Development and testing of an unmanned aircraft system for environmental science

For some environmental science applications, Unmanned Aircraft Systems (UASs) are increasingly recognized for their capacity to collect remotely sensed data in a safer, more efficient and effective manner than is permitted with manned aircraft and satellite remote sensing platforms. To date, however, technological, human, and other challenges have constrained adoption of UASs in the environmental sciences. This study developed and tested a new UAS for an archetypical environmental science research group (stakeholder) composed of non-UAS experts. Specifically, this thesis: 1) Assessed the research and operational needs of the stakeholder to determine the optimum UAS platform; 2) Developed an Unmanned Aerial Vehicle (UAV) and sensor payload; 3) Developed a new software tool for UAV operation planning, control, and optimized data acquisition; and 4) Tested the operational performance of the newly developed UAS. A powered paraglider was selected as the optimal UAV platform for the stakeholder. Development and testing of the UAS overcame many technical, human, and other challenges. A relatively stable and useful UAS with a 30lb payload was engineered and appears to meet the needs of the stakeholder. A new scalable operational software tool was engineered that integrates operation planning, UAV and sensor specification, and map based real time flight optimization with a relational database. An assessment of the Unmanned Aircraft System was performed by asking four different pilots, ranging in pilot skills from skilled to novice, to perform two flights (one unassisted and one assisted with the newly designed operational software) to best capture photographic coverage of an experimental area (200 x 400 meters) within a fifteen minute time limit). The capacity of the operational software to improve the spatial coverage of data acquisition was also assessed. Coverage of aerial photography was enhanced 4.41% to 35.76% when the operational software was used when compared to non software assisted flights, and the Unmanned Aerial Vehicle spent between 3.75% and 12.29% more time in the predefined sampling area when guided by the operational software. An experienced pilot outperformed inexperienced pilots but still benefitted from the guidance offered by the operational software