Agricultural process data as a source for knowledge: Perspective on artificial intelligence

Abstract

In visions of the future of agriculture, it is predicted that Artificial Intelligence and Robotics will revolutionize farming. Artificial Intelligence (AI) is not always clearly visible to the end users, who are in this case farmers. AI methods are usually incorporated in existing Farm Management Information Systems or in some cases in separate Decision Support Systems; or the AI is part of the machine or robot operation. Currently, AI methods are used mostly in machine vision applications. Practical applications that use methods other than image data are scarce. However, agricultural machines produce an ever-increasing amount of process data during agricultural operations. There is huge potential to gain useful information from this data. This paper presents a case example of treatment-zone determination based on agricultural process data using clustering methods. This paper also describes how AI is incorporated into the Cropinfra research data collection platform. Based on the experiences gained from Cropinfra-related research projects and a review of research papers, we predict the future directions of Artificial Intelligence in agriculture.

Keywords

• Decision support systems
• Farm management information systems
• Precision farming
• Future directions
• Interactive machine learning