Semantic Simultaneous Localization based on deep-learning algorithm, Mapping and Tracking

During the last few years, the area of greatest development has been that of Artificial Intelligence(AI), given the fact that it is applicable in the vast majority of situations that also involve those of daily life. The field in which it is applied by us is the robotic one where, a fundamental point, is the perception of the environment by the robot. The problem that has been analyzed is the simultaneous localization and mapping (SLAM) which has been addressed through the integration of deep-learning algorithms such as object detection. In this thesis is analyzed and compared the state of the art of those algorithms and is given a solution as robust as possible to the above stated problem.