In Australia, Queensland University of Technology researchers debuted Agbot II, a fully-autonomous robot farm worker that sees and kills weeds before they can impact crop health; potentially reducing weed control by 90 percent and saving Australian farmers $1.3 billion per year. The robot’s sensors, software and other electronics enable it to navigate through a field, detect and classify weeds and then kill them either mechanically or chemically.
Central to Agbot II’s efficiency is cutting edge robotic vision and artificial intelligence for the detection and classification of different weed species. A key problem is weeds that are becoming increasingly immune to chemical control, making it important for the robot to not only identify but classify and decide what action to take for effective treatment. Concentrating efforts on three weeds relevant to Queensland, the researchers reported 99% accuracy in the classification of the correct species based on the images collected by the robot cameras.