Automatic Learning in Agriculture:A Survey

Abstract

Agriculture plays a pivotal role in the growth of any nation. Nowadays, with the advancements of information technologies, big data are generated and executed using faster computing techniques. machine learning techniques have been used to automate and improve agricultural activities for a long time. This paper presents a review of existing applications of machine learning in agriculture with a focus on the applications of Deep Reinforcement Learning techniques in agriculture. The conventional solutions for agricultural machine learning decision-making problems are using supervised approaches. In supervised learning, the machine needs to be trained on samples of inputs and outputs to support decision making. While, in reinforcement learning, sequential decision making happens and the next input depends on the decision of the machine. In this paper, we perform a survey of 49 publications of which 10 were secondary research e orts that discussed a variety of machine learning approaches applications in agriculture and 39 research e orts that used machine learning approaches to support the automation of agricultural activities.