Application development using modeling and dynamical systems analysis

Abstract

Research on context aware systems is handicapped by the lack of readily available large scale data sets, as well as by the lack of tools by which researchers can interact effectively with such data sets across a range of scales and granularities. We show how virtual reality combined with a dynamical systems analysis approach can start to address these gaps. Simulation allows for generation of simulated sensor data at runtime and actuations of entities in the virtual world. The ease of sensor deployment and configurations in a simulated environment allows for rapid reconfigurations enabling generation of the required large scale data sets for analysis. Using these data sets, dynamical systems analysis can determine if a given application is functioning in a manner that is deemed to be correct.