Weed whacker to protect the Reef
JAMES Cook University is to oversee the development of a smart weed-spraying robot designed to reduce the use of herbicide on Far North cane farms.
The device – which is the focus of a two-year project funded by a $400,000 Great Barrier Reef Foundation and Australian Government’s Reef Trust grant – will reduce the use of chemicals in the Reef’s catchment areas by 80 per cent.
JCU, AutoWeed and Sugar Research Australia are combining to refine the smart spot spraying systems with the objective of reducing the herbicide run-off that presents a threat to plants and animals in rivers, creeks, coastal and inshore areas.
Mostafa Rahimi Azghadi, the senior JCU engineering lecturer leading the project, says researchers will work with AutoWeed engineers while Sugar Research Australia will assess water quality.
“Most herbicides, being mobile in soil, are carried in river run-off and have been detected in GBR ecosystems at concentrations high enough to affect organisms,” he says.
“Sugarcane farms are only 1.4 per cent of the GBR catchment area but contribute 95 per cent of the pesticide load draining to the GBR.
“We’re aiming to design, develop and trial the spot spraying method and fit it to a high-rise self-propelled boom to be used on a sugarcane farm.”
During the project’s first year thousands of images of sugarcane crops will be collected, labelled by a human expert, and fed into deeplearning models to train the weed and crop detection system.
This will be an ongoing process and, every time the spraying system is used, it will collect additional data so the deep-learning models can continue to tweak performance.
The second year will focus on developing and trialling the herbicide delivery component.
AutoWeed co-founder and engineer Jake Wood says the new system will use “stored images of weeds to detect and spray them without hitting non-target crops”.
“Extending our AutoWeed spot spraying technology to sugarcane requires significant new research and development (and) we aim to reduce knockdown herbicide usage on sugarcane farms by at least 80 per cent,” he says. “This will incentivise water-quality improvements in reef catchment areas by reducing weed management costs for farmers while also lowering the concentration of herbicides in run-off to support a healthy Reef.
“AutoWeed technology previously targeted weeds in cattle farm pastures and broadacre crops but sugarcane presents unique challenges.
“The project will use deep convolutional neural networks – the very same used by Facebook to detect faces and Google to optimise image searches – but it will be the first time it’s been applied to sugarcane.”
WE’RE AIMING TO DESIGN, DEVELOP, AND TRIAL THE SPOT SPRAYING METHOD AND FIT IT TO A HIGH-RISE SELF-PROPELLED BOOM TO BE USED ON A SUGARCANE FARM