Fresh funding for AI, drone research in weed management
CQUniversity tech experts are supporting farmers to target weeds with an AI-driven and drone-delivered solution – and the innovative world-first approach has been backed by fresh Federal Government funding.
A new $1,136,194 grant will help take the smart approach to patent and commercialisation, after Central Queensland field trials proved successful in cotton crops earlier this year.
The CQU project offers Australian farmers a green and cost-saving game-changer in their $5 billion battle against weeds every year.
It’s among 39 research initiatives awarded grants under round one of Australia’s Economic Accelerator (AEA) Innovate program, announced by Australian Minister for Education Jason Clare MP last week.
The new two-year research initiative is aimed at transforming weed management practices through advanced technology.
The project is led by CQUniversity researchers Dr Nahina Islam, Dr Jahan Hassan and Associate Professor Biplob Ray, who bring deep expertise in Artificial Intelligence (AI), Internet of Things (IoT) and agricultural informatics.
Supporting the development and application of advanced robotics, drone engineering and agronomic design are Dr Md Nurun Nabi, Dr Lasi Piyathilaka, Dr Zhenglin Wang, and Associate Professor Surya Bhattarai.
The research will also benefit from cross-institutional collaboration with Professor Stephen Xu from Charles Darwin University, whose leadership in sustainable agriculture will strengthens the project’s national impact.
The innovative project builds on previous work developing a world-first prototype that integrates image processing, AI, GPS and lightweight autonomous drones for precision herbicide spraying.
This collaborative effort extends beyond academia, engaging with Australian industry partners including Camzilla, Impexpo Pty Ltd, ConnectAuz Pty Ltd, Foxwell Farms, DQ Commodities, and Logical Analysis, as well as Northern Territory’s Department of Agriculture and Fisheries.
Together, the team aims to advance the technology to commercialisation, helping to deliver sustainable, cost-effective solutions for the agricultural sector.
“Traditional methods to combat weed infestations, such as blanket herbicide spraying, are costly and harmful to the environment,” Dr Islam said.
“Our project will take our proven technology intervention, using a drone-based, AI-driven weed detection and management system, and grow our capabilities in real-world agricultural environments for green-on-green targeted weed management.”
Early industry-partnered development, led by CQU’s AI, drone, IoT, mechatronics and agriculture specialists, held field trials at an Alton Downs cotton farm in February.
Minister Clare said the grants strategically targeted future-proofed solutions.
“These investments allow our world-class universities and researchers to work on game-changing projects that are good for our economy and good for Australia,” he said.
Australia’s Economic Accelerator brings together research and industry to create investor-ready commercial enterprises to provide jobs of the future in priority industries, and the Innovate grants support proof-of-scale commercialisation projects.
Learn more at the Innovate Grants website, and explore CQUniversity projects across agriculture and innovation at CQU’s Institute for Future Farming Systems and Centre for Machine Learning – Networking and Education Technology.
Transcript
This project is about developing a a solution for smart herbicide spraying for green on green weed management in cotton farm solutions where drones are used for weed location identification and spot spraying exist in isolation.
So there is a need for a coordinated approach where um there would be communication from one drone which captures farmland imagery to you know use image processing and AI to detect weed locations and then the identified weed locations coordinates needs to be automatically transferred to the spraying drone so that the spraying drone can spray only on the locations where weed is identified.
So internet reliance dependence is a hindrance in many remote locations in Australia where the farmlands may not have stable internet connection.
So traditionally speaking the image processing or downloading the images to the cloud requires that stable internet connectivity.
But our solution actually has avoided the need for that internet dependence so it can work in a localised manner in remote areas.
Today we are at Peter Foxwell's farmland at Alton Downs in Central Queensland Rockhampton we are very thankful to Peter of course for allowing us to use his cotton farmland as our test site look it's it's an interesting tool that would probably save a lot of chemical if we're if we're targeting weeds specifically spot spraying instead of blanket spraying or broad spraying like we do in our cotton so there's potential there but I'm probably thinking more about the opportunities in other crops as well.
There's we we're just focusing on cotton now but why not why not other broad acre crops why not horticulture crops yeah I guess I guess it's limitless it' be a lot easier if I could sit in the office and control a sprayer that's for sure Um I got plenty plenty to do, Would save time Would if I'm using less chemical it's going to save money so it makes sense in that regard.
Just another tool I mean I've got tractors in the shed that cost a lot of money but yeah this could be a big saving It's fantastic It's so exciting that yes we have done it We have developed a intelligent autonomous targeted sprayer system which reduces the manual intervention that means reducing labor cost.
We are reducing the use of herbicide by 1/4 that is we are making it to 25% compared to the blanket spraying which is a huge benefit for the farmer and then we are reducing the time needed the battery life that means we're reducing the energy consumption to half of the regular traditional users.
Oh look I'm I'm the first farmer to see this this stuff happening we get to see some amazing things great ideas and new technologies and and we're nice and handy to the uni which is nice but you know I get to see these things first that's pretty exciting the future is really wonderful this is a very good beginning but we would like to see these kind of products or solutions become accessible to the end users for that a lot more research is needed and development is needed but the ultimate goal would be that the farmers benefit from it from this wonderful solution.
Look I think it's it's probably the way of the future with ground-based aerial based remote sensing It it all makes sense if you can identify problems remotely and deal with them remotely It's it it makes a lot of sense to me So yeah it's it's the way of the future I think.
