PlantFeel: A Real-Time System to Translate Plant Bioelectrical Signals Into Visual Emotional Feedback

Overview

Current plant monitoring systems often rely on reactive visual inspections or invasive analysis, detecting issues only after physiological damage has occurred. To address this, we present a non-invasive, real-time framework to detect early physiological stress in six Epipremnum aureum specimens, as bio-potential fluctuations provide unique waveform signatures long before physical symptoms emerge. To capture this data, we engineered an integrated system using an Arduino-based unit and an AD8232 analog front-end for high-sensitivity acquisition. After applying digital filters to eliminate environmental noise, we stream the processed data through a Web Serial API to an interactive dashboard featuring a dynamic digital twin. We evaluated the system under baseline, mechanical, salinity, and drought conditions to prove its sensitivity. Specifically, signal volatility (Std Dv) surged from a mechanical baseline of 7.25 and a healthy average of 44.33 to 73.69 under salinity stress and 72.73 during drought, marking a significant physiological shift. This framework is ideal for live applications in greenhouse and hydroponic environments, where sentinel plants act as early-warning nodes, enabling automated responses that advance sustainable agriculture.

Competition history

• ISEF 2026 Plant Sciences · Entry PLNT042T

Resources

• ISEF abstract page ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.