Date of Award
Master of Science
Department of Systems Engineering and Management
John M. Colombi, PhD.
This research addresses the flight path optimality of Small Unmanned Aerial Systems (SUAS) conducting overwatch missions for convoys or other moving ground targets. Optimal path planning algorithms have been proposed, but are computationally excessive for real-time execution. Using the Arduino-based ArduPilot Mega Unmanned Aerial Vehicle (UAV) autopilot system, Hardware-in-the-Loop (HIL) analysis is conducted on default mobile target tracking methods. Designed experimentation is used to determine autopilot settings that improve performance with respect to path optimality. Optimality is characterized using a weighted combination of stand-off range and aircraft roll-rate. Finally, a state-based heuristic navigation strategy is designed, developed, and tested that approximates optimal path solutions and can be used for real-time execution. A 66% improvement in mean performance is achieved over default target tracking methods. Finite state machine improvements are found to be statistically significant and it is concluded that heuristic strategies can be a viable approach to realizing near-optimal SUAS flight paths utilizing onboard processing capabilities.
DTIC Accession Number
Neal, Charles J., "Feasibility of Onboard Processing of Heuristic Path Planning and Navigation Algorithms within SUAS Autopilot Computational Constraints" (2014). Theses and Dissertations. 717.