Direct Chemo-Mechanical Actuation

Not Available Clifton Labs, Inc. proposes to model and implement a framework for the construction of self-adaptive software systems. Specifically, a framework for the creation of self-improving software systems that are able to dynamically reconfigure themselves with the aim of improving performance or exploiting better functionality. To accomplish these goals, the framework promotes the use of on-line monitoring techniques and permits developers and end users to experiment with alternative program configurations, to play what if games with selected program attributes. The objective of this research is the modeling, development, and demonstration of the applicability of a workbench for constructing on-line configured software systems. Clifton Labs, Inc. proposes to model, develop, and evaluate the proposed workbench in an implementation independent manner. The Phase I objective are to (1) develop predictive and recovery based control polices for specific software systems (support for building new control techniques will also be provided), (2) automate evaluation of software functionality and performance, and (3) develop a workbench to tailor predictive and recovery based techniques to specific software systems for run-time configurability.