Networked Cyber Physical Systems at SRI

 

 

 

http://ncps.csl.sri.com/image/side_icon.gif

Home

http://ncps.csl.sri.com/image/side_icon.gif

Framework

http://ncps.csl.sri.com/image/side_icon.gif

Publications

http://ncps.csl.sri.com/image/side_icon.gif

People

http://ncps.csl.sri.com/image/side_icon.gif

Download

http://ncps.csl.sri.com/image/side_icon.gif

Case Study1

Case Study2

http://ncps.csl.sri.com/image/side_icon.gif

Reading List

http://ncps.csl.sri.com/image/side_icon.gif

Sponsors

 

In an NCPS, information about the physical world (e.g., wireless signal strength, sensor readings) is usually acquired via physical devices, which can be geographically scattered.  Hence, the management of an NCPS entails the integration between a physical layer, which includes the underlying physical devices as well as the network, and a higher layer, which includes the knowledge dissemination layer and the application layer that is designed to achieve global objectives

 

 

To deal with uncertainty, complexity, and resource constraints in both communication and computation, the cyber-framework is concerned with decisions at each layer. For example, assume an application operating a networked team of mobile robots, i.e., mobile computational resources with sensors and actuators.  At the application layer, it needs to decide when to perform a local action (e.g., move forward, sense the temperature of a certain area) to achieve local goals, which are typically obtained by decomposition from higher-level goals. Local computations are also needed for tasks such as information fusion, based on which new goals may be triggered. In [1][3], we presented a declarative approach to avoid low-level programming by combining distributed reasoning and asynchronous control.  In [2], we presented a case study of a dynamic distributed optimization and control problem that is not solved using a distributed logic but directly built on top of the cyber-framework.

 

At the knowledge dissemination layer, a specific strategy (e.g., deterministic flooding) is chosen to propagate knowledge on top of the underlying physical network layer.  The global objective needs coordination among the nodes.  Knowledge about local solutions needs to be disseminated through the network to improve overall solution quality. What kind of knowledge and how often the knowledge should be disseminated depends on the specific application and its objective. 

 

1.     M-O. Stehr, M. Kim, C. Talcott

Toward Distributed Declarative Control of Networked Cyber-Physical Systems

            7th International Conference on Ubiquitous Intelligence and Computing (UIC'10)

            Oct. 2010, Xian, China.

 

2.     M. Kim, M-O. Stehr, J. Kim, S. Ha,

            An Application Framework for Loosely Coupled Networked Cyber Physical Systems

            8th IEEE/IFIP Conference on Embedded and Ubiquitous Computing (EUC'10)

            Dec. 2010, Hong Kong, China.

 

3.     M. Kim, M-O. Stehr, C. Talcott

A Distributed Logic for Networked Cyber-Physical Systems

            IPM International Conference on Fundamentals of Software Engineering (FSEN'11),

            Apr. 2011, Teheran, Iran.

 

 

 

 

 

Last updated: July 12, 2012