National Science Foundation invests $5.1M into Mcity for autonomous vehicle simulations

Breana Noble
The Detroit News

The National Science Foundation is investing $5.1 million into Mcity, the public-private mobility research partnership at the University of Michigan, to enhance testing for autonomous vehicles.

The funds from the government agency charged with advancing science will add virtual reality software and the ability to generate real-world datasets to the Ann Arbor facility. This "Mcity 2.0" will allow academic researchers and public users to create simulations tailored to scenarios, helping to address challenges in autonomy from insufficient access to real-world data and testing facilities.

“Testing these vehicle technologies in real-world scenarios is an essential step for transferring innovations to businesses, communities, and drivers," Susan Margulies, assistant director for engineering at the National Science Foundation, said in a statement.

Researchers can send their algorithms and programs to Mcity, plug them into the facility’s operating system, request specific conditions for testing and participate remotely as those scenarios play out in Mcity's combined real and virtual setting.

“The new digital infrastructure combining real-world data sets with high-quality simulation capabilities and a physical test track will set Mcity apart from other AV test facilities, and enable remote use,” Henry Liu, a professor of civil and environmental engineering at UM and director of Mcity, said in a statement.

“The proliferation of AVs and connected vehicles will create a safer travel environment, and by making this next generation version of Mcity available to a wider range of researchers, we believe we can help accelerate adoption.”

Mcity works with partners to collect data from intersections in Ann Arbor and Detroit for its simulations and has plans to equip more crossroads. These locations have sensors to capture anything that moves from cars to pedestrians. The technology then categorizes the road user and identifies its speed and direction.

It can send the information instantaneously to connected vehicles in the area. That can trigger onboard warnings when cars are facing dangerous conditions. The largest projected involved nearly 3,000 connected vehicles, and the research has suggested the technology can help to reduce crashes by 80%, according to the University of Michigan Transportation Research Institute.

Earlier this year, UM completed a $1.5 million upgrade to a building on the Mcity Test Facility site that will be used to monitor testing.

bnoble@detroitnews.com

Twitter: @BreanaCNoble