This page describes my previous research in control systems design and robotics. Specifically, my undergraduate thesis dealt with optimizing a UAV swarm control algorithm for convoy escorting while my Master’s thesis research focused on enabling a small UAV to autonomously track a moving ground target. All processing was computed on-board the robot and the localization was done with and without external localization systems.
Additionally, I performed proof of concept experiments with the 3DRobotics quadrotor and ArduCopter software. This relatively cheap hobby line quadrotor is easily assembled and runs off open-source software with strong online community support. The ArduCopter and ROS Integration pages will be updated with data and videos from the latest experiments.
Further, I have been investigating the modeling and control of autonomous vehicles. This platform has different dynamics than a quadrotor, making it less maneuverable. However, these robots typically have higher payload capacities, allowing for more sensors and onboard computing power. This allows them to leverage more advanced algorithms such as machine learning for computer vision, which I am also investigating. My research on this topic will be updated on the Autonomous Vehicles page.
W. Selby, “Here are the disturbing reasons ISIS marketing is so effective,” Yahoo! Finance, 27 Aug, 2106. link
W. Selby, “Operating in an Era of Persistent Unmanned Aerial Surveillance,” Small Wars Journal, 31 Jul, 2015. link
W. Selby, “Print, Plug, and Play Robotics,” Center for International Maritime Security, CIMSEC.org, 10 Jun, 2014. link
W. Selby, “Unprepared for the Unmanned: Addressing the enemy unmanned aircraft threat,” Marine Corps Gazette, 01 November 2013: 45-49. pdf
W. Selby, “Autonomous Navigation and Tracking of Dynamic Surface Targets On-board a Computationally Impoverished Aerial Vehicle,” Master’s thesis, Massachusetts Institute of Technology, 2011. pdf bibtex