Back to search

Georgia Institute of Technology

Driving Like an Experienced, Well-Behaved Human Driver: Designing and Evaluating a Spatial Social Agent for Autonomous Vehicles Through Co-Design and Trust Measurement

Abstract

dc:description.abstract

To promote people’s acceptance of autonomous vehicles (AV), designers have been trying to calibrate user trust in AV through novel human-AV interaction designs. In this thesis, we explore what kind of anthropomorphic AV interface designs may better enhance user trust in AV systems. More specifically, we propose the novel concept of “an AV interface that is perceived as an experienced, well-behaved human driver” since people tend to place more trust in experienced, well-behaved human drivers than in autonomous driving systems. We explored this design concept through two sessions of co-design workshops with 12 industrial designers across all levels. We employed “Annotated Portfolio” to annotate the graphical design materials collected in the workshop and analyzed these annotations together with the transcripts generated from the workshop through Reflexive Thematic Analysis. The results suggest novel design insights and guidelines for an AV interface that plays the social role of an experienced, well-behaved human driver. The design insights from this analysis informed the development of a spatial social agent prototype, which was implemented and tested within a virtual reality (VR) simulation to assess its effectiveness in fostering user trust in AV systems. The final prototype (EWDA) integrates key anthropomorphic elements that emulate the behaviors and communication patterns of an experienced, well-behaved human driver, creating a more relatable and reassuring AV interaction experience. To evaluate its impact, user testing with human-vehicle perception and human-vehicle trust measurement was conducted, examining whether the proposed design was perceived as a well-behaved, experienced human driver and significantly enhances user trust levels. The findings indicate that incorporating anthropomorphic and social interaction features into AV systems can effectively boost user trust and acceptance, highlighting the potential of the proposed anthropomorphic AV design strategies in addressing trust calibration challenges.

Degree

thesis:*
Level thesis:degree_level
Masters
Department dc:contributor.department
Industrial Design
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Yang, Huaizhuo
Advisor dc:contributor.advisor
  • Wang, Yixiao
Committee members dc:contributor.committeemember
  • Li, Mengyao
  • Li, Wayne
  • Purdy, Timothy
  • Lynn, David

Subjects

dc:subject × 4

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/77878

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Yang, Huaizhuo. Driving Like an Experienced, Well-Behaved Human Driver: Designing and Evaluating a Spatial Social Agent for Autonomous Vehicles Through Co-Design and Trust Measurement. Masters thesis, Georgia Institute of Technology, 2025. https://hdl.handle.net/1853/77878