Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Sustainable Agricultural Systems Laboratory » Research » Research Project #443466

Research Project: Building Resilient Organic Weed Management Systems with Precision Smart Sprayer Technologies

Location: Sustainable Agricultural Systems Laboratory

Project Number: 8042-22000-167-063-A
Project Type: Cooperative Agreement

Start Date: Oct 1, 2022
End Date: Jul 31, 2026

Objective:
Weeds are a big challenge in organic field crop production. The overall goal of this project is to automate weed management with smart technologies that are driven by robots and tractor platforms. Specifically: 1. Regionalize weed detection, identification, and mapping functionality for autonomous robots 2. Quantify optimal combinations of robot mounted mechanical weed management implements in organic corn, soybeans, and cotton across diverse soil and climate conditions: 3. Quantify effects of targeted weed control on economics, crop yield loss, and subsequent weed seedbank dynamics and species shifts 4. Facilitate knowledge exchange about alternative weed control tactics using robotics among farmers, researchers, and educators

Approach:
The collaborative team will build a digital image repository for identifying weeds from crop. This will expand on an existing effort to build plant image repositories of crops, cover crops, and weeds. A synthetic image pipelines will be constructed to scale up overall training data for weed detection. The training images will be used to build a model on a OAK-D camera that has RGB and stereo camera capabilities. This camera is important for species and size classifications. Efforts will then be made to link the camera species detection with a robotic system that delivers micro-volumes of organic herbicides. Field testing of the entire system will be conducted on both robotic platforms and tractor mounted systems. The efficacy of weed targeting, herbicide activity, and overall economics will be quantified.