Abstract
This report presents the design, implementation, and evaluation of an Autonomous Control System (ACS) for a quadcopter drone. The system employs a stationary stereovision camera communicating with a Jetson Nano Development Kit for real-time object detection, identification, and location processing. The generated data is passed to flight control software running in a Software in the Loop (SIL) simulation. The primary objective is to enable a quadcopter to autonomously track and follow a designated object while maintaining a safe distance and position. The report discusses the hardware setup, software architecture, and algorithms utilized for object detection, tracking, and flight control. Additionally, it provides steps for training and validation of the custom You Only Look Once (YOLOv7) model used in the object detection algorithm. The results show the system's performance for both object detection and autopilot tracking.