We describe CoSMo, a Cognitively Inspired Service and Model Architecture for situational awareness and monitoring of vehicular traffic in urban transportation systems using a network of wireless sensors. The system architecture combines (i) a cognitively inspired internal representation for analyzing and answering queries concerning the observed system and (ii) a service oriented architecture that facilitates interaction among individual modules, of the internal representation, the observed system and the user. The cognitively inspired model architecture allows one to effectively respond to deductive as well as inductive queries by combining simulation based dynamic models with traditional relational databases. On the other hand the service oriented design of interaction allows one to build flexible, extensible and scalable systems that can be deployed in practical settings. To illustrate our concepts and the novel features of our architecture, we have recently completed a prototype implementation of CoSMo. The prototype illustrates advantages of our approach over other traditional approaches for designing scalable software for situational awareness in large complex systems. The basic architecture and its prototype implementation are generic and can be applied for monitoring other complex systems. CoSMo's architecture has a number of features that distinguish cognitive systems. This includes: dynamic internal models of the observed system, inductive and deductive learning and reasoning, perception, memory and adaptation.
This thesis describes the service oriented model and the associated prototype implementation. Two important contributions of this thesis include the following:
The Generic Service Architecture - CoSMo's service architecture is generic and can be applied to many other application domains without much change in underlying infrastructure.
Integration of emerging web technologies - Use of Web Services, UPnP, UDDI and many other emerging technologies have taken CoSMo beyond a prototype implementation and towards a real production system. / Master of Science
| Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/33624 |
| Date | 02 October 2006 |
| Creators | Tupe, Sameer Vijay |
| Contributors | Computer Science, Marathe, Madhav Vishnu, Barrett, Christopher L., Ribbens, Calvin J., Bisset, Keith |
| Publisher | Virginia Tech |
| Source Sets | Virginia Tech Theses and Dissertation |
| Detected Language | English |
| Type | Thesis |
| Format | application/pdf |
| Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
| Relation | SameerTupeFinalThesis.pdf |
Page generated in 0.0092 seconds