Study of simulation and modelling tools for autonomous vehicle driving based on stereoscopy

Abstract

The subject of this thesis is the analysis and study of the tools available today that assist engineers in the modeling and development of autonomous vehicles with automatic navigation. A comprehensive explanation of the basic concepts will be made to familiarize the reader with the specific technology. The individual systems of autonomous vehicles will be analyzed from the hardware and software perspective, and the tools that can be used to simulate autonomous vehicle operations and their underlying algorithms will be explored. In addition, the application of stereoscopic vision in these functions will be analyzed. Then a summary review of the utilization of artificial intelligence in combination with current technology review of stereoscopic vision using cameras will be presented along with references to literature. Finally, a review of a specific stereoscopic camera will take place (Intel RealSense D435), and a simple example of integrating the camera with the MATLAB platform will be given. Following that, MATLAB will be used to calibrate the camera. In addition some measurements will be performed to test the accuracy of the specific camera.