Techno-economic analysis of implementing unmanned aerial systems for corn crop health monitoring

Ashish Manandhar
Category: 
Postdoctoral
Advisor: 
Ajay Shah
Department: 
Food, Agricultural and Biological Engineering
Abstract: 

Crop health needs to be monitored regularly to detect crop stresses in time and implement proper crop management measures to maximize crop yield. Conventional crop health monitoring by crop scouts is time and labor-intensive with limited observations, which oftentimes, results in excess or uniform application of resources, such as fertilizer, regardless of within-field variability. This can increase crop production costs, as well as result in negative environmental impacts. Sensors onboard unmanned aerial systems (UAS), some further augmented with machine learning tools (ML), can be used to monitor crop health. High-resolution data collected using UAS can be used to optimize the application of resources by focusing only on certain areas when necessary, thus, minimizing the resources requirements, costs, and environmental impacts of crop production. Thus, the objective of this study was to evaluate the techno-economics of crop health monitoring in corn production using four crop scouting approaches: 1) no scouting, 2) traditional crop scouting, 3) UAS-based comprehensive scouting, and 4) UAS-ML-based scouting. A techno-economic analysis model was developed for a corn farm to estimate requirements for resources such as UAS, energy use, labor and costs for different crop scouting scenarios, and corn production costs under these scenarios. The data were obtained from literature, market study, and drone surveys. UAS-ML approach had the lowest labor, battery swaps, and energy use due to flight path and time reductions. UAS- and UAS-ML-based scouting costs were 9 and 26% lower than conventional crop scouting costs. UAS- and UAS-ML-based crop scouting enabled substantial reductions in fertilizer use, which reduced production costs and increased the net profit for corn production. Annual cost savings of ~$43-78 per acre can be obtained using UAS- and ML-based crop scouting. In perspective, the total annual savings in corn production costs by implementing UAS- and ML-based crop scouting in 20% of 82.5 million acres corn production in the U.S. could be $0.7-1.2 billion. Similar approach could be implemented for weed and pest control which can further improve crop health and minimize crop production costs and environmental impacts.