Enabling Smart Farming Through Edge Artificial Intelligence (AI)

2. Fundamentals Of Edge Ai

Edge AI, is the process of deploying artificial intelligence algorithms on local or edge devices instead of relying on centralised cloud servers. This process will reduce the latency allowing data to be processed in the local edge device itself. Edge AI can be used for applications requiring low latency and prone to high risks of privacy and security. This technique is used much in applications involving machine learning models. The flow of the steps involved in Edge AI is specified in figure 1. The training process of machine learning models involving large real-time datasets requires high computational facility for processing. After training process the model is deployed on the edge AI process facilitating AI features for real-time applications (Singh et al., 2023). Thus, the flow of Edge AI involves the deployment and execution of machine learning models on edge devices performing data processing and decision-making locally as shown in Figure 1.

Figure 1.

Edge AI process flow

Alt-text: Figure 1 displays the flow of Edge AI (Prajapati et al., 2023) is characterized by its ability to process data locally, reducing the dependency on cloud infrastructure, provide real-time decision-making capabilities, and address privacy and security concerns.

Also, it is considered a paradigm that brings intelligence closer to the source of data, enabling innovative applications across various domains such as agriculture, healthcare and more. With the increased demand for GPUs, NPUs and more edge AI technology has grown significantly. This demand is noticeable, as machine learning and artificial intelligence are currently trending technologies.