Driver behavior and state changes in autonomous vehicles
Abstract/Contents
- Abstract
- While human drivers are currently responsible for almost all of the transport of goods and people, autonomous vehicles are near-at-hand and are expected to replace human drivers in the next few decades. Several vehicle manufacturing and tech companies have begun testing their autonomous vehicles on public roads. During this period of testing, human drivers are employed to ensure the safety of the passengers and those in the immediate vicinity in the event of a failure of automation. Moreover, lower levels of automation, that are currently available to the general public, also require constant human supervision to ensure safety. This makes the study of the cognitive, behavioral and psycho-physiological state of drivers of automation crucial. In this thesis, I review existing literature on the study of driver state and behavior and on the methodologies employed by researchers to study them. I present a series of studies that investigate the changes in driver state while using automated vehicles. To this end, I employed various physio- logical sensors such as near-infrared spectroscopy, galvanic skin response, and electrocardiography. I argue the case for driver state monitoring in automated vehicles to ensure vigilance and driver availability for a take over of control. The first study I present investigates the reasons for the presence of sleepy and drowsy behavior in drivers of automated vehicles. I found that the reason for drowsy behavior was prolonged intervals of low cortical activity. The second and third study investigate the changes in cortical activity across different levels of automation and over time. These studies showed that drivers who used partially automated vehicles had the highest levels of cognitive activity and drivers of partial and fully automated vehicles showing significant decreases in cortical activity over time. In the last study, I employ a panel design to investigate the impact of prior and knowledge and training on driver preparedness and behavior. I found that drivers are prepared when they are trained on the possible failure modes of automation. However, this preparedness decreases over time in both drivers with and without any training. These studies add to the existing knowledge of driver state investigations while providing much needed insight into the longitudinal changes in driver state. It is my hope that this thesis provides the foundation for future investigations into training programs for safety drivers in autonomous vehicles and serve as guidance for developers of automated driving technology.
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 | 2020; ©2020 |
| Publication date | 2020; 2020 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Sibi, Srinath |
|---|---|
| Degree supervisor | Leifer, Larry J |
| Thesis advisor | Leifer, Larry J |
| Thesis advisor | Cutkosky, Mark R |
| Thesis advisor | Follmer, Sean |
| Thesis advisor | Ju, Wendy, 1975- |
| Degree committee member | Cutkosky, Mark R |
| Degree committee member | Follmer, Sean |
| Degree committee member | Ju, Wendy, 1975- |
| Associated with | Stanford University, Department of Mechanical Engineering |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Srinath Sibi. |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering. |
| Thesis | Thesis Ph.D. Stanford University 2020. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2020 by Srinath Sibi
- 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...