109/SBR-LARS.2012.8
