YOLOv8-based Agricultural Robots for Precision Pollination and Yield Prediction

Abstract

Precision agriculture uses advanced technology to enhance agricultural yield and sustainability. This research presents a YOLOv8-based agricultural robotic system for precise pollination monitoring and production forecasting. The system incorporates high-resolution cameras and IoT-enabled sensors to identify flowers, monitor pollinators, and evaluate fruit growth in real time. Utilizing YOLOv8's deep learning framework, the system attains exceptional detection accuracy, with an average precision (AP) of 99.4% for flower identification and 98.7% for pollinator tracking. The obtained data is analyzed using a cloud-based machine learning framework, which facilitates yield prediction with 99.4% accuracy, surpassing conventional human estimation techniques by 22.8%. Experimental findings from tomato greenhouse research indicated a 25.6% enhancement in pollination efficiency, resulting in improved fruit set and increased yield. The AI-driven robotic system improves precision agriculture via real-time monitoring, less labor reliance, and enhanced decision-making capabilities. It highlights the potential of YOLOv8-enabled agricultural robots to transform smart farming via precise pollination monitoring and accurate production prediction. Future work will concentrate on scalability for open-field crops and real-time adaptive pollination strategies to enhance agricultural efficiency.