Spatio-temporal hybrid automata for safe cyber-physical systems

Abstract

Interactions between the computing units and the physical environment in Cyber-Physical Systems (CPSes) are considered to verify safety properties, i.e. ensuring the un-intentional side-effects of cyber-physical interactions are within desired limits. A Linear 1 space dimension Spatio-Temporal Hybrid Automata (L1STHA) is defined to capture the effects of the interactions, in both time and space. Aggregate effects of interactions due to concurrent operations in the computing entities are expressed as a set of interdependent partial differential equations associated with dedicated modes of the L1STHA model. A time and space bound L1STHA reachability analysis algorithm is proposed for safety verification, which provides reachable states of the L1STHA with an arbitrary accuracy . The runtime of the algorithm depends on the requested accuracy. The usage of the L1STHA modeling and analysis is demonstrated for medical CPSes such as infusion pumps.

Towards the verification of cyber-physical processes based on time and physical properties

“…Using temporal formal formalisms, the designers consider time-related properties, and they can specify the collaborative behaviour. In other studied approaches (Shao et al, 2013;Banerjee and Gupta, 2013;Li et al, 2014;Shao et al, 2013;Krishna and Trivedi, 2015), the authors use hybrid automata and their extended versions to model CPS. However, in these papers, the researchers do not support the physical properties that can be associated with a physical behaviour, and they do not study the impact of these properties on the timing behaviour of the collaborative processes.…”

Internet of Medical Things: A Review of Recent Contributions Dealing With Cyber-Physical Systems in Medicine

“…Our selection represents the whole range of modeling that has to be realized when designing cyberphysical systems. Indeed, papers focused on the development of models describing physiological (and thus continuous time) processes, such as the modeling of cardiac activity [13], [14]; models of the blood glucose rate for artificial pancreas design [12]; or the development of models describing the human body's absorption rate of a given medicine [47], [48]. Figure 8 exhibits a detailed example about how authors of [14] have used finite state automata to model the electrical behavior of the heart, intended for cardiac fault detection in implantable pacemakers.…”

Cyber physical systems for predictive production systems

“…In energy, Stanovich et al developed a testbed for smart-grid [13] while Taneja has discussed the challenges of using CPS in smart-grid [14]. In healthcare, Banerjee et al proposed a linear 1-space dimensional spatio-temporal hybrid automata that can capture the spatio-temporal aggregate cyber-physical interactions [15]. Wang et al proposed a human-centered cyber-physical systematic approach for post-stroke monitoring [16].…”