Yiqi Zhang, assistant professor of industrial and manufacturing engineering at Penn State, aims to make the roads a safer place for all drivers by better understanding the interactions between human drivers and autonomous vehicle (AV) technologies.
Zhang received a grant from the National Science Foundation Directorate for Computer and Information Science and Engineering (CISE) Research Initiation Initiative (CRII) program to research the individual differences between human and AV interactions, especially in mixed transportation systems. The NSF program specifically recognizes early-career scientists by funding research in the first five years after completing a doctorate.
According to the National Highway Traffic Safety Administration, 94 percent of fatalities are due to human error. It is predicted that AV technology could significantly decrease the number of accidents each year, according to Zhang.
“Autonomous vehicles are expected to grow in number over the coming years and dominate our future transportation systems,” Zhang said. “However, the public isn’t ready yet. AAA’s 2018 Vehicle Technology Survey of more than 1,000 participants indicated that around 70 percent of drivers would be afraid to ride in autonomous vehicles.”
The goal of Zhang’s study is to quantify the effects of the designed AV driving styles on drivers’ trust and decision-making under the synthesized consideration of drivers’ own driving styles. Zhang and her team will observe driver behavior in two simulated environments: while a human is riding in an AV (human-AV) and while a human is driving alongside an AV (HV-AV).
Zhang stressed that a driver’s individual behaviors can impact their experience with AV technology. For example, a careful driver, being unsure of how well the AV will obey the law, may not trust the AV well; or an aggressive driver, being unsatisfied with the conservative nature of AV driving, may give up the AV technology.
The study will be conducted in Zhang’s Human-Technology Interaction Laboratory, where researchers work to address fundamental issues of human behavior in order to promote our understanding of human cognition and performance. Zhang aims to develop computational models of driver behavior and apply these models to design intelligent systems and user interfaces to improve human performance and safety.
“It may help us to explore ways to understand and promote drivers’ trust of AVs in mixed traffic conditions; mitigate driver aggressiveness to improve safety; address the public acceptance of autonomous vehicles; and smooth the transition into a future where autonomous driving is prevalent,” Zhang said.
According to Zhang, most of the current research associated with AVs recognizes trust as the essential factor influencing drivers’ acceptance of automation technology, as well as a key determinate in understanding how to promote successful interactions between human and AVs.
“This study will fill an important gap between literature and expectations of cyber transportation systems in the next few decades,” Zhang said. “In particular, this work will address the potential offsetting of driver behavior induced by the HV-AV interaction in mixed transportation systems.”
Zhang plans to use the results to propose design guidelines for driving styles of AVs to improve driver trust and acceptance of AV technology, ultimately creating safer roads.