Each node in the Array of Things was equipped with an Nvidia Graphics Processing Unit (GPU), which performed image computations in the field and sent only the processed data to the network (a form of edge computing). As an added privacy protection, nodes are designed to be installed temporarily. “I don’t want to see edge computing all over the city, with cameras analyzing your behavior wherever you walk,” he says. “For me, it’s more dystopian than I’d like to see it. But I think there’s room for diagnostics in these edge devices.
Between 2016 and 2019, the team installed 140 AoT nodes on Chicago streetlights. In a participatory process, Argonne and a team from a local university worked with Chicago citizens and city departments to decide where to place the sensors.
Basket/URBANCCD
Since then, dozens of studies have used sensor data. Nodes are used to assess railroad crossing safety, monitor pedestrian crossing usage, and detect flooding along the Chicago River. A collaborator on the project, Kathleen Cagney, who directs the Institute for Social Research at the University of Michigan, used environmental data from the sensors in public health studies, finding that where the sensors detected more air pollution, It was found that the asthma rate was high.
Since then, Catlett’s team has been working on low-tech projects. For example, last year he and his colleagues partnered with his Microsoft Research to install 115 of his low-cost, solar-powered air sensors at bus stops around the city. The resulting data showed pollution hotspots near Chicago’s south and west industrial corridors in unprecedented high resolution. Environmental and community groups are now pressing the city to change its policies. The team plans to expand to thousands of air quality nodes over the next few years.
Array of Things is expanding beyond Chicago through a project called SAGE. Unlike other urban sensing systems that tend to be proprietary, SAGE allows anyone to write software for the nodes, including high-resolution hyperspectral cameras, lidar, and audio recorders.
Catlett said the team is now in the deployment phase. By the end of the year, Chicago will have 50 of its $10,000 nodes, replacing his Array of Things nodes from the previous generation. Dozens have already been deployed across Southern California to detect wildfires, and in towers across the country to analyze weather and climate change. The National Science Foundation wants a total of 80, one for each tower in the National Ecological Observatory Network. Oregon wants 100 to help detect earthquakes. It was ordered by the Australian scientific agency CSIRO. The ever-growing library of open-source applications available on GitHub includes programs that identify birds by their calls and classify funnel clouds from images.
The “urban fitness tracker” has become a global phenomenon just in time to study our changing world.
Christian Elliott is a freelance science journalist based in Chicago, Illinois.
.
Comments
Post a Comment