In the RELIEF project we will utilize Unmanned Ground Vehicles (UGVs), which will perform full coverage of an a-priori known environment and at the same time RFID tags localization. Specifically, two vehicles types will be employed, a lighweight with small dimensions, suitable to cover narrow spaces and a larger one, perfect for convering large industrial environment, such as warehouses. ach of the robots will take as input a 2D or 3D representation of the environment it operates in, and it must accurately calculate its pose in this representation. The most usual 2D representation is an OGM (Occupancy Grid Map), each cell of which holds the probability of the corresponding part of the environment to be occupied, whilst the most common 3D representation is Octomap. The most used robot localization algorithm is called AMCL (Adaptive Monte Carlo Localization) and is based on particle filters to continuously improve its assumption about the robot's pose while it moves.

Finally, in order for pose localization and safe robot motion to be supported, we utilize two types of sensors. The first one is called Lidar and is capable of accurately measuring a large number of distances perimetrically of the robot's body, and the second is a depth camera, which generates depth image, where each pixel represents the distance of the depicted point from the camera.