Artificial Intelligence in the IoT Era: A Review of Edge AI Hardware and Software

Abstract

The modern trend of moving artificial intelligence computation near to the origin of data sources has increased the demand for new hardware and software suitable for such environments. We carried out a scoping study to find the current resources used when developing Edge AI applications. Due to the nature of the topic, the research combined scientific sources with product information and software project sources. The paper is structured as follows. In the first part, Edge AI applications are briefly discussed followed by hardware options and finally, the software used to develop AI models is described. There are various hardware products available, and we found as many as possible for this research to identify the best-known manufacturers. We describe the devices in the following categories: artificial intelligence accelerators and processors, field-programmable gate arrays, system-on-a-chip devices, system-on-modules, and full computers from development boards to servers. There seem to be three trends in Edge AI software development: neural network optimization, mobile device software and microcontroller software. We discussed these emerging fields and how the special challenges of low power consumption and machine learning computation are being taken into account. Our findings suggest that the Edge AI ecosystem is currently developing, and it has its own challenges to which vendors and developers are responding.

Edge AI: A survey

“…Smart cities rely on edge computing, which allows sensors on traffic lights and other public utilities to communicate with smartphones and apps. It also helps sensor-equipped machines in "smart factories" report on its status and performance in real-time [63]. The potential to deploy AI on the Edge has sprung up and flourished in recent years, thanks to the growing complexity of activities that must be done by edge devices [59].…”

“…Instead of a data centre, edge computing processes data at the network's edge. This provides real-time processing and lower latency, but AI implementation is complicated by constrained CPU cores and network access [63]. We will examine these problems and potential and highlight crucial considerations to consider when implementing AI systems in edge computing and other resource-constrained situations in this section [64].…”

“…Moreover, there are a number of challenges that must be overcome in order to successfully implement AI systems in edge computing as well as other resource-limited situations [50,[63][64][65].…”

A Survey on Optimization Techniques for Edge Artificial Intelligence (AI)

“…In Ref. [ 7 ], the authors performed a survey with a specific focus on AI hardware platforms and software packages. In Ref.…”

“…Original private data cannot be used for model optimization. Distribution of computation to devices is also limited by limited communication resources of the devices [ 7 ]…”