Integration of potential field theory and proportional navigation theory to autonomously guide an unmanned aerial vehicle.

Abstract

Industrial robotics, military, surveying, and delivery applications have laid a foundation for research into full autonomy of machines, including Unmanned Aerial Vehicles (UAV). This thesis supports this research by surveying the methods used to guide UAVs, and developing a new method by combining potential fields, typically used for obstacle avoidance, and proportional navigation, a popular missile guidance algorithm. The new algorithm modifies the old algorithms to allow a UAV to track an optimal path to, and rendezvous with, a moving target while avoiding obstacles in its path. A model for a quad rotor style UAV is developed and controlled using feedback linearization. A simulator is built for deploying a number of environments and taking performance measurements. Aspects of the hardware implementation are introduced.