2022
DOI: 10.3390/su14159680
|View full text |Cite
|
Sign up to set email alerts
|

Vision Transformer for Detecting Critical Situations and Extracting Functional Scenario for Automated Vehicle Safety Assessment

Abstract: Automated Vehicles (AVs) are attracting attention as a safer mobility option thanks to the recent advancement of various sensing technologies that realize a much quicker Perception–Reaction Time than Human-Driven Vehicles (HVs). However, AVs are not entirely free from the risk of accidents, and we currently lack a systematic and reliable method to improve AV safety functions. The manual composition of accident scenarios does not scale. Simulation-based methods do not fully cover the peculiar AV accident patter… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
15
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
7
1
1

Relationship

1
8

Authors

Journals

citations
Cited by 27 publications
(15 citation statements)
references
References 66 publications
0
15
0
Order By: Relevance
“…Studies on accident prevention technology development and scenario generation for vehicle safety evaluation have been conducted based on real-data analysis [ [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] ]. Nitsche et al (2017) [ 32 ] derived 34 crash scenarios using clustering and association analysis and extracted 12 scenarios of accidents with a high risk of injury.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…Studies on accident prevention technology development and scenario generation for vehicle safety evaluation have been conducted based on real-data analysis [ [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] ]. Nitsche et al (2017) [ 32 ] derived 34 crash scenarios using clustering and association analysis and extracted 12 scenarios of accidents with a high risk of injury.…”
Section: Related Workmentioning
confidence: 99%
“…However, it is important to acknowledge certain limitations associated with the use of HV data in this research. Kang et al (2022a) [ 39 ] used a vision transformer to detect critical situations involving AVs. The vision transformer caught critical situations at an F1 score of 94 %.…”
Section: Related Workmentioning
confidence: 99%
“…The experiment was conducted using a model based on the proposed framework as a traffic-accident detector and three additional models for comparison. The models included a DCNN [15], an LSTMDTR [29], the Vision Transformer for Traffic Accidents (ViT-TA) [32], and the proposed framework. Table 2 lists the hyperparameters used for the training and validation of each model.…”
Section: Experimental Environmentmentioning
confidence: 99%
“…In the vehicular context, different sensors are crucial sources of digital evidence [28]. One of these sensors is the dashcam, which has been used for multiple purposes [29][30][31][32]. Dashboard cameras have been used to increase road safety [33].…”
Section: Dashcam Technology and Forensic Analysismentioning
confidence: 99%