Intra-Canopy Sensing Using Multi-Rotor sUAS: A New Approach for Crop Stress Detection and Diagnosis

Abstract

HighlightsA novel platform was developed for intra-canopy insertion of sensors from a multi-rotor sUAS.The system enables real-time data acquisition from inside the crop canopy comparable to an in-person view.The system provides ideal data for use in modern CNN-based stress diagnosis in row crop production.Abstract. Remote sensing is a critical tool in precision agriculture, giving producers the ability to monitor field conditions throughout the growing season. Although several remote sensing platforms are in use today, small unmanned aerial systems (sUAS) provide the greatest flexibility with the highest resolution. As sUAS capabilities continue to increase (i.e., payload, flight time, and speed), their potential in commercial row crop production is substantial. However, like other forms of remote sensing, traditional sUAS are limited to a nadir view of the target and only capture the top of the crop canopy. Although disease epidemiology and stress origins vary, this limited view usually does not capture the impact of stress at the initial manifestation. For example, stresses such as macronutrient deficiencies in corn originate at the base of the plant and then move upward as nutrients translocate. By the time the stress is detectable at the top of the canopy, the opportunity to mitigate yield loss is limited. A new sUAS platform is needed for sensing beneath the upper portion of the canopy. The Stinger platform, developed to meet this need, consists of a 4.0 m fiberglass rod, custom sensor mount, communication network, and radio link. Using this platform, a variety of sensors can be inserted into the crop canopy from a hovering sUAS. The Stinger platform, when combined with artificial intelligence (AI), significantly expands the capabilities of sUAS for diagnosis of crop stress in row crop production. Keywords: Intra-canopy sensing, Remote sensing, RGB imagery, Stress diagnosis, sUAS.