Document type : Review published in Computers and Electronics in Agriculture
Authors: Rodrigo García, Jose Aguilar, Mauricio Toro, Angel Pinto, Paul Rodriguez
Preview: This article presents a systematic literature review of recent works on the use of machine learning (ML) in precision livestock farming (PLF), focusing on two areas of interest: grazing and animal health. This review: (i) highlights opportunities for ML in the livestock sector; (ii) shows the current sensors, software and techniques for data analysis; (iii) details the increasing openness of data sources. It was found that the use of ML in PLF is in a stage of development and has several research challenges. Examples of such challenges are: (i) to develop hybrid models for diagnosis and prescription as a tool for the prevention and control of animal diseases; (ii) to bring together the grazing and animal health issues; (iii) to give autonomy to PLF using autonomous cycles of data analysis tasks and meta-learning; and (iv) to bring together soil and pasture variables because, for both, animal health and animal grazing, the variables used are only behavioral and environmental.
