Student team places second in inaugural SAE AI Mini-Challenge
A team of students led by Andrew Perrault, assistant professor in the Department of Computer Science and Engineering along with Qadeer Ahmed, research associate professor in the Department of Mechanical and Aerospace Engineering placed second out of 12 teams in the first ever SAE MobilityForward Challenge: AI Mini-Challenge.
According to SAE.org, The SAE MobilityForward Challenge is a student engineering design competition in the SAE Emerging Technology Series that provides an opportunity for students to think critically about current and future emerging technologies in the mobility engineering industry. Each program within the Series challenges universities to create a team of multi-disciplinary engineers to collaborate, design, present and defend conceptual designs to industry professionals.
The AI Mini-Challenge asked student teams to apply data science techniques to emerging technology issues relevant to the mobility industry in order to uncover how mobility could reflect, impact or help resolve social issues in a defined geographic area. The teams then presented their solutions via the Solutions Presentation and a virtual showcase booth on SAE’s Digital Platform.
The Ohio State team, made up of data analytics students Chenwei Xu and Hui Miao, geographic information science student, Rui Zhang and mathematics students, Yushan Qu and Ruijie Zhao, was given a dataset for Allegheny County in Pennsylvania and could create a problem of their choosing to solve. “After consulting with Professor Perrault, we decided to focus on Drive-by Sensing, researching how to find a cheap way to sense whole city air pollution,” said Xu. “We also wanted to focus on how minority groups would benefit from this idea after analyzing their living area in the entire county.”
The team organized itself into two groups, one is for modeling and one for visualization. The modeling team built work on how to optimize sensor placement across vehicles. The visualization team built multiple visualizations about the sensor area, shared driving distribution and minority groups distribution. They then organized the visualizations into one report which described their method for solving the problem of how to place the drive-by sensor into shared driving to have the maximum coverage.
“It’s exciting to see an opportunity for undergrads to analyze real data to see how it can inform socially impactful decisions,” said Perrault. “I’m impressed by what our team accomplished.”