AUTHOR=Guerrero-Higueras Ángel Manuel , Álvarez-Aparicio Claudia , Calvo Olivera María Carmen , Rodríguez-Lera Francisco J. , Fernández-Llamas Camino , Rico Francisco Martín , Matellán Vicente 

TITLE=Tracking People in a Mobile Robot From 2D LIDAR Scans Using Full Convolutional Neural Networks for Security in Cluttered Environments

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 12 - 2018

YEAR=2019

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2018.00085

DOI=10.3389/fnbot.2018.00085

ISSN=1662-5218

ABSTRACT=Tracking people has many applications, such as security or the safe use of robots. Many on-board systems are based on Laser Imaging Detection and Ranging (LIDAR) sensors. Tracking people's legs using only information from a 2D LIDAR scanner in a mobile robot is a challenging problem because many legs can be present in an indoor environment. There are frequent occlusions and self-occlusions, many objects in the environment (table legs, plants, columns, etc.) resemble legs given the limited information provided by a two-dimensional LIDAR, usually mounted at knee height in mobile robots, etc. On the other hand, LIDAR sensors are affordable in terms of acquisition price and processing requirements. In this article, we describe a tool named PeTra, based on an off-line trained, full Convolutional Neural Network capable of tracking pairs of legs in a cluttered environment. We describe the characteristics of the system proposed and evaluate its accuracy using a dataset from a public repository.