A Cyber Sensor Model for Cyber-Physical-Social Systems

Introduction

Cyber-Physical Systems (CPSs) (Delicato et al., 2020; Liu et al., 2017) envisage a tight coupling between computing processes and the physical world. Engineering efficient, robust and secure CPSs pose many challenges; minor faults in design may result in catastrophic and tragic events in the physical world, as documented experiences with autonomous vehicles testify. Nonetheless, CPS are an indispensable and often invisible enabler of many services in the modern world. As the complexity of CPSs invariably increases, a need for more sophisticated models, methodologies and tools will be needed (Lee, 2015). For example, Intelligent CPSs (iCPSs) advocate embedding artificial intelligence techniques in the feedback loops for enabling decision-making in next-generation CPS (Vijayakumar et al., 2019). A complementary approach, that of Digital Twins (Rasheed et al., 2020), advocates constructing replicas of physical processes; in this way, more sophisticated models of reasoning and learning may emerge.

CPSs may leverage pre-existing cyberinfrastructures by incorporating them into their technology stack. The Sensor Web (Zhang et al., 2018) is an exemplar of such infrastructure; it enables programmatic access to networks of heterogeneous sensors for use in system development, sometimes in systems with global reach. Thus, when building CPSs, software engineers can call on an array of tried and trusted technologies such as Wireless Sensor Networks (WSNs), middleware, and Internet of Things (IoTs) along with a range of mature standards, for example, those of the Open Geospatial Consortium (OGC). The net result is that the CPSs concept has been demonstrated in a wide range of domains including environmental monitoring (Mois et al., 2016), landslide detection (Liu et al., 2019), smart agriculture (Liu et al., 2020), health (Ashutosh Sharma et al., 2019) and even space technologies (Akyildiz & Kak, 2019).

Cyber-Physical-Social Systems (CPSSs) are, in contrast to their CPSs peers, less-well developed. Such systems offer an intriguing vision – the intrinsic incorporation of the human dimension in a variety of facets into system operation. Such incorporation may be relatively passive, for example, data or information provision, or may involve active human-in-the-loop participation that enables human-automation symbiosis as envisaged by Industry 4.0 (Rauch et al., 2020). Just as CPSs may leverage the Sensor Web, a CPSS may likewise leverage the Social Web (Russell & Klassen, 2018). In practice, challenges with data collection arise, especially if a big data solution is envisaged. However, even within relatively constrained circumstances, difficulties arising from a lack of software engineering toolkits and the inherent complexity. This paper demonstrates that in viewing CPSSs as an extension of CPSs, rather than a singular paradigm in its own right, core technologies that underpin CPSs may be refactored and augmented to form an intuitive foundation for either or both paradigms.

This paper is structured as follows. A review of the state-of-the-art is presented in the next section. A model for a CPSS incorporating social web precepts is then proposed. An implementation of this model is then validated using a cohort of software engineers. After discussing the results of the evaluation, the paper is concluded.