World Watchers

The National Science Foundation has awarded a multi-institutional team led by Northwestern University a $9 million grant to launch the Sage project, a novel cyberinfrastructure created to exploit dramatic improvements in artificial intelligence technology. The goal: to build a continent-spanning network of smart sensors.

With Sage, advanced machine learning algorithms will be moved to “the edge.” Edge computing is a way to streamline data flowing from Internet of Things devices by providing data analysis very near the site where the data is gathered. By linking small, powerful, computers directly to high-resolution cameras, air quality and weather sensors, and experimental Light Detection and Ranging (LIDAR) systems, this new distributed infrastructure will enable researchers to analyze and respond to data almost instantly. From early detection of wildfire smoke plumes in California to identifying ultrasonic calls of bats or the patterns of pedestrians in a busy crosswalk, Sage’s artificial intelligence-enabled sensors will give scientists a new tool to understand our planet.

Northwestern-Argonne Institute of Science and Engineering Codirector Pete Beckman believes that these distributed, intelligent sensor networks will prove essential for understanding the impacts of global urbanization, natural disasters, such as flooding and wildfires, and climate change on natural ecosystems and city infrastructure. However, many of today’s distributed systems struggle with the immense volume of streaming data that high-fidelity sensors can provide. Some systems resort to saving the data on hard drives that a technician traveling to the instrument might retrieve just a few times a year. Other systems manage the data deluge by only collecting a small portion of the valuable input and uploading it to a cloud server for analysis via a slow wireless link. To address this challenge, Sage will embed computers directly into the sensor network and rely on advancements in edge computing to analyze the torrent of sensor data as it streams past.

Sage’s name is inspired by nature: the perennial mint plant. But it also conveys other connotations associated with the word, such as wisdom and good judgment. The new cyberinfrastructure project will be enhanced by a set of partnerships with existing scientific instruments that span contexts from urban (NSF-funded Array of Things) to continental-scale (NSF’s National Ecological Observatory Network, or NEON), and with regional instruments focused on understanding and responding to emergencies, such as severe storms (Atmospheric Radiation Measurement, or ARM) and wildfires (High-Performance Wireless Research and Education Network). Sage will be able to integrate measurements from these multiple modalities.

Sage test nodes will be deployed in environmental locations in California, Colorado, and Kansas and in urban settings in Illinois and Texas. The project will build on the open source technology platform used in the Array of Things project, which has deployed more than 100 sensors with edge computing capabilities within Chicago.

“With the development of a new intelligent, or ‘software-defined’ sensor network, researchers will be able to better collect and analyze data that is essential for understanding the impacts of wildfires, global urbanization, and climate change,” says Beckman, a globally recognized expert in high-end computing systems and principal investigator of the Sage project.

Other NAISE members involved in the project include Jennifer Dunn, who is part of the project management team; Nicola Ferrier, who is deputy director of the project; Rajesh Sankaran, who is leading the Sage node development; and Scott Collis, who is leading the atmospheric science work. Several Northwestern doctoral, graduate, and undergraduate students have also played a role in the project, as has Aaron Packman, civil and environmental engineering, who is on the advisory committee.

Reusable Cyberinfrastructure

The research team that includes the University of Chicago, George Mason University, the University of California San Diego, Northern Illinois University, the University of Utah, the Lincoln Park Zoo, and collaborators at Argonne National Laboratory will design and build reusable software components and cyberinfrastructure services to support the new scientific measurement functions. These measures will range from in-situ analysis of high-bandwidth sensor data streams to adaptive system behaviors, such as adjusting the sampling rates and directional settings of LIDAR instruments or high-resolution cameras to capture events in more detail.

“Particularly with remote instruments, such as those on NEON towers with low-bandwidth network connectivity, the traditional method of reducing data volume works well for steady state conditions,” says Gene Kelly, a professor from Colorado State University and lead scientist at NEON. “But Sage will give us the ability to detect ecosystem changes of interest at continental scale, allowing us to apply more sophisticated analytics in the field, capturing much greater detail on these changes.”

“We’re designing a system with software that can be programmed to monitor everything from wildfires to wolf calls, and from urban traffic flows to rural lightning strikes,” says Ilkay Altintaş, chief data science officer at the San Diego Supercomputer Center at UC San Diego and director of the WIFIRE Laboratory that uses computer simulations and integrated data streams to develop new cyberinfrastructure to support fire science. “With Sage, mountaintop networking towers in Southern California will be able to analyze camera data for natural hazards and even autonomously redirect cameras to zoom in on regions of interest.”

The reusable cyberinfrastructure running on these Sage testbeds will give climate, traffic, and ecosystem scientists new data for building models to study these coupled systems, says Beckman. The software and hardware components developed in Sage will be open-source and provide an open architecture to enable scientists from a range of fields to build their own intelligent sensor networks. The Waggle edge computing platform, developed at Argonne, will be the starting point for the intelligent nodes deployed across the country. Also led by Beckman, Waggle was launched in 2013 to permit a new breed of sensor-driven environmental science and smart city research.

The Sage team will also extend the current educational curriculum used in Chicago and will inspire young people — with an emphasis on women and underrepresented populations — to pursue science, technology, and mathematics careers by providing a platform for students to explore measurement-based science questions related to the natural and built environments.

“It’s perfect timing for this project because of the capabilities of machine learning and artificial intelligence that didn’t exist a decade ago,” says Beckman. “We have flung open the door to scientists who are now tasked with writing codes that can measure a limitless number of urban and environmental elements.”

NAISE is part of Northwestern’s robust ecosystem of University Research Institutes and Centers, a collection of more than 40 interdisciplinary knowledge hubs that harness talent from across all areas of the institution.

By Roger Anderson