CQUniversity researchers help teach AI to see the Australian paddock
CQUniversity researchers are helping an Australian agricultural technology company build smarter artificial intelligence systems designed specifically for the unique conditions of Australian farming.
Through the Queensland Government-funded Regional University Industry Collaboration (RUIC) program, delivered by CSIRO, CQUniversity partnered with Queensland-based agtech company InFarm to tackle one of the biggest challenges facing agricultural AI: ensuring the technology can perform reliably in the highly variable conditions of Australian farms.
The collaboration brought together CQUniversity researchers Professor Philip Brown, Associate Professor Paul Kwan and Dr Mitchell Woodbright with InFarm’s commercial team to develop foundational AI models that can learn and adapt to local farming environments.
InFarm is best known for its weed-detection platform, which combines drone imagery with a farmer’s existing tractor equipment to target weeds with precision and reduce herbicide use by up to 95 per cent.
However, developing agricultural AI that works across different crops, farming systems and regions requires much more than identifying weeds in a single paddock.
Agriculture presents a particularly difficult environment for artificial intelligence, with constantly changing variables including lighting conditions, crop growth stages, weed species, row spacing and seasonal differences. In Australia, those challenges are amplified by the diversity of farming systems and environmental conditions across regions.
Many existing agricultural AI models have been developed using data from Europe and North America, making it difficult for those systems to consistently perform in Australian paddocks.
To address this, the project focused on teaching AI models to learn from large volumes of unlabelled images, reducing the time and cost associated with manually tagging thousands of images while improving the accuracy and adaptability of the systems.
The team also developed methods to identify unusual scenarios and "edge cases" that frequently occur in agriculture and can cause conventional AI models to fail in real-world settings.
Associate Professor Paul Kwan from CQUniversity’s School of Science and Technology said the project highlights how advanced research can be translated into practical solutions for industry.
“This project is about making AI work in the real-world complexity of Australian agriculture,” Associate Professor Kwan said.
“By developing models that can learn from unlabelled data and adapt to highly variable conditions, we’re helping bridge the gap between cutting-edge research and practical tools farmers can rely on in the paddock.
“The collaboration with InFarm has allowed us to test and refine these approaches in a genuinely applied setting, which is critical for building robust, scalable agricultural AI.”
Another key focus of the project was ensuring the technology could operate in environments with limited internet connectivity, enabling farmers in remote and regional areas to benefit from AI-powered tools in the field.
CQUniversity Director, Office of Research Jessica Suna said the partnership demonstrates the value of connecting regional research expertise with industry challenges.
“Collaborations like this demonstrate the value of connecting regional research expertise with industry challenges,” Ms Suna said.
“Through the RUIC program, CQUniversity researchers are working alongside innovative companies like InFarm to translate advanced AI capabilities into tangible outcomes for Australian agriculture.
“It’s a strong example of how applied research can drive productivity, sustainability and regional impact.”
The RUIC program supports collaborative research and development projects by connecting Queensland startups and small-to-medium enterprises with researchers from regional universities. The initiative provides matched funding and embeds researchers directly within businesses to solve real-world challenges.
InFarm Managing Director Jerome Leray said the collaboration had demonstrated the value of bringing researchers and industry together to tackle practical challenges.
“Working with CQUniversity through the RUIC program has been a great experience for InFarm. We’ve really enjoyed bringing researchers into industry and working alongside them to solve some of our biggest real-world challenges,” Mr Leray said.
“We strongly encourage these types of industry and research collaborations. Bridging the gap between researchers and industry is something Australia can leverage to achieve better outcomes for society, industry and the research community.
“That’s why we’re so supportive of CQUniversity and the RUIC program, and we hope this collaboration is the start of closer industry and research collaboration.”
With the project helping strengthen InFarm’s AI capabilities, the company is now expanding the technology into new areas including grazing systems and animal health. The collaboration highlights the growing role CQUniversity researchers are playing in helping industry partners solve complex challenges and deliver practical innovations for regional Australia.
