Breaking the field phenotyping bottleneck in maize with autonomous robots.

Area of Science:

• Agricultural Science
• Plant Biology
• Robotics

Background:

• Phenotypic plasticity in maize is crucial for crop improvement but is hindered by the high cost and time required to measure genotype-by-environment-by-management (GxExM) interactions.
• Traditional phenotyping methods are a bottleneck for advancing maize yield due to their labor-intensive nature and scale limitations.

Purpose of the Study:

• To demonstrate the feasibility of using autonomous ground robots for high-throughput, low-cost, and large-scale phenotyping in maize.
• To assess the capability of robotic platforms in capturing biologically relevant data for dissecting GxExM interactions.

Main Methods:

• Deployed TerraSentia autonomous ground robots equipped with low-cost sensors across 142 research fields in the USA and Canada over five years.
• Collected in-canopy data from nearly 200,000 experimental units.
• Utilized computer vision and machine learning algorithms to analyze multi-sensor data and derive phenotypes like plant height, ear height, stem diameter, and leaf area index.

Main Results:

• Autonomous robots accurately and reliably measured key plant phenotypes at high volume and scale.
• The collected data enabled the dissection of interactions between maize genotypes and nitrogen rates across diverse environments.
• Demonstrated the potential for robots to significantly reduce human labor in plant phenotyping.

Conclusions:

• Autonomous robotic platforms are a viable solution for high-throughput, low-cost maize phenotyping.
• This technology can accelerate the understanding of GxExM interactions, paving the way for improved crop yields.
• Robotic phenotyping promises to revolutionize agricultural research by overcoming current data collection limitations.