The overall aim of this graduation project is to present a compact robot that uses open-source probabilistic algorithms and mapping to process information obtained from a closed environment through the fusion of two sensors: An Intel Realsense T265 tracking camera capable of performing visual odometry and a Hokuyo URG-04-LX-UG01 2D LiDAR capable of identifying the distance of objects with high precision. The specific objectives are: To study the SLAM problem and the methodologies considered state of the art; To design and build a compact mobile robot with differential drive; To integrate the robot s mechanical, electronic and software; To combine the sensory information from the camera s visual inertial odometry and the distances obtained by the LiDAR to perform simultaneous localization and mapping in indoor environments; To demonstrate that the compact robot developed is capable of generating a occupancy grid map of an environment while estimating its own location, solving the SLAM problem in real time and to validate the results by comparing the estimated trajectory by the robot with a developed ground truth. The ROS framework was used to help distribute tasks between hardware and software. The robot was entirely 3D printed following the open-source project developed by Raffaelo Bonghi. The system was tested with indoor experiments, demonstrating the robot’s ability to perform loop closures, recognize the edges of a node and create a occupancy grid map in real time. The robot traveled through a closed environment, with its movement controlled by teleoperation, obtaining a 2D map of the environment, as well as its complete estimated trajectory. Despite being a small robot with hardware limitations, the system proved capable of solving the SLAM problem in real time and performing the task continuously for long periods of time. The quality of the mapping in indoor and static environments proved to match the topology of the environment.