A Swarm-Coordinated Multi-Robot System for Early Stress Detection in Agricultural Rows Using Multimodal Leaf Sensing

Problem

The paper addresses the gap in affordable and accessible crop stress detection methods for medium and small-scale farmers. Current solutions often lack the necessary affordability and accessibility, which limits their adoption in these farming communities. This work is presented as a preprint and has not undergone peer review.

Method

The authors propose CropSentry, a low-cost, ground-based multi-robot system designed for real-time crop health monitoring. The system comprises two autonomous robots equipped to detect leaf color and gather environmental data. Data collection involved 63 observations during experimental trials. The output is a real-time web-based dashboard that displays crop health metrics. The system achieved a 100% wireless communication success rate across 10 slave observations, ensuring reliable data transmission.

Results

The system demonstrated the following classification accuracies for crop health:

• Overall Crop Health Classification Accuracy: 84.12%
• Healthy Plants Accuracy: 82.60%
• Nutrient-Deficient Plants Accuracy: 88%
• Diseased Plants Accuracy: 80%
• Wireless Communication Success Rate: 100% No baseline comparisons were reported for these metrics, indicating that the results stand alone without direct competition.

Limitations

The authors do not explicitly state any limitations in their work. However, potential limitations could include the system's reliance on specific environmental conditions, which may affect its performance in diverse agricultural settings. Additionally, the scalability of the system for larger agricultural operations remains unaddressed.

Why it matters

This research has significant implications for the agricultural sector, particularly for small-scale farmers who may benefit from low-cost, efficient monitoring solutions. By providing a reliable method for early stress detection, CropSentry could enhance crop management practices, leading to improved yields and resource utilization. The findings may also inspire further research into scalable robotic systems for agricultural applications.