How Drones and Augmented Reality Can Help Cities Map Air Pollution
Researchers are helping to tell urban stories through data visualization, flying sensors, and digital storytelling.
When mapping a city's air pollution, sensors are typically placed four feet or so above the ground, at the level where people inhale the air. But a new project recently used drones to collect urban pollution data by gathering detailed information on the vertical structures and environments of Shanghai's Tongji University. The project, initiated by Biayna Bogosian, a PhD student at the University of Southern California's Media Arts & Practice program, and Princeton University researcher Maider Llaguno-Munitxa, featured collaborations with assistants and students at Tongji. The hope is that these data-collecting drones, along with the creative data visualization video and augmented reality applications, will offer a new way for scientists and other people to learn about the air pollutants occupying urban spaces.
As Bogosian tells Creators, sensors that map urban micro-climates or micro-environments are typically fixed to a building, or traveling with a researcher who is either walking or biking through an environment. While bicycles outfitted with sensors were used for this project, the drones allowed Bogosian and her collaborators to collect vertical data from buildings and tall vegetation. Using photogrammetry, the drones were also used to create a 3D scan of the university. These 3D visuals were then fed into the Microsoft's HoloLens augmented reality headset worn by the film's female character, as a way of visualizing how she might learn about her environment.
"Maider and I both trained as architects, so there is this interest in the architectural scale, so we're interested in the details of buildings," says Bogosian. "We're interested in how knowing more about your immediate environment can help you make better decisions in terms of the design of buildings. But, ultimately we're fascinated with different visualization methods that can basically allow you to see these individual parameters like air quality."
With this video, Bogosian wanted to explore the story of two students with similar interests but different backgrounds. Both are trying to discover their environment and make sense of different parameters.
While the woman maps temperature, humidity, CO, CO2, nitrogen dioxide, ozone, and particle matter, the man maps surface temperature, surface types, and environmental textures. The two never interact, but continue on their own personal scientific journeys, together building a more complete picture of the Tongji micro-environment.
"We wanted to create a correlation between the environment and the data that we collected," Bogosian notes. "So, for instance, we were interested in looking at building typologies and vegetation on campus as well as surfaces like asphalt or stone, which absorb heat and trap pollution in different ways… But we were also interested in the different ways air pollution circulates in urban environments during day and during night."
The drone's vertical gradient allows Bogosian and Llaguno-Munitxa to better understand how specific type of buildings and environments repel or trap air pollution. To do this, they used drone swarms outfitted with Arduinos and various sensors to make the data collection as spatial and temporal as possible.
For Bogosian, Llaguno-Munitxa, and their collaborators at Tongji, the film is an exercise in in storytelling through data. It was important for them to think of how they could communicate this data to an average person, but also develop a spatio-temporal visualization technique that even a scientist could find useful.
"We're also interested in ways that we can come up with an efficient sensing module, and we would have different versions of this that perhaps someone could purchase and assemble for sensing," says Bogosian. "We're interested in the visualization aspect of nature, and this isn't really practiced in the field of urban sciences."
"The amount of spatio-temporal resolution you get from drones is unlike anything else we could achieve," she adds. "It's easier and cheaper to give people the sensors to do the data collection, but with drones we can program them to have a certain speed and be more consistent. So, spatio-temporal mapping is often complimentary to the field work of mapping we would be doing either by walking or using bicycles."
Click here to see more of Biayna Bogosian's architectural and interactive design work.