Back to search

University of Central Florida

Comprehensive Analytics of Automated Vehicles’ Effectiveness, Benefits, and Crash Severity

Abstract

dc:description.abstract

Automated Vehicles (AV) are increasingly becoming a reality, offering major road safety improvements and advancements through the development of Advanced Driver Assistance Systems (ADAS) and Automated Driving Systems (ADS). Although research on AVs is progressing, several key gaps remain. There is a lack of sufficient matched data between AV and HDV crashes to compare their differences, along with a need for more detailed datasets to analyze ADS and ADAS crash severity and assess ADAS safety effectiveness. First, the research investigates overall trends within the AV dataset by utilizing 1,099 ADS, 1,001 ADAS and 35,113 HDV crash data. including ADS (Society of Automotive Engineers (SAE) Level 4) and ADAS (SAE Level 2). And then, a matched case-control design is employed to compare differences between ADS and Human-Driven Vehicle (HDV). Second, this research focuses on analyzing the injury severity of 1,001 ADAS and 548 ADS crashes. Two random parameters multinomial logit models with heterogeneity in the means are considered to analyze the contributing factors of injury severity for the ADAS and ADS. Lastly, this research assesses the safety effectiveness of ADAS using 3,549 ADAS-equipped and 4.51 million non-ADAS vehicles involved in two-vehicle crashes across 34 models and 41 brands, with these ADAS always activated. The Quasi-Induced Exposure (QIE) is applied to assess the safety effectiveness across various conditions. Also, a series of Bayesian hierarchical models are compared and the best fit Bayesian Hierarchical Random-Parameter Model (BHRPM) considering unobserved heterogeneity is selected to investigate the determinants influencing the ADAS-equipped vehicle responsibility in a crash. iv Through these objectives, this research assesses ADAS and ADS vehicle safety using real-world crash data, offering insights to guide deployment strategies and future advancements in AV safety.

Degree

thesis:*
Grantor dc:publisher
University of Central Florida

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Ding, Shengxuan
Contributors dc:contributor
  • Mohamed Abdel-Aty

Subjects

dc:subject × 7

Identifiers

dc:identifier.*
Identifier
DP0029697

Chain of custody

source
Harvested from
Central Florida
Base URL
stars.library.ucf.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Ding, Shengxuan. Comprehensive Analytics of Automated Vehicles’ Effectiveness, Benefits, and Crash Severity. University of Central Florida, https://stars.library.ucf.edu/etd2024/399