Autonomous Fixed-Point Landing for Quadrotor Aerial Vehicles

Using a student-built quadcopter, the navigation capabilities of autonomous aerial vehicles were explored. The project implemented an artificially intelligent quadcopter that could autonomously identify, navigate to, and eventually land at a marked location. Through the use of an onboard camera and Raspberry Pi control board, image processing techniques were developed in Python in order to analyze and understand the quadcopter’s environment. By binarizing and contouring the image feed from the camera, a red landing target was identified. Using this environmental model, control algorithms were developed in order to intelligently navigate to and land on the target autonomously