People Detection and Localization in Real Time during Navigation of Autonomous Robots

Abstract

Currently the navigation involves the interaction of the robot with its environment, this means that the robot has to find the position of obstacles (natural brands and artificial) with respect to its plane. Its environment is time-variant and computer vision can help it to localization and people detection in real time. This article focuses on the detection and localization of people with respect to plane of the robot during the navigation of autonomous robot, for people detection is used Morphological HOG Face Detection algorithm in real-time, where our goal is to localization people in the plane of the robot, obtaining position information relative to the X-Axis (left, right, obstacle) and with the Y-Axis (near, medium, far) with respect robot, to identify the environment in that it's located in the robot is applied the vanishing point detection. Experiments show that people detection and localization is better in the medium region (201 to 600 cm) obtaining 93.13% of accuracy, this allows the robot has enough time to evade the obstacle during navigation, the navigation getting 97.03% of accuracy for the vanishing point detection. © 2016 IEEE.