SENSELET provides sensory-driven IoT network for scientific instruments

11/14/2018 4:32:30 PM Kim Gudeman

When fabricating new semiconductors for computer chips, the external environment – such as the humidity, temperature, or vibration from nearby machines – can make or break the experiment. That holds true for many types of research involving scientific instruments, including the discovery of new materials or biomedical processes.

A team of researchers at the University of Illinois at Urbana-Champaign is building a wireless and scalable sensory infrastructure that will help scientists better understand the conditions under which their experiments succeed or fail, with the idea of accelerating scientific discovery and the introduction of new technology to the marketplace. The National Science Foundation is funding the two-year, $500,000 project, titled “SENSELET: Sensory Network Infrastructure for Scientific Laboratory Environments."

“Our goal is to help scientists not have to repeat experiments so many times to find the winning formula,” said Klara Nahrstedt, principal investigator of the project, director of the Coordinated Science Laboratory and the Ralph and Catherine Fisher Professor of Computer Science at Illinois. “Ultimately, we want to speed up the discovery cycle for new innovations.”

The team plans to deploy a diverse wireless and sensory infrastructure that collects data and sends it to the cloud, where it will be correlated and synchronized with instrument data in real time and on demand. Along with providing valuable information to researchers, the technology also will help lab managers better understand conditions in the facilities they are in charge of maintaining.

“Lab environment data gives present status, information on changes and trends, and conditions during a process or measurement,” said Mark J. McCollum, principal research engineer at the Micro & Nanotechnology Laboratory. “Having this history and using it with 4CeeD (a data analysis and acquisition strategy) provides important meta data as well as a predictive path for management and operational needs in a physical device laboratory. It is exciting how SENSELET will make this conveniently available.

SENSELET builds on previous NSF-funded work, which includes the development of the 4CeeD Distributed Data Infrastructure, a Web-based distributed capture, curation, and coordination tool that bridges the network and computing gap between scientific instruments and the remote cloud infrastructure. As part of the work, researchers also developed BRACELET, an edge device infrastructure that enhances security and performance issues that arise when connecting old instruments with the modern 4CeeD cloud infrastructure. 4CeeD operates on a cloud-based smart data management system called Clowder, developed by the National Center for Supercomputing Applications at Illinois.

In addition to Nahrstedt, the UI research team includes Roy Campbell, the Sohaib and Sara Abbasi Professor of Computer Science; Kenton McHenry, research scientist at the NCSA: John Dallesasse, professor of Electrical and Computer Engineering, and Tracy Smith, director for research IT.