Artificial Intelligence for Digitising Industry - Technologies

Industry 5.0 is the next major evolution of intelligent automated and autonomous industrial environments. Advances in artificial intelligence, robotics, the industrial internet of things, edge computing, digital twins, connectivity, and nature-inspired innovation to enhance human-machine and machine-machine collaboration are achieving novel manufacturing solutions that are both efficient and environmentally friendly. Advances in autonomous real-time manufacturing systems enabled by AI, robotisation, and hyperconnectivity are essential in the transition to Industry 5.0.The first chapter provides an overview of the foundations of artificial intelligence, highlighting concepts and technologies used in industrial environments and the emergence of Industry 5.0 applications. The shift of AI methods from cloud to edge as the basic approach for digitising industries marks the start of a general transition in the control of industrial processes and in the functionality of devices.From the perspective of various industry practitioners and researchers, the next two chapters look at several new developments in machine learning, deep learning, and the hardware and software technologies that are anticipating the development of Industry 5.0. The chapters discuss end-to-end AI technology stacks that reflect research and innovation in AI hardware (e.g., neuromorphic, CPUs, GPUs, FPGAs, ASICs), interfaces, platforms (e.g., frameworks, algorithms, architectures/topologies, methods), training, applications, and services.Chapters four and five consider the ethics of AI technology and the standardisation of applications in industrial environments. These chapters present an overview of the ethical considerations involved in digitising industry and the potential impact of AI on economic growth and productivity, the workforce, the digital divide, and such features of productive and economic processes as trustworthiness, transparency, and fairness. Relevant issues include the concentration of AI technology within only a few companies, human-machine relationships, and behavioural and operational misconduct.We discuss the role of AI standardisation in digitising industry and the objectives of industrial sectors, including the requirements and challenges of standardisation; the landscape of AI standards, the activities of Standards Development Organisations (SDOs), and the approaches to AI standards of industrial stakeholders; and a recommended roadmap for this standardisation. Finally, three chapters focus on three projects (AI4DI, TEMPO and ANDANTE) that address state-of-the-art developments in AI-based hardware and software, highlighting research challenges and open issues regarding the development and implementation of AI-based technologies in industrial sectors.The concepts and AI technologies presented in this book are essential for facilitating sustainable industrial development, reducing energy consumption, and- by supporting the optimisation of products, services, and processes through real-time data collection and intelligent analytics - mitigating the carbon footprints and environmental impact of industrial processes.