Spring 2022 Series

Student Presenter: Dr. Xiao Shi, Research Fellow at the National Institute of Health.

Student Advisor: Professor Anne Moudon, Department of Urban Design and Planning at University of Washington.

Original Air Date: March 31, 2022

Description

Childhood obesity has become a growing concern in the United States, with one in five school-age children being overweight or obese. Walking to or from school provides children with an opportunity to engage in physical activity and establish habits for leading an active lifestyle in their adult life. Factors influencing children’s ability to walk can differ from those that apply to adults, yet few studies have explicitly considered how to measure school walkability related to children walking to and from school. In collaboration with WSDOT, we constructed a walkability index based on built environment factors known to influence the walking behavior of children and validated the index using a uniquely large student travel survey from Washington State. Using the index, we predicted the percentage of students walking to school for all K-8 schools in Washington State. Because data availability differs by counties and communities, both indices were tested under different data availability scenarios. The developed indices filled in a significant data gap in school environment research and program development regarding active travel to schools. The index building process has the potential to be applied to other states interested in benchmarking school walkability.

Student Presenter: Dr. Shuo Feng, Assistant Research Scientist at the University of Michigan.

Original Air Date: May 16th, 2022

Description

Safety assessment is critical to the development and deployment of autonomous vehicles (AVs). The prevailing approach tests AVs in the naturalistic driving environment. However, due to the high dimensionality of the environment and the rareness of safety-critical events, hundreds of millions of miles would be required to demonstrate the safety performance of AVs, which is severely inefficient. To address this challenge, we have developed a scenario generation toolbox, which includes the augmented reality (AR) testing platform and the naturalistic and adversarial driving environment (NADE). With AR, a real AV can be tested at a test track with interaction from the virtual traffic flow. With NADE, the maneuvers of virtual background vehicles will be controlled intelligently, in that most scenarios are generated from naturalistic driving data, and only at selected moments, adversarial scenarios are generated to challenge the AV under test. The theory behind NADE ensures both the unbiasedness and the efficiency of the safety assessment. With this toolbox, every testing mile at test tracks can be converted into thousands of equivalent miles on public roads, which can significantly reduce the development costs and shorten the development cycle.