Field Phenotyping Robots for Accurate Crop Trait Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for measuring agricultural traits like corn ear height and soybean pod count in dynamic environments are inefficient and unreliable, relying heavily on human labor and introducing selection bias, leading to inaccurate data.
Innovation Solution
A robotic system equipped with sensors and a computing device that navigates and analyzes image data to detect and measure agricultural objects of interest, determining characteristics such as height and geometry without human intervention, using LIDAR, cameras, and deep learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional manual sampling methods are used to measure agricultural traits, then human labor can obtain measurements, but the measurements are inefficient and unreliable with selection bias
Solution Approach 1:
The patent replaces manual mechanical measurement systems with an automated robotic system that uses sensors (cameras, LIDAR), GPS positioning, and computer vision algorithms to detect and measure agricultural objects. This substitution eliminates human labor while improving both measurement accuracy through consistent algorithmic processing and productivity through continuous automated operation across entire fields.
Solution Approach 2:
The robotic system performs self-navigation using GPS and onboard sensors to autonomously traverse the field, automatically detects agricultural objects using computer vision, and self-corrects positioning errors through integration of multiple sensor data streams. This self-service capability eliminates the need for human operators while maintaining high measurement standards.
2Productivity
If random sampling of a few plants is used to estimate average traits, then data collection is quick, but selection bias skews statistics and reduces reliability
Solution Approach 1:
The robotic system is designed to measure all agricultural objects in the entire field rather than sampling subsets, making the measurement process universal across the complete population. This multi-functional approach simultaneously achieves comprehensive coverage for reliability and efficient automated processing for productivity, eliminating the trade-off between sampling speed and statistical accuracy.
3Measurement precision
If manual measurement devices like tape measures and poles are used, then measurements can be obtained, but human labor requirements make the process inefficient and costly
Solution Approach 1:
The patent replaces simple manual mechanical devices (tape measures, poles) with a complex automated robotic system equipped with sensors, processors, and navigation systems. This increase in device complexity enables full automation, eliminating human labor while maintaining or improving measurement accuracy through consistent algorithmic processing and multi-sensor validation.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides accurate, efficient, and reliable measurements of agricultural traits, enabling fully automated data collection across entire fields without human intervention, reducing errors and biases, and allowing for various characteristics to be analyzed and used for actionable decisions.
Implementation Method 1
image data of an environment captured by a sensor device
Implementation Method 2
using LIDAR, cameras, and deep learning algorithms
Data Source
AI summary
A method, non-transitory computer readable medium, and system that manage agricultural analysis in dynamic environments includes detecting a location of one or more agricultural objects of interest in image data of an environment captured by a sensor device during active navigation of the environment. An orientation and position of the sensor device with respect to the image data is determined. Each of the one or more agricultural objects of interest is analyzed based on the image data, the detected location of the one or more agricultural objects of interest, and the determined orientation and position of the sensor device to determine one or more characteristics about the one or more agricultural objects of interest. At least one action is initiated based on the determined one or more characteristics about the one or more agricultural objects of interest.