Human operator-autonomous vehicle interactions with system bias and transitions of control
Abstract/Contents
- Abstract
- Autonomous vehicles are considered the future of transportation. They offer several potential long-term benefits, including reduction in driving and navigation chores, enhanced mobility for individuals with disabilities, improved safety and decreases in traffic collisions, and reduced infrastructure costs. Currently, however, autonomous vehicle systems in development are still limited by their abilities to detect their surroundings, interpret sensory information, and fully navigate without human input. The human operator needs to both supervise the imperfect system and collaborate with it to navigate the driving environment. The research presented in this dissertation was aimed at better understanding human operator and autonomous vehicle interactions and how they affect system design and implementation. An immersive driving simulator was modified to develop experimental studies that analyzed interactions between the human operator and the autonomous vehicle system. The first part of this research focused on the inherent algorithmic biases present in designing autonomous vehicle systems, and the effects of changes in the system's sensitivity level and autonomy level were analyzed. The results show that a system biased towards lower sensitivity can improve operator vigilance and performance. The results further demonstrate that higher levels of autonomy in vehicles can result in reduced operator performance during potentially fatal events. Subsequent research focused on transitions of control from automation, in which the system hands control of the vehicle over to the human operator because there is a failure in the autonomous driving system. The effect of varying alert frequencies and levels of failure severity on the human operator's ability to successfully navigate a transition was analyzed. Significant differences in operator performance between the presence or lack of an alarm were observed, as well as the fact that an initial system failure's severity level did not seem to affect operator performance in a later failure event. Overall, the results of this work show that engineers and designers of autonomous vehicle systems need to consider a number of scenarios when specifying system actions and responses. This research provides a starting point for further exploration of the implications of system bias, particularly with varying levels of automation. It also provides insights that can help inform the design of transitions of control. A better understanding of human operator and autonomous vehicle interactions remains critical as vehicles with higher levels of automation become more prevalent on the road.
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 | Fu, Ernestine |
|---|---|
| Degree supervisor | Fischer, Martin, 1960 July 11- |
| Degree supervisor | Leifer, Larry J |
| Thesis advisor | Fischer, Martin, 1960 July 11- |
| Thesis advisor | Leifer, Larry J |
| Thesis advisor | Cutkosky, Mark R |
| Thesis advisor | Sirkin, David Andrew, 1960- |
| Thesis advisor | Weyant, John P. (John Peter) |
| Degree committee member | Cutkosky, Mark R |
| Degree committee member | Sirkin, David Andrew, 1960- |
| Degree committee member | Weyant, John P. (John Peter) |
| Associated with | Stanford University, Civil & Environmental Engineering Department. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Ernestine Fu. |
|---|---|
| Note | Submitted to the Civil & Environmental Engineering Department. |
| Thesis | Thesis Ph.D. Stanford University 2019. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2019 by Ernestine Fu
- 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...