2025-06-11 12:00:00 2025-06-11 13:00:00 America/Indiana/Indianapolis Summer 2025 Seminar Series Finding Preferred Automated Vehicle Driving Style via Exposure to Various Types of Vehicle Behavior Myeongkyu Lee, Ph.D. Student GRIS 134
Summer 2025 Seminar Series
Finding Preferred Automated Vehicle Driving Style via Exposure to Various Types of Vehicle Behavior
Summer 2025 Seminar Series
Finding Preferred Automated Vehicle Driving Style via Exposure to Various Types of Vehicle Behavior
Event Date: | June 11, 2025 |
---|---|
Speaker: | Myeongkyu Lee |
Sponsor: | Dr. Brandon Pitts |
Time: | 12:00pm |
Location: | GRIS 134 |
Priority: | No |
School or Program: | Industrial Engineering |
College Calendar: | Show |
ABSTRACT
As automated vehicle (AV) technologies advance, drivers will become increasingly able to personalize their interactions with them. One aspect that drivers desire to customize is the driving style of AVs. While previous work suggests that personalization of AVs can enhance driver trust and satisfaction, little is known about how drivers perceive the behavior of AVs and the ways they wish to modify AV driving styles to match their preferences. This study defined four driving behaviors based on two factors: safety (cautious vs. risky) and efficiency (slow vs. fast) to understand how drivers find their preferred driving style. Thirty-two participants completed 10 trials and experienced one of four driving styles in a simulator: very cautious and slow, moderately cautious but fast, very fast and risky, or risky but slow. After each drive, participants provided feedback to the AV to either maintain or change its driving behavior. The success rate of executing this feedback was 80%, and participants could take over control of the vehicle at any time. Overall, drivers preferred a balanced driving style, avoiding too aggressive or too conservative behaviors. Additionally, driver takeover frequency increased when the AV adopted a more aggressive style consisting of high speed and risk. Finally, driver trust in the AV peaked at the balanced driving style, suggesting that neither overly conservative nor overly aggressive behavior maximized trust. These findings can contribute to the development of personalized AV driving style selection methods that prioritize safety and satisfaction.
BIOGRAPHY
Myeongkyu Lee (he/him/his) is a PhD student in the Edwardson School of Industrial Engineering at Purdue University, advised by Dr. Brandon J. Pitts. He holds a B.S. in Automotive Engineering and an M.S. in Automobile and IT Convergence. His research focuses on human factors in the automated vehicle (AV) domain, including human-vehicle interaction and vehicle/driver safety.In his talk, he will discuss how drivers respond to personalized driving styles in AVs. Using a driving simulator, participants experienced four AV driving styles that varied in safety and efficiency and were able to provide feedback to adjust the AV’s behavior. He will share how drivers find their preferred driving styles, and how these preferences influence trust and takeover behavior.