Effective response and adaptation to the physical world, and rigorous management of such behaviors through programmable computational means, are mandatory features of cyber physical systems (CPS). However, achieving such capabilities across diverse application requirements surpasses the current state of the art in system platforms and tools. Current computational platforms and tools often treat physical properties individually and in isolation from other cyber and physical attributes. They do not adequately support the expression, integration, and enforcement of system properties that span cyber and physical domains. This results in inefficient use of both cyber and physical resources, and in lower system effectiveness overall.
This work investigates novel approaches to these important problems, based on modularizing and integrating diverse cyber-physical concerns that cross-cut physical, hardware, instruction set, kernel, library, and application abstractions. The three major thrusts of this research are 1) establishing foundational models for expressing, analyzing, enforcing, and measuring different conjoined cyber-physical properties, 2) conducting a fundamental re-examination of system development tools and platforms to identify how different application concerns that cut across them can be modularized as cyber-physical system aspects, and 3) developing prototype demonstrations of our results to evaluate further those advances in the state of the art in aspect-oriented techniques for CPS, to help assess the feasibility of broader application of the approach. The broader impact of this work will be through dissemination of academic papers, and open platforms and tools that afford unprecedented integration of cyber-physical properties.
Project Overview: Effective response and adaptation to the physical world, and rigorous management of such behaviors, are mandatory features of cyber-physical systems (CPS). However, achieving such capabilities across diverse application requirements surpasses the current state of the art in system platforms and tools. Existing systems do not support the expression, integration, and enforcement of such properties that span cyber and physical domains. In this work, we examined mechanisms to enable conjoining of cyber-physical properties within a system through: 1) plastic data structures, 2) tightly coupling SW/HW resources, and 3) integrating system implementation artifacts with control theory. Outcomes: 1) With respect to tightly coupling SW/HW resources. Two outcomes were: a) the development of a hardware co-processor to off-load real-time scheduling from software, and b) a cache was developed that was aware of how "critical" a given task was to give highly critical tasks preferential treatment over lower critical tasks. 2) With respect to integrating system implementation artifacts with control theory. A plant-on-chip (PoC) approach was developed to allow the model of the physical plant being controlled to be deployed on the same chip as the software. Thus, allowing control algorithms to be evaluated on their final computing platform to give early feedback on how inherent artifacts (e.g. bus delays, cache miss rates, etc.) of the actual computing platform will impact the stability of the controller.
