Date of Award

1-3-2020

Access Type

Thesis - Open Access

Degree Name

Master of Science in Unmanned and Autonomous Systems Engineering

Department

Electrical, Computer, Software, and Systems Engineering

Committee Chair

Dr. Hever Moncayo

First Committee Member

Dr. Richard Stansbury

Second Committee Member

Dr. Richard Prazenica

Abstract

In this thesis, a formation flight architecture is described along with the implementation and evaluation of a state-of-the-art vision-based algorithm for solving the problem of estimating and tracking a leader vehicle within a close-formation configuration. A vision-based algorithm that uses Darknet architecture and a formation flight control law to track and follow a leader with desired clearance in forward, lateral directions are developed and implemented. The architecture is run on a flight computer that handles the process in real-time while integrating navigation sensors and a stereo camera. Numerical simulations along with indoor and outdoor actual flight tests demonstrate the capabilities of detection and tracking by providing a low cost, compact size and low weight solution for the problem of estimating the location of other cooperative or non-cooperative flying vehicles within a formation architecture.

Share

COinS