Abstract
Simultaneous localisation and mapping (SLAM) is a process by which robots build maps of their environment and simultaneously determine their location and orientation in the environment. In recent years, SLAM research has advanced quickly. Researchers are currently working on developing reliable and accurate visual SLAM algorithms dealing with dynamic environments. The steps involved in developing a SLAM system are described in this article. We explore the most-recent methods used in SLAM systems, including probabilistic methods, visual methods, and deep learning (DL) methods. We also discuss the fundamental techniques utilised in SLAM fields.