Mobile robots: a study on sensing and perception systems

Abstract

Sensors traditionally used in mobile robotics provide raw data. Most of this data is processed and converted into information. This thesis had to goal study traditional sensing techniques, using RGB and 3D sensors, and propose a new sensing approach embedded in a compact hardware, named the DeepSpatial sensor. To reach the goal, a study of point cloud processing in land robots and air flights was carried out, and later the hardware and software architecture was proposed for a new sensing approach. The thesis was developed in the collection of articles model, consisting of 4 articles published in journals, where each chapter presents the summary of each published paper.