The design and the development of a hyperspectral unmanned aircraft mapping system for the detection of invasive plants

Cost-effective tools for rapidly identifyingthe spatial distribution of invasive plants in naturaland managed landscapes are needed by land managersacross the world to identify and monitor the spread ofweeds. This paper describes a low-cost unmanned aircraftsystem (UAS) designed for the rapid acquisitionof hyperspectral image data. The system integrates aHeadwall Hyperspectral Visible Near-Infrared (VNIR)sensor covering 400'1000 nm range, a PCO EdgeCMOS camera, a Cyber Technology autopilot andInertial Navigation System (INS) including Real TimeKinematic (RTK) GPS, a data acquisition computerand custom software for operating the sensor. Theimaging and autopilot systems are mounted on lowcostfixed wing and helicopter platforms. The Zealous2 helicopter kit was developed by RC Helicopters andbuilt by Cyber Technology in Western Australia. TheZealous 2 consists of a jet turbine engine poweringmain rotors spanning 2.1 m and is capable of lifting 15kg. The fixed wing plane known as a 'Super Hauler'is made from balsa wood, has a wing span of 3.6 mand is 3 m long. The plane was designed and builtby BTE Engineering in the USA and is capable oflifting 20 kg. The GPS/INS system provides aircraftposition, pitch, roll and yaw information allowing forrapid geo-registration of the imagery. This paper willdiscuss the challenges associated with developmentand integration of the components for mapping invasiveplants using unmanned aircrafts and the spectralcharacteristics of a number of target invasive species.