Σχεδιασμός και υλοποίηση αυτοκατευθυνόμενου οχήματος με ανίχνευση λωρίδας
Design and implementation of self-navigated vehicle with lane detection
KeywordsΑυτοκατευθυνόμενο όχημα ; Ανίχνευση λωρίδας ; Υπολογιστική όραση ; Αλγόριθμος ανίχνευσης λωρίδας
The purpose of this thesis is to study the construction of a self-navigated vehicle which detects and follows a lane. Τhis is an attempt to couple hardware and software in an appropriate way so that the resulting model vehicle has the ability to steer autonomously through lane detection. The core hardware consists of camera paired with a single board computer used for lane tracking, while both placed on a custom 3D–printed deck. Additionally, two servo motors are responsible for moving the vehicle. In terms of software, the vehicle is based on algorithms for locating the lane and applying the necessary corrections during movement so that the vehicle stays in course. These algorithms are implemented in the Python programming language. In addition, computer vision techniques are applied to locate the lane. The thesis is divided into six thematic sections: Τhe first part where computer vision is presented from a theoretical aspect and in combination with self-navigated vehicles, as well as other applications which use computer vision. The second section describes the hardware that we used for the construction of the vehicle. The third section of the thesis mentions the suitable software and libraries of the operations of the specific vehicle. The fourth section presents in detail all the connections between the parts of the vehicle. In the fifth section, the algorithms for lane detection are presented. Lastly, the sixth section gathers the conclusions of the project as well as the difficulties that arose during the construction and opportunities for future improvements.