The growing phenomenon of crop resistance to herbicides and the spiralling costs of weed management continue to increase the cost of food production and reducing farmers' ability to maintain profitability. The cost of weed management to Australian farmers now exceeds $4 billion dollars per year in addition to the indirect costs that result from yield loss, poor returns, loss of productive areas and contamination as a result of weed growth.
Site-specific crop management is dependent on sophisticated monitoring technologies that can detect the presence of weeds and therefore enable precision spraying of herbicides. This creates a demand for real-time monitoring solutions that are usable at common operating speeds in order to detect individual weeds and release herbicides only when weeds are detected.
Through an industry project with Photonic Detection Systems (PDS) Pty Ltd, the Electron Science Research Institute successfully demonstrated an advanced photonics-based weed sensor for differentiation between various plant types.
This successful research program demonstrated the potential for development of a ground based sensor engine and associated systems that could enable real-time detection of weeds within crops allowing for highly precise spraying of weeds. This enables, for the first time, spraying of only weeds in the post emergent stages of crop growth. It is this stage in the crop cycle where weed control is critical to increasing crop yield.
This project is currently focussing on the development of a pre-production prototype that is capable of discriminating key commercial crops from the majority of weed species in a real-world environment at regular operating speeds. The prototype is a world leader in that it is designed to perform the following functions:
The technology can be used across a variety of weed control applications beyond the farm including roadside verges, footpaths, railways and any situation requiring selective weed management.
Please leave a comment about your rating so we can better understand how we might improve the page.