A Bayesian method for estimation of plant soil water content with application to low-cost horticultural robotics
Southgate, A. J.
Show abstract
Climate change represents a challenge to food security by interfering with the environmental conditions needed for productive plant growth. While technology can be used for partial mitigation, access to technology is inequitable. Low-cost microcontrollers, such as the ESP32, have recently lowered the barrier for entry into prototyping smart devices. ESP32s equipped with capacitive moisture sensors have been suggested for low-cost smart plant watering systems. However, measuring moisture in soil is complex, potentially destructive, and requires careful calibration in order to characterise the response curve mapping soil water content to sensor measurements. Here, we developed a Bayesian method for estimating the inverse response curve from capacitive moisture sensor data, known water doses, and prior uncertainty, bypassing the need for destructive gravimetry. This method constitutes the core calibration module of the open-source OpenHCult software system for low-cost horticultural automation.
Matching journals
The top 9 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- Classification of daily crop phenology in PhenoCams using deep learning and hidden markov models 91%
- Individual tree-crown detection in RGB imagery using semi-supervised deep learning neural networks 90%
- A satellite-based spatio-temporal machine learning model to reconstruct daily PM2.5 concentrations across Great Britain 89%
Similar papers in this journal
- An Assistive Computer Vision Tool to Automatically Detect Changes in Fish Behavior In Response to Ambient Odor 91%
- Statistical analysis of three data sources for Covid-19 monitoring in Rhineland-Palatinate, Germany 91%
- Recommendations to address uncertainties in environmental risk assessment using toxicokinetics-toxicodynamics models 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.