Internet of Intelligent Things: A convergence of embedded systems, edge computing and machine learning

This article comprehensively reviews the emerging concept of Internet of Intelligent Things (IoIT), adopting an integrated perspective centred on the areas of embedded systems, edge computing, and machine learning. With rapid developments in these areas, new solutions are emerging to address previously unsolved problems, demanding novel research and development paradigms. In this sense, this article aims to fulfil some important research gaps, laying down the foundations for cutting-edge research works following an ever-increasing trend based on embedded devices powered by compressed artificial intelligence models. For that, this article first traces the evolution of embedded devices and wireless communication technologies in the last decades, leading to the emergence of IoT applications in various domains. The evolution of machine learning and its applications, along with associated challenges and architectures, is also discussed. In this context, the concept of embedded machine learning (TinyML) is introduced within the context of the Internet of Intelligent Things paradigm, highlighting its unique characteristics and the process of developing and deploying such solutions. Furthermore, we perform an extensive state-of-the-art survey to identify very recent works that have implemented TinyML models on different off-the-shelf embedded devices, analysing the development of practical solutions and discussing recent research trends and future perspectives. By providing a comprehensive literature review across all layers of the Internet of Intelligent Things paradigm, addressing potential applications and proposing a new taxonomy to guide new development efforts, this article aims to offer a holistic perspective on this challenging and rapidly evolving research field.