Internet of Plants: Machine Learning System for Bioimpedance-Based Plant Monitoring.

Area of Science:

• Agricultural Science
• Biotechnology
• Sensor Technology

Background:

• Traditional crop monitoring relies on wide-scale wireless sensor networks (WSNs) for real-time data collection.
• WSNs gather environmental data (soil moisture, temperature, sunlight, nutrients) to optimize cultivation.
• Existing methods require extensive infrastructure and can be costly.

Purpose of the Study:

• To propose a novel "plant-to-machine" interface for direct plant monitoring.
• To reduce reliance on extensive wireless sensor networks (WSNs) in agriculture.
• To simplify crop monitoring infrastructure and decrease digitization costs.

Main Methods:

• Utilizing a plant-based biosensor as the primary data source.
• Employing Electrical Impedance Spectroscopy (EIS) for non-invasive physiological monitoring.
• Developing simple data-gathering hardware and a machine learning framework for data analysis.

Main Results:

• Demonstrated a novel "plant-to-machine" interface for crop monitoring.
• Showcased a non-invasive, single-wire connection for data acquisition.
• Preliminary results confirm the model's feasibility in monitoring plant responses to sunlight.

Conclusions:

• The proposed plant-based biosensor and EIS approach offers a simplified and cost-effective alternative to traditional WSNs.
• This method enables direct monitoring of plant physiological parameters and environmental interactions.
• The technology has the potential to significantly advance precision agriculture and digital farming.