Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset

Sungmoon Ahn; Shiho Kim

doi:10.5220/0011139800003274

Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset

Sungmoon Ahn, Shiho Kim

School of Integrated Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We propose a method to evaluate the RSS model using data obtained from real roads. Recently, the Responsibility-Sensitive Safety (RSS) model representing the minimum safety distance has been proposed. After that, there were studies to evaluate the RSS model using simulators. Most virtual simulation studies showed that the RSS model guarantees safety but adversely affects traffic flow by estimating the distance too long than necessary. We evaluated the RSS model using data obtained in natural situational environments, unlike previous studies. First, we found correlations representing distances between vehicles from the data using Graph Neural Networks. Using the obtained correlations, we expressed it as a mathematical model through symbolic regression. As a result of comparing the model we found with the RSS model, we verified that the RSS model has a significant trade-off between safety and traffic flow.

Original language	English
Title of host publication	SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
Editors	Gerd Wagner, Frank Werner, Floriano De Rango
Publisher	Science and Technology Publications, Lda
Pages	210-217
Number of pages	8
ISBN (Print)	9789897585784
DOIs	https://doi.org/10.5220/0011139800003274
Publication status	Published - 2022
Event	12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications , SIMULTECH 2022 - Lisbon, Portugal Duration: 2022 Jul 14 → 2022 Jul 16

Publication series

Name	Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications
Volume	1
ISSN (Print)	2184-2841

Conference

Conference	12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications , SIMULTECH 2022
Country/Territory	Portugal
City	Lisbon
Period	22/7/14 → 22/7/16

Bibliographical note

Publisher Copyright:
© 2022 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved.

All Science Journal Classification (ASJC) codes

Computational Theory and Mathematics
Computer Science Applications
Safety, Risk, Reliability and Quality
Modelling and Simulation

Access to Document

10.5220/0011139800003274

Cite this

Ahn, S., & Kim, S. (2022). Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset. In G. Wagner, F. Werner, & F. De Rango (Eds.), SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (pp. 210-217). (Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications; Vol. 1). Science and Technology Publications, Lda. https://doi.org/10.5220/0011139800003274

Ahn, Sungmoon ; Kim, Shiho. / Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset. SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications. editor / Gerd Wagner ; Frank Werner ; Floriano De Rango. Science and Technology Publications, Lda, 2022. pp. 210-217 (Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications).

@inproceedings{ea9b34f981404ca6a4f3062d9a6e8416,

title = "Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset",

abstract = "We propose a method to evaluate the RSS model using data obtained from real roads. Recently, the Responsibility-Sensitive Safety (RSS) model representing the minimum safety distance has been proposed. After that, there were studies to evaluate the RSS model using simulators. Most virtual simulation studies showed that the RSS model guarantees safety but adversely affects traffic flow by estimating the distance too long than necessary. We evaluated the RSS model using data obtained in natural situational environments, unlike previous studies. First, we found correlations representing distances between vehicles from the data using Graph Neural Networks. Using the obtained correlations, we expressed it as a mathematical model through symbolic regression. As a result of comparing the model we found with the RSS model, we verified that the RSS model has a significant trade-off between safety and traffic flow.",

keywords = "Data Analytics and Simulation, Intelligent Transportation, Neural Nets and Fuzzy Systems",

author = "Sungmoon Ahn and Shiho Kim",

note = "Publisher Copyright: {\textcopyright} 2022 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved.; 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications , SIMULTECH 2022 ; Conference date: 14-07-2022 Through 16-07-2022",

year = "2022",

doi = "10.5220/0011139800003274",

language = "English",

isbn = "9789897585784",

series = "Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications",

publisher = "Science and Technology Publications, Lda",

pages = "210--217",

editor = "Gerd Wagner and Frank Werner and {De Rango}, Floriano",

booktitle = "SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications",

}

Ahn, S & Kim, S 2022, Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset. in G Wagner, F Werner & F De Rango (eds), SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications. Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications, vol. 1, Science and Technology Publications, Lda, pp. 210-217, 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications , SIMULTECH 2022, Lisbon, Portugal, 22/7/14. https://doi.org/10.5220/0011139800003274

Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset. / Ahn, Sungmoon; Kim, Shiho.
SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications. ed. / Gerd Wagner; Frank Werner; Floriano De Rango. Science and Technology Publications, Lda, 2022. p. 210-217 (Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset

AU - Ahn, Sungmoon

AU - Kim, Shiho

PY - 2022

Y1 - 2022

N2 - We propose a method to evaluate the RSS model using data obtained from real roads. Recently, the Responsibility-Sensitive Safety (RSS) model representing the minimum safety distance has been proposed. After that, there were studies to evaluate the RSS model using simulators. Most virtual simulation studies showed that the RSS model guarantees safety but adversely affects traffic flow by estimating the distance too long than necessary. We evaluated the RSS model using data obtained in natural situational environments, unlike previous studies. First, we found correlations representing distances between vehicles from the data using Graph Neural Networks. Using the obtained correlations, we expressed it as a mathematical model through symbolic regression. As a result of comparing the model we found with the RSS model, we verified that the RSS model has a significant trade-off between safety and traffic flow.

AB - We propose a method to evaluate the RSS model using data obtained from real roads. Recently, the Responsibility-Sensitive Safety (RSS) model representing the minimum safety distance has been proposed. After that, there were studies to evaluate the RSS model using simulators. Most virtual simulation studies showed that the RSS model guarantees safety but adversely affects traffic flow by estimating the distance too long than necessary. We evaluated the RSS model using data obtained in natural situational environments, unlike previous studies. First, we found correlations representing distances between vehicles from the data using Graph Neural Networks. Using the obtained correlations, we expressed it as a mathematical model through symbolic regression. As a result of comparing the model we found with the RSS model, we verified that the RSS model has a significant trade-off between safety and traffic flow.

KW - Data Analytics and Simulation

KW - Intelligent Transportation

KW - Neural Nets and Fuzzy Systems

UR - http://www.scopus.com/inward/record.url?scp=85178567974&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85178567974&partnerID=8YFLogxK

U2 - 10.5220/0011139800003274

DO - 10.5220/0011139800003274

M3 - Conference contribution

AN - SCOPUS:85178567974

SN - 9789897585784

T3 - Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications

SP - 210

EP - 217

BT - SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

A2 - Wagner, Gerd

A2 - Werner, Frank

A2 - De Rango, Floriano

PB - Science and Technology Publications, Lda

T2 - 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications , SIMULTECH 2022

Y2 - 14 July 2022 through 16 July 2022

ER -

Ahn S, Kim S. Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset. In Wagner G, Werner F, De Rango F, editors, SIMULTECH 2022 - Proceedings of the 12th International Conference on Simulation and Modeling Methodologies, Technologies and Applications. Science and Technology Publications, Lda. 2022. p. 210-217. (Proceedings of the International Conference on Simulation and Modeling Methodologies, Technologies and Applications). doi: 10.5220/0011139800003274

Assessment of the RSS Model Suitability using Graph Neural Network based on a Naturalistic Driving Dataset

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this