Seminar : Automation and Deep Learning to Advance Digital Agriculture

Details

To sustainably intensify agricultural production and food supply while preserving the environment, we must radically improve the efficiency and resilience of our agri-food systems through automation and AI-driven digital agriculture. In this talk, I will go over multiple research projects that leverage agricultural robotics and deep learning to address challenges spanning the food chain from breeding to harvest and postharvest handling. I will present a novel modular agricultural robotic system (MARS) that is an autonomous, multi-purpose, and affordable robotic platform for in-field automated phenotyping and precision farming. The robotic system is empowered by machine learning-based vision intelligence, including object detection and semantic/instance segmentation for detecting plants and plant parts in 2D images, video frame-based multi-object tracking for plant organ counting, and 3D deep learning models for point cloud segmentation and architectural trait extraction. Another project will highlight a patented sensor to emulate berry fruit and quantify mechanical impacts during the mechanical harvesting and postharvest handling processes, as well as a deep learning-based hyperspectral imaging approach for berry internal bruise detection and quantification.