Using AI in Australia to detect weeds in sugarcane fields

Designed in partnership with James Cook University researchers, agricultural technology company AutoWeed and Sugar Research Australia, the machine (in the above video) uses artificial intelligence (AI) to detect and spray weeds, like nutgrass, in a cane paddock, generating savings compared to conventional blanket spraying methods. “In the best cases where weed infestation has been lower, we have saved up to 65 per cent in herbicide usage,” said Lead project researcher Associate Professor Mostafa Rahimi Azghadi (JCU).

A water quality assessment produced by Sugar Research Australia during testing also found that paddocks which used the AutoWeed AI sprayer reduced their chemical concentration in water run-off by an average of 46 per cent.

Some of the weeds in the trials have included nutgrass. Terry Granshaw, SRA District Manager Burdekin said: “Herbicides that control nutgrass are expensive. Weeds like nutgrass come in patches, so instead of a total blanket spray of the farm, we are only spraying the areas we need to. If we can spray less of it, we’re saving money and we’re doing a better job for the environment.”

SRA’s Chief Investigator, Emilie Fillols, said the team has now progressed trials on other weed species and are working on technical issues with shadows.

The project, funded by a grant through the partnership between the Great Barrier Reef Foundation and the Australian Government’s Reef Trust is a collaboration between JCU, AutoWeed and Sugar Research Australia.

To read more about this project, view our Research Missions page .