Location: Sugarbeet Research
Project Number: 3060-21000-045-014-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Sep 1, 2023
End Date: Dec 30, 2024
Objective:
(1) Develop artificial intelligence algorithms based on Edge-AI computing technology for weed species identification (susceptible and resistant) in sugarbeet fields; (2) Develop a UAV-based system that can recognize weed geo-location/species and create a prescription map to feed the UGV system; (3) : Develop a UAV-UGV based weed identification system that can recognize weed location and species in real time between crop rows and target spray/mechanical remove the weed automatically.
Approach:
For Objective 1, weed and sugarbeet greenhouse and field imagery data will be collected to train Edge-AI technology and deep learning neural network models. The model will then be deployed to include onboard camera systems to enable autonomous weed identification and mapping during field operations. For Objective 2, a UAV equipped with a high-resolution camera and GPS will capture grower fields to capture crop and weed images, which will processed to enable the geo-location of weeds in the field. For Objective 3, the technologies developed in previous objectives will be utilized to develop an unmanned ground vehicle that can autonomously detect and apply herbicide or mechanically remove the identified weeds.
