Sensing Architecture for Terrestrial Crop Monitoring: Harvesting Data as an Asset

Area of Science:

• Agricultural Engineering
• Robotics
• Environmental Science

Background:

• Sustainable food production and environmental issue mitigation are hampered by decisions made with unreliable or insufficient data.
• Data-driven agriculture aims to address information gaps in critical farming decisions, but manual measurements lead to undersampling, limiting AI applications.
• Current crop monitoring lacks standardized guidelines for vehicle configuration to achieve high-resolution parameter tracking.

Purpose of the Study:

• To propose a sensing architecture for automating crop scouting from ground vehicles, combining crop proximity with massive data sampling.
• To provide guidelines for configuring monitoring vehicles for optimal high-resolution crop parameter tracking.
• To demonstrate the architecture's capability in generating real-time crop maps for data-driven agricultural actuation.

Main Methods:

• Structuring a sensing architecture into four subsystems with examination of common components and their interactions.
• Developing an autonomous robot embodying the proposed architecture for automated vineyard harvesting zone sorting.
• Generating high-density field data maps integrating global positioning, agronomical traits, and ambient conditions.

Main Results:

• The proposed architecture facilitates real-time generation of comprehensive crop maps.
• An autonomous robot successfully sorted vineyard harvesting zones based on generated data, producing wines of distinct characteristics.
• The system demonstrated suitability for massive monitoring and subsequent data-driven actuation in agriculture.

Conclusions:

• An efficient and reliable sensing architecture is essential for leveraging novel sensors and overcoming limitations in non-invasive crop parameter measurement.
• The automated crop scouting system enables high-density data acquisition, supporting statistically significant analysis and AI implementation.
• This approach supports systematic differential harvesting and data-driven actuation, enhancing sustainable agriculture practices.