Open Collections

Gill, Gurdiljot Singh

University of British Columbia

Cities are developing facilities for walking and cycling to improve safety and meet mode share goals for active transportation. People’s perceptions of safety and comfort (PSC) on those facilities mediate the influence on travel behaviour. For pedestrians, a critical component of PSC is interactions with vehicles at crosswalks. Today, pedestrians mainly interact with bicycles and cars, but they will also need to interact with self-driving vehicles (SDVs) in the future. A key gap in understanding of pedestrian PSC is how the type of interacting road user (bicycle vs. car vs. SDV) influences PSC. The goals of this dissertation are to improve understanding of PSC in intermodal interactions for a diverse and representative array of people, and to apply that knowledge to inform policy for responsible introduction of SDVs in a way that maintains walkable streets. I develop statistical models using data from two online surveys in which participants rate safety and comfort for sample videos of pedestrian interactions with bicycles, cars, and SDVs in unsignalized crosswalks. Results show that perception of yielding plays a crucial role in directly and indirectly influencing PSC, and yielding is most strongly (but not exclusively) determined by whether the pedestrian crossed before or after the interacting road user. All else equal, people perceive pedestrian interactions with SDVs as less safe and less comfortable than similar interactions with human-driven cars, which in turn are perceived as less safe and less comfortable than similar interactions with bicycles. To ensure comfort for a majority of pedestrians, SDVs must provide crossing pedestrians with more passing time than cars, and car drivers must yield more than bicycles. In terms of attitudes towards SDVs, British Columbians are almost evenly split on policies allowing SDVs, but a large majority want restrictions on SDVs on operating near pedestrian priority areas, without a human in the driver seat, or without clear identification. To ensure pedestrian-friendly introduction of SDVs, especially for older individuals, people of colour, and women, I recommend a cautious, tiered approach beginning with restricted pilot testing to allow road users to develop familiarity with SDV interactions.

Text

2024-02-22

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/87466

Civil Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/