A cyber-physical system (CPS) forms the borderline between the physical world in all its diversity and complexity, and the world of information processing that models the physical world, comprehends and… Click to show full abstract
A cyber-physical system (CPS) forms the borderline between the physical world in all its diversity and complexity, and the world of information processing that models the physical world, comprehends and predicts it for ultimately controlling it. CPSs, just like the physical world itself, are distributed, but where the data analysis, processing, computing, and control should take place remains a question that every application has to address for itself to find a reasonable solution that depends on physical distances, complexity of the task, requirements on response time, safety, security, and privacy considerations. Computing efficiency, generality, and flexibility of allocation of vast computing resources favor the realization of computing tasks in server farms, but various constraints on available communication bandwidth and energy supply as well as demands on latency, safety, and privacy drive distributed processing and the provision of local intelligence. The vision of self-aware computing is that a computing system has a comprehension of itself and its environment that is complete in the sense that it can relate to all relevant observations and make all required decisions. The field of self-aware computing systems emerged in the early 2000s, with IBM’s vision of autonomic computing formulated in 2003 as one main driver and has become a diverse and rather mature area of research with various application domains and versatile techniques. It is often inspired by cognitive sciences and the remarkable capabilities of animals and humans. By first focusing on what is possible in principle the field has so far not prioritized questions of efficiency and what can be accomplished within tight constraints of energy and physical limits. In contrast, cyber-physical systems consider resource constraints as first class citizens that shape and drive all design decisions. It has emerged at about the same time with its name being coined around 2006 by Helen Gill at the National Science Foundation. Although both research domains have already a rich literature and long history, only recently we have seen proposals to use concepts and techniques of self-aware computing in cyber-physical systems under tight constraints. The following articles give an overview of this emerging area and exemplify some of the application cases. In “Self-Aware Cyber-Physical Systems,” Bellman et al. give a comprehensive overview of all relevant topics, issues, challenges, and some potential solution. Their article can serve as introduction to the field and summary of the state of the art. Observing that CPSs often operate in groups and interact with other agents, some of them like and some of them very unlike themselves, Esterle and Brown, in “I Think Therefore You Are,” discuss interaction patterns and models that CPSs build up about their partners based on the interactions observed. The authors postulate that, what they call networked self-awareness is a solid
               
Click one of the above tabs to view related content.