Sistema autônomo de predição de rotas para veículos terrestres

Abstract

The present work has as its main purpose to develop a system that utilizes computer vision to detect obstacles in the environment and plan routes to navigate around them in order to find a target object in the scene. With the increasing growth of vehicle fleets, it becomes essential to develop solutions that facilitate transportation on roads, optimizing travel time and ensuring the safety of drivers and pedestrians. The methodology of the study consists of a series of steps, including depth detection in a scene, analysis of the pixel intensity spectrum, aerial mapping of the environment, and route prediction. The use of computer vision algorithms allows for a more precise analysis of the vehicle’s surroundings, enabling the recognition of obstacles and the selection of the best path to follow. The research contributes to the field of vehicular navigation, as well as to the development of technologies applied to traffic safety.