While current monocultural farming methods have helped ensure the stability of food supplies in developed countries, this dominant way of farming has come at a cost. Emissions from land use, largely agriculture, forestry and land clearing, make up around 21% of the world’s greenhouse gas emissions. To address these issues, CSIRO Data61 has launched the Reimagine Farming Program, as part of the Reinvent Science stream of research. The program aims to advance agricultural science, robotics, and AI to create a more sustainable future for farming. In the past, autonomous and intelligent technologies have been developed by Data 61 for a broad range of agriculture applications, including 3D canopy condition monitoring, aerial phenotyping, and early plant disease detection.

Current farming robots and sensors are limited to performing singular, fixed tasks, and they lack the ability to adapt to changing environments. To create profitable and climate-resistant farms in the future, robots and AI must possess multiple skills and work collaboratively with humans, leveraging on humans’ skills, to manage dynamic environments. We address this challenge in the Collaborative Intelligent Future Science Platform by developing the Human-Machine Collaborative Learning platform. This platform will enable multiple agents to collaborate with humans, selectively sharing knowledge gained through actions and experiences to continuously improve human and machine interactions. Human-Robot Teams will thus be able to cope with the complexities of unstructured, non-rigid environments.