UAV path automation using visual waypoints acquired from the ground

摘要

This article models and implements a simple system to automate the flight path of an UAV (Unmanned Aerial Vehicle) using ground image recognition. Its operation is based on two phases. In the first phase, we performed a manual flight, via joystick, to capture the visual waypoints of the trajectory that we want to automate. In the second phase, the system performs autonomous flights following the visual waypoints captured to retrace the original trajectory. To test the developed system, we performed two sets of experiments. The first set consists of simulations of the software developed using the Unity platform. The second set consists of indoor real flights using a commercial UAV robotic platform.