Plataforma robótica móvil para experimentos de mapeo y localización simultanea (SLAM) en base a sensores de rango.

Abstract

The objective of this research was to design a mobile robotic platform that allows students of the Mechatronics career at Universidad Tecnica del Norte (UTN), to perform Mapping and Simultaneous Location (SLAM) experiments in two dimensions with range sensors. For this, the approaches and techniques used for this purpose were researched, highlighting probabilistic approach with estimation techniques and Bayesian filters. They were researched the hardware elements that allow to develop such performance. At this point, the Sensor of Light Detection and Range (LIDAR) stand out strongly and in conjunction with a mini computer called Raspberry Pi 3 as robot brain and a commercial robotic platform modified as a base structure, it was implemented the Raspi_mBot. In which, computational algorithms were applied in the Robotic Operation System (ROS) environment. It is concluded that with the application of the following packages: ROS, hector_slam, gmapping and test protocol, the Mapping and the Simultaneous Localization were successfully developed in controlled environments similar to those existing in the laboratories at Universidad Tecnica del Norte (UTN). It is recommended to include other sensors in combination with navigation techniques and trajectory planning to provide the Raspi_mBot with complete autonomy.