Tackling New Frontiers in Modeling and Analysis of Cyber-Physical Systems

With the advent of smart sensors and actuators, computing units are interacting with the physical environment at a much larger scale than ever before, through feedback and controlled actuation. Such cyber-physical systems (CPSes) are by definition mission critical and have strict requirements on safety, security, sustainability, reliability, and long term sustainable operation. Guaranteeing these requirements needs accurate understandings of the large scale complex interactions of the computing units with the physical environment and an inter-disciplinary approach towards problem solving in several domains such as control systems, hybrid systems, and operating systems. Common assumptions of linearity, time invariance, dimensionality, non-emergent behavior are no longer applicable. This paper presents some outcomes of NSF funded CPS projects to illustrate inapplicability of existing mathematical tools to characterize CPS behavior with desired accuracy, and new frontiers in theoretical and practical computer science that have to be overcome in order to design CPSes with the desired guarantees on requirements.