Robotic Monitoring of Colorimetric Leaf Sensors for Precision Agriculture

Current remote sensing technologies that measure crop health e.g. RGB, multispectral, hyperspectral, and LiDAR, are indirect, and cannot capture plant stress indicators directly. Instead, low-cost leaf sensors that directly interface with the crop surface present an opportunity to advance real-time direct monitoring. To this end, we co-design a sensor-detector system, where the sensor is a novel colorimetric leaf sensor that directly measures crop health in a precision agriculture setting, and the detector autonomously obtains optical signals from these leaf sensors. This system integrates a ground robot platform with an on-board monocular RGB camera and object detector to localize the leaf sensor, and a hyperspectral camera with motorized mirror and an on-board halogen light to acquire a hyperspectral reflectance image of the leaf sensor, from which a spectral response characterizing crop health can be extracted. We show a successful demonstration of our co-designed system operating in outdoor environments, obtaining spectra that are interpretable when compared to controlled laboratory-grade spectrometer measurements. The system is demonstrated in row-crop environments both indoors and outdoors where it is able to autonomously navigate, locate and obtain a hyperspectral image of all leaf sensors present, and retrieve interpretable spectral resonance from leaf sensors.

@article{hopkins2025_2505.13916,
  title={ Robotic Monitoring of Colorimetric Leaf Sensors for Precision Agriculture },
  author={ Malakhi Hopkins and Alice Kate Li and Shobhita Kramadhati and Jackson Arnold and Akhila Mallavarapu and Chavez Lawrence and Varun Murali and Sanjeev J. Koppal and Cherie Kagan and Vijay Kumar },
  journal={arXiv preprint arXiv:2505.13916},
  year={ 2025 }
}