A Comprehensive Overview of AI and Visual Data Applications in Mobile Robotics

Abstract

Mobile robots are increasingly being suggested as solutions to various problems on their own, thanks to the significant progress in the field of mobile robotics in recent years. In recent years, advancements in computer vision hardware and algorithms have made visual sensors an integral part of mobile robots operations. It is worth noting that mobile robots has made great strides in using AI technologies to address a wide range of issues, with photos serving as the primary or secondary source of data, or in conjunction with other sensors like lasers or GPS. The scientific community is closely watching efforts to enhance the autonomy of mobile robots. There has been a wide range of methods developed as a result of the large number of planned works in recent years. Any mobile robot worth its salt needs to be able to map its surroundings, pinpoint its exact location inside the model (localization), and navigate itself from one spot to another. Keeping this in mind, the purpose of this review is to examine these issues, the ways in which academics have used AI tools and visual information to tackle them, and the ways in which these techniques have developed over the past decade. There are a lot of studies in the connected literature, and the topic is open at the moment. Consequently, it could be worthwhile to assess where things stand with the subject at the moment. Information retrieval from scenes, mapping, localization, and exploration are only a few of the mobile robotics jobs that have benefited greatly from AI, as this review has shown. Still, it's worthwhile to keep digging into this area of study in the hopes of uncovering more comprehensive answers to the navigation challenge, allowing mobile robots more independence in complicated, diverse, and expansive settings.