Smokey Bear is mostly right — you can help prevent wildfires — but it’s not all up to you. Use fireproof building materials and avoid vegetationignition zoneand avoiding fire-related activities when it’s hot, dry and windy is something that ideally we can all do.
The scientific community also has a precautionary role in avoiding natural disasters, urbanization and climate change. Recent advances in this area are the result of ambitious projects to build continent-wide networks of intelligent sensors that monitor environmental changes, such as attempting to predict the behavior of fires before they get out of control.
Partnered with the University of Utah, the Sage Project is a $9 million National Science Foundation (NSF)-funded initiative launched in 2019 by researchers at the Northwestern Argonne Institute of Science and Engineering (NAISE). being led. Between Northwestern University and the U.S. Department of Energy’s Argonne National Laboratory. Other Sage partners include the University of Colorado, University of California, San Diego, Northern Illinois University, and George Mason University.
“Sage is a next-generation software infrastructure flexible enough to handle both urban and environmental monitoring. Chair of the National Science Board (NSB), the policy-making body of the NSF. chief architect.
Meanwhile, Dr. Manish Parashar, director of U’s Scientific Computing and Imaging (SCI) Institute, is working to code scheduling and analysis functions on three sensor nodes managed by U, one at Taft-Nicholson. is supervising. Center for Environmental Humanities in Lakeview, Montana. One atop the 102 Tower in downtown Salt Lake City. The other is on the roof of the Rio Tinto Center, which houses the Utah Natural History Museum (NHMU).
When Reed presented the project to the university’s council of deans, NHMU executive director Jason Cryan said, “We immediately raised our hand to offer NHMU as the first installation site in Utah.” rice field.
The idea behind Sage is to move advanced machine learning algorithms to “edge computing”. The traditional method is to place sensors and collect data later. A few times a year, our previous system saved the data from the sensors to a hard drive or uploaded only a portion of the data to our cloud server over a slow wireless connection. Edge computing, on the other hand, uses equipment near or at the data collection site to analyze and measure data almost immediately. According to Reed, Sage devices process images, sounds, vibrations, and other data to create measurements that aren’t easily obtained from traditional sensor networks.
The Sage project draws heavily on lessons learned from the Array of Things (AoT) project, part of the smart city initiative announced by former President Barack Obama in 2015. AoT leverages an open-source intelligent sensing and edge computing platform called Waggle, developed at Argonne National Laboratory.
Funded primarily by the NSF, AoT was a collaborative effort among scientists, universities, federal and local governments, industry partners, and communities to collect real-time data on urban environments. AoT is the brainchild of Charlie Catlett, a senior his computer scientist in Argonne’s Mathematics and Computer Science (MCS) department, a co-principal investigator at Sage, and a longtime friend and colleague of Reid. did.
Catlett’s vision was to create an “urban fitness tracker.” A vast network of low-cost sensors deployed throughout Chicago can measure everything from urban heat islands to noise pollution.
“Given the way we’ve traditionally done the social sciences, people go out and do research. When you research people, sometimes they say what you want to hear.” Reed said, “It’s a bit like a doctor advising you to eat a balanced diet and exercise more. With AoT, instead of surveys, you measure what’s really going on.” Why don’t you try it?”
Reed said getting AoT projects off the ground requires a fair amount of public debate about acceptable uses of data.
“One of the rules agreed upon was that the raw image never left the camera, leaving only derived and anonymized statistics,” says Reed. “Another realization was that the amount of data was too large to push back to the central site. rice field.”
Sage’s cyber infrastructure connects small, powerful computers directly to nodes. Most of the nodes are equipped with high-definition cameras (including thermal cameras), microphones, weather measurements, air quality sensors, and send information back to a central server. Distributed systems allow researchers to quickly analyze and act on vast amounts of data without having to move it all back to the lab.
Sage devices are intended to measure local and regional environmental changes, but sensors are often sensitive enough to detect changes thousands of miles away. For example, Reed said Sage’s sensors detected pressure waves from a volcanic eruption in Tonga that caused the largest atmospheric explosion on record.
In addition to the U-maintained nodes, Sage technology will continue to be tested in the AoT environment. His WIFIRE project at the University of California, San Diego provides real-time wildfire prevention and response data on nearly 80 towers in Southern California. NSF’s National Ecological Observatory Network (NEON) is an array of hundreds of terrestrial and aquatic measurement sites across the United States that collect data on plants, animals, soil, water, and atmosphere.
“NEON will provide expert ecological data from sites across the continent to enhance the most important science being done today,” Reid said. “NEON is essential for tracking and understanding how human activities affect local flora and fauna as the environment changes.”
Sage sensor node overview (Source: Dan Reed)
- Each node runs on a wireless or wired Ethernet connection or Starlink, the Taft Nicholson Center’s highest bandwidth option.
- Nodes periodically send data back to the Waggle platform and other Argonne hosted infrastructure.
- A real-time information dashboard displays still images recorded by the upper and lower cameras, infrared views, 30-second audio snippets, and environmental conditions such as pressure, temperature, and humidity.
- The sensors are managed by an AI engine powered by a Raspberry Pi microcontroller and an Nvidia graphics processing unit (GPU) with plug-in connectors that can support additional sensors.
- You can schedule code to run on a node.
- The nodes perform a myriad of environmental monitoring tasks, from detecting changes in air quality and weather patterns to recognizing different species of birds and their migratory habits.
Reed is also excited about Sage’s potential to “bring K-12 students into real citizen science.”
“There’s a place in the node where young people can plug in their own low-cost sensors,” said Reed. “We are excited about the idea that students can interact with this technology and not just read examples in textbooks, but do real science, write code, build sensors, and get data in real time. Please try to imagine.”
