Key points are not available for this paper at this time.
Artificial intelligence (AI) has long been applied in agriculture and has become especially prevalent in recent years. AI especially deep learning technique have progressed to have much stronger learning ability to learn more useful features to handle even more complicated task in the field of precision agriculture. Challenges in agriculture such as disease, infestation, inadequate irrigation and soil treatment, and poor crop management have brought about crop losses and adverse effects on the environment. Not to mention, the ever-increasing demand of agricultural product due to the increasing global population and the limited amount of arable land. To be overcome, those challenges need innovative approaches, ones that could benefit from AI’s flexibility, accuracy, cost-effectiveness, and generally superior efficiency. AI technology whose aim is to mimic the ability of humans to solve problems especially in decision making enables agricultural activities to be done more efficiently while reducing human interference. The use of AI in agriculture has evolved from the application of fuzzy logic, then into machine learning and deep learning. Some deep learning methods that have been applied in precision agriculture are convolutional neural network, transformer learning, meta deep learning, and lightweight deep learning. This paper presents a review of 100 research papers addressing the application of AI in overcoming challenges in agriculture from the year 2000 to 2023. The paper selection for this review paper is done by using the SALSA method to effectively identify relevant research papers. In the near future, AI will be ubiquitous in the global agricultural sector and will bring about new technologies, new knowledge, and endless possibilities.
Mustaza et al. (Fri,) studied this question.