Collaborative Research: HCC: Small: Toolkits for Creating Interaction-powered Energy-aware Computing Systems

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

The explosion in the number of “smart” computing devices has led to the need to design power solutions that reduce their ecological footprint and electronic waste. One possible answer is “self-sustaining” systems that harvest power from electromagnetic, solar, and other sources, reducing the need for batteries and external charging. This project looks at a novel kind of self-sustaining power source, one where systems draw energy from how people use them, for instance through button clicks or motion of the device. These user interactions are a promising source of power, but require careful management on the part of device designers since both the timing and types of interactions can be unpredictable and must be designed primarily to meet users’ needs rather than power goals. The research team seeks to deepen the understanding of information and resource needs of novice designers of self-sustaining systems and to create tools that help developers manage those needs. This in turn will improve sustainability in the design of computing systems, with the goal of developing power sources and development methods that generalize to a wide set of device design contexts. The project is structured around three research thrusts. The first is to better understand user needs, through a combination of observations of novices and experts programming self-sustaining systems, interviews, and surveys. The second is to create modules to harvest energy from interactions, providing developers with profiling tools and composable mechanisms enabled by a generic backbone system. The third is to develop toolkits that link the first two thrusts: supporting the development, deployment, and assessment of self-sustaining computing systems by providing guidance to developers on harvester selection, mechanism design, energy profiling, debugging, and beyond. These three research thrusts will be complemented by a comprehensive evaluation with a series of technical validations and user studies. This project will adopt common qualitative and quantitative evaluation techniques including A/B tests, Likert-scale questionnaires, semi-structured interviews, experimenters’ observations, walk-through demonstrations, evaluation through demonstrations, and pair programming. Overall, this research effort will guide the creation of tools to support future developments of self-sustaining computing systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.