Ph.D XXXVIII

Supervisor: Fabio Remondino

Mail: padkan.nazanin@spes.uniud.it

Research Project

Vision and metrology techniques for advanced manufacturing

SLAM (Simultaneous Localization and Mapping) is a technique used in robotics and computer vision to create a map of an unknown environment while simultaneously estimating the position and orientation of a robot or sensor in that environment. SLAM typically relies on multiple sensors, such as cameras, lidars, and sonars, to gather information about the environment. Visual SLAM (VSLAM) is a specific type of SLAM that uses visual sensors, such as cameras, to perform both mapping and localization. VSLAM algorithms extract features from camera images and match them between consecutive frames to estimate the camera’s movement and generate a map of the environment. VSLAM is a real-time process and is typically used in applications such as robotics, augmented and virtual reality, and autonomous driving.

The aim of our research is to integrate artificial intelligence and deep learning with visual SLAM to gain better accuracy in creating a semantic map, with a specific focus on crack detection in visual SLAM applications.