The design of implicit pedestrian-autonomous vehicle interactions
Abstract/Contents
- Abstract
- Automotive manufacturers and researchers have posited that pedestrians will feel uncomfortable crossing in front of driverless vehicles—autonomous vehicles that lack visible human operators. This belief is often based on the lack of driver-related cues such as eye contact and hand waves that can communicate when it is safe for a pedestrian to cross. As such, interface concepts based on explicit communication such as displays, lights, or projections have been developed to indicate to pedestrians when it is safe to cross. However, the underlying need for such interfaces, and their efficacy, have not been thoroughly explored in on-road settings. It is likely that autonomous vehicles can simply depend on legacy behaviors, such as slowing appropriately, to interact with pedestrians. This dissertation explores the design of motion and engine sound as implicit communication between autonomous vehicles and pedestrians. We begin by investigating motor sound's subtle influence on human-robot interactions, working towards identifying characteristics of sound that contribute to particular subjective experiences. We show that changing the timbre of sound alone, while leaving all other dimensions constant, can change someone's perception of a robot. We then apply these findings in the context of autonomous vehicles, where engine sound could implicitly cue that the vehicle will yield for a pedestrian, particularly in electric vehicles that will soon be required to project artificial engine sound while moving at low speeds. Through a field study, we demonstrate how a visibly driverless vehicle's motion and engine sound can communicate to pedestrians that it is safe to cross. During these studies, we drove a simulated driverless vehicle in everyday traffic, and researchers standing by interviewed unsuspecting pedestrians who interacted with the vehicle. We then analyzed video of each interaction to evaluate responses to the vehicle. While the novelty of a driverless vehicle did surprise some, many did not even notice its autonomous nature. Most pedestrians reported positive experiences, and all crossed in front of the vehicle without the need for explicit communication. While not inclusive of all crossing situations, this work challenges the assumption that driverless vehicles will need explicit displays to replace driver-pedestrian communication at crosswalks. In addition, it contributes design requirements for socially appropriate stopping behavior and artificial engine sound for autonomous electric vehicles. In doing so, we bring automotive design a step closer to solving a critical human factors problem that could hinder the widespread adoption of autonomous vehicles.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2019; ©2019 |
| Publication date | 2019; 2019 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Moore, Dylan James |
|---|---|
| Degree supervisor | Leifer, Larry J |
| Degree supervisor | Sirkin, David Andrew, 1960- |
| Thesis advisor | Leifer, Larry J |
| Thesis advisor | Sirkin, David Andrew, 1960- |
| Thesis advisor | Cutkosky, Mark R |
| Thesis advisor | Ju, Wendy, 1975- |
| Thesis advisor | Shanks, Michael |
| Degree committee member | Cutkosky, Mark R |
| Degree committee member | Ju, Wendy, 1975- |
| Degree committee member | Shanks, Michael |
| Associated with | Stanford University, Department of Mechanical Engineering. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Dylan James Moore. |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering. |
| Thesis | Thesis Ph.D. Stanford University 2019. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2019 by Dylan James Moore
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...