UAS-Based Phenotyping Platform for Identifying Abiotic Stress Tolerant Crop Genotypes
A novel phenotyping and modeling platform using unmanned aircraft systems (UAS) with multispectral and thermal imaging, combined with machine learning, to accelerate identification of drought and heat stress tolerant crop genotypes. Designed for use in replicated field trials of small grains and broadleaf crops, correlating aerial imagery with yield to pinpoint tolerant varieties.
Overview
This solution offers a phenotyping and modeling platform designed to accelerate the identification of crop genotypes tolerant to abiotic stresses, specifically drought and high temperature. By deploying unmanned aircraft systems (UAS) equipped with visual, multispectral, and thermal cameras over replicated no-till semiarid field trials, the platform captures high-resolution aerial imagery across more than 30 genotypes of small grains and broadleaf crops. Machine learning algorithms process this imagery to estimate crop growth, physiology, and phenology, which are then correlated with yield data to identify stress-tolerant genotypes. The approach reduces the time and labor required by traditional manual phenotyping methods and enables more precise selection of resilient crop varieties for breeding programs and agricultural deployment.
Technical specifications
Core capabilities:
- UAS-based aerial imaging using visual, multispectral, and thermal cameras to capture crop canopy data at field scale
- Machine learning integration for automated extraction of crop physiology metrics, including NDVI (Normalized Difference Vegetation Index), NDRE (Normalized Difference Red Edge), and canopy temperature
- Phenology tracking to identify key growth stages such as heading, flowering, and maturity dates
- Yield correlation analysis linking aerial phenotyping data to ground-truth yield measurements to identify tolerant genotypes
- Replicated no-till semiarid field trial design enabling robust evaluation under realistic stress conditions
- Broad crop applicability across small grains and broadleaf field crops
Technology readiness level
The platform is currently at an early-to-mid stage of development. The underlying hypothesis—that genotypic variation for drought and high temperature tolerance exists in field crops and can be detected through physiological responses captured via UAS imaging—has been established based on evolutionary and physiological rationale. Future validation involves deploying the UAS system over replicated field trials containing more than 30 genotypes, with machine learning models to be refined in collaboration with industry scientists. The approach is positioned for validation in real-world field environments, with the next steps focused on demonstrating correlation between UAS-derived metrics and yield outcomes across diverse crop genotypes.
About North Dakota State University (NDSU)
North Dakota State University is a comprehensive public land‑grant research university (R1) in Fargo with a pragmatic, industry‑engaged culture. Companies engage through an adjacent research and technology park offering incubator space and prototyping resources, and shared high‑performance computing supports data‑intensive work. Pilot‑scale processing facilities and co‑located labs in the Peltier Complex and the campus pilot plant help de‑risk scale‑up and validate manufacturing processes. A statewide Extension network provides access to field sites, stakeholders, and dissemination channels for real‑world testing and deployment. Research is backed by competitive federal funding from agencies such as the National Science Foundation and the U.S. Department of Agriculture, and a dedicated technology transfer office streamlines IP, licensing, and startup formation.
delivered to your inbox.