Automating agriculture: using UASs to monitor for environmental and management benefits

This project focuses on black-grass, a weed that has developed resistance to a range of herbicides that have been used to control it. In 2014, 58% of the UK’s wheat crop contained black-grass with 22% expected to suffer a moderate yield loss of 5% or more. We propose to develop a system for monitoring black-grass with a Unmanned Ariel System (UAS), more commonly referred to as a drone.

Image data will be collected by flying the UAS over fields across the UK, making this study the largest of its kind undertaken in this country. The results will come from image analysis, using ‘feature learning’ techniques so that software can develop the ability to identify black-grass from images of entire fields. The fields, or even patches of fields, will then each be automatically assigned different management strategies. This allows farmers to act sooner against black-grass, making them more likely to have an impact on the problem, resulting in better yields and greater food security.

Supervisor

Professor Rob Freckleton

Department of Animal and Plant Sciences

Co-Supervisors

Dr Dylan Childs

Department of Animal and Plant Sciences

Professor Sandor M. Veres

Department of Automatic Control and Systems Engineering

Professor Tony Prescott

Department of Psychology