Architecture, inertial navigation, and payload designs for low-cost Unmanned Aerial Vehicle-based Personal Remote Sensing.

摘要

This thesis presents work done towards a Personal Remote Sensing (PRS) system: small Unmanned Aerial Vehicles (UAVs) with electronic, control, and sensing subsystems. Based on papers presented to conferences (AutoTestCon2008 and MESA2009), as well as other work on PRS, multiple levels of engineering are detailed: complex multi-UAV data flow; attitude estimation filters; real-time microprocessor functionality; and small, mobile power systems. Wherever possible, Open-Source tools and designs have been used, modified, or studied, providing excellent cost to performance ratios in most cases. First, the overall PRS UAV architecture, AggieAir, is presented with a motivating examples (Ghost-Eye and EagleEye camera payloads). Then, AggieNav, an inertial navigation system for small UAVs, is detailed, along with information about a Kalman filter for estimation of UAV navigation, position, and attitude. Finally the Spatial Environment Autonomous Logger (SEAL), a general-purpose wireless datalogger for small UAV applications, is presented, with application examples with and without small UAVs. This work represents designs based on two years of organic small UAV system growth, and provides integrated solutions to many problems of small UAV communication, sensing, and control.