“Smart” windshield wipers that know when it’s raining have been common since the mid-1990s. But one pair of smart windshield wipers is still pretty dumb. Network all the windshield wipers on the road, however, and you have the power to predict the weather.
The Pilot Program
In 2014, the University of Michigan Transportation Research Institute conducted Safety Pilot, a connected vehicle test program in Ann Arbor, Michigan. At the time, it was the world’s largest connected car test. Some 3,000 participants were enrolled and their cars outfitted with GPS, wireless communications systems, and a black box data acquisition system. The objective of the study, conducted in conjunction with the United States Department of Transportation, was inherent in the name: improving safety.
While cars with the ability to talk with infrastructure and communicate with each other can improve vehicle and pedestrian safety in obvious ways, they also create a powerful network. Researchers are eager to tap it, developing secondary applications including one idea that was trialed in the Ann Arbor study: Using data from windshield wipers to predict and prevent flooding.
Going Beyond Safety
In a paper published in January, Michigan Engineering researchers Matthew Bartos, Hyongju Park, Tian Zhou, Branko Kerkez and Ramanarayan Vasudevan explain that they outfitted 70 of the Safety Pilot vehicles with dashboard cameras and sensors embedded in the windshield wipers. By monitoring this fleet, they were able to acquire faster and more accurate rainfall data than is currently available from radar and rain gauges.
“These vehicles offer us a way to get rainfall information at resolutions we’d not seen before,” said Branko Kerkez, assistant professor in the Department of Civil and Environmental Engineering. “It’s more precise than radar, and allows us fills gaps left by existing rain gauge networks.”
By combining weather radar with the connected car data stream, rainfall estimates are improved. This includes capturing localized rainfall that conventional measuring methods might miss, as shown in the maps above. Such rain can lead to flash flooding, which commonly causes sewage overflows in urban areas. And with better abilities to make rainfall predictions, municipalities can better take preventive actions against flooding.
Not coincidentally, Ann Arbor and the surrounding areas of Southeastern Michigan see heavy rainfall during much of the year and have a history of such environmental disasters.
“Because of the sparseness of radar and rain gauge data, we don’t have enough information about where rain is occurring or when it’s occurring to reduce the consequences of flooding,” said Ram Vasudevan, a U-M assistant professor of mechanical engineering. “If you have fine-grain predictions of where flooding occurs, you can control water networks efficiently and effectively to prevent all sorts of dangerous chemicals from appearing inside our water supply due to runoff.”
“One day, when everything is connected”
The researchers imagine an improved version of their procedure could benefit from optical rain sensors. These are already being installed in many new cars and would enable direct measurement of rain intensity. In the near future a network of connected cars could even provide the sensors and data stream for an autonomous stormwater system. Such “smart” infrastructure would operate its own gates, valves and pumps. With scientific data pointing to recent increases in flooding, these types of systems may be even more useful. And rain sensing is just a start.
“One day, when everything is connected, we’re going to see the benefits of this data collection at a system scale,” Kerkez said. “Right now, we’ve made connections between cars and water, but there will surely be more examples of data sharing between interconnected infrastructure systems.”