Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services

Beopgwon Jung; Joonwoo Kim; Sangheon Pack

doi:10.1109/ICOIN53446.2022.9687254

Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services

Beopgwon Jung, Joonwoo Kim, Sangheon Pack

School of Electrical Engineering

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

7 Citations (Scopus)

Abstract

Collective perception service (CPS) is one of the most fundamental services in intelligent transportation systems. Since it can incur significant overhead in exchanging perceived object containers (POCs), european telecommunications standards institute (ETSI) introduced several redundancy mitigation schemes; however, there are several limitations in application to the vehicular environment. In this paper, we propose a deep reinforcement learning (DRL)-based context-Aware redundancy mitigation (DRL-CARM) scheme where various vehicular contexts (i.e., location, speed, heading, and perception area) are employed for redundancy mitigation. To derive the optimal policy on redundancy mitigation, the DRL-CARM scheme employs a deep Q-network (DQN) with a reward function on the usefulness of POC. Evaluation results demonstrate that the DRL-CARM scheme can improve the average usefulness of POC by 254% and reduce the network load by 49.4%, compared with conventional redundancy mitigation schemes.

Original language	English
Title of host publication	36th International Conference on Information Networking, ICOIN 2022
Publisher	IEEE Computer Society
Pages	276-279
Number of pages	4
ISBN (Electronic)	9781665413329
DOIs	https://doi.org/10.1109/ICOIN53446.2022.9687254
Publication status	Published - 2022
Event	36th International Conference on Information Networking, ICOIN 2022 - Virtual, Jeju Island, Korea, Republic of Duration: 2022 Jan 12 → 2022 Jan 15

Publication series

Name	International Conference on Information Networking
Volume	2022-January
ISSN (Print)	1976-7684

Conference

Conference	36th International Conference on Information Networking, ICOIN 2022
Country/Territory	Korea, Republic of
City	Virtual, Jeju Island
Period	22/1/12 → 22/1/15

Bibliographical note

Funding Information:
ACKNOWLEDGEMENT This research was supported by National Research Foundation (NRF) of Korea Grant funded by the Korean Government (MSIT) (No. 2020R1A2C3006786).

Publisher Copyright:
© 2022 IEEE.

Keywords

Collective Perception Service
Deep Reinforcement Learning
ETSI Redundancy Mitigation Scheme
Intelligent Transportation System

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems

Access to Document

10.1109/ICOIN53446.2022.9687254

Cite this

Jung, B., Kim, J., & Pack, S. (2022). Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services. In 36th International Conference on Information Networking, ICOIN 2022 (pp. 276-279). (International Conference on Information Networking; Vol. 2022-January). IEEE Computer Society. https://doi.org/10.1109/ICOIN53446.2022.9687254

Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services. / Jung, Beopgwon; Kim, Joonwoo; Pack, Sangheon.
36th International Conference on Information Networking, ICOIN 2022. IEEE Computer Society, 2022. p. 276-279 (International Conference on Information Networking; Vol. 2022-January).

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

Jung, B, Kim, J & Pack, S 2022, Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services. in 36th International Conference on Information Networking, ICOIN 2022. International Conference on Information Networking, vol. 2022-January, IEEE Computer Society, pp. 276-279, 36th International Conference on Information Networking, ICOIN 2022, Virtual, Jeju Island, Korea, Republic of, 22/1/12. https://doi.org/10.1109/ICOIN53446.2022.9687254

@inproceedings{cccb4169143940b297d614bf8f083a6a,

title = "Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services",

abstract = "Collective perception service (CPS) is one of the most fundamental services in intelligent transportation systems. Since it can incur significant overhead in exchanging perceived object containers (POCs), european telecommunications standards institute (ETSI) introduced several redundancy mitigation schemes; however, there are several limitations in application to the vehicular environment. In this paper, we propose a deep reinforcement learning (DRL)-based context-Aware redundancy mitigation (DRL-CARM) scheme where various vehicular contexts (i.e., location, speed, heading, and perception area) are employed for redundancy mitigation. To derive the optimal policy on redundancy mitigation, the DRL-CARM scheme employs a deep Q-network (DQN) with a reward function on the usefulness of POC. Evaluation results demonstrate that the DRL-CARM scheme can improve the average usefulness of POC by 254% and reduce the network load by 49.4%, compared with conventional redundancy mitigation schemes.",

keywords = "Collective Perception Service, Deep Reinforcement Learning, ETSI Redundancy Mitigation Scheme, Intelligent Transportation System",

author = "Beopgwon Jung and Joonwoo Kim and Sangheon Pack",

note = "Funding Information: ACKNOWLEDGEMENT This research was supported by National Research Foundation (NRF) of Korea Grant funded by the Korean Government (MSIT) (No. 2020R1A2C3006786). Publisher Copyright: {\textcopyright} 2022 IEEE.; 36th International Conference on Information Networking, ICOIN 2022 ; Conference date: 12-01-2022 Through 15-01-2022",

year = "2022",

doi = "10.1109/ICOIN53446.2022.9687254",

language = "English",

series = "International Conference on Information Networking",

publisher = "IEEE Computer Society",

pages = "276--279",

booktitle = "36th International Conference on Information Networking, ICOIN 2022",

}

TY - GEN

T1 - Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services

AU - Jung, Beopgwon

AU - Kim, Joonwoo

AU - Pack, Sangheon

N1 - Funding Information: ACKNOWLEDGEMENT This research was supported by National Research Foundation (NRF) of Korea Grant funded by the Korean Government (MSIT) (No. 2020R1A2C3006786). Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - Collective perception service (CPS) is one of the most fundamental services in intelligent transportation systems. Since it can incur significant overhead in exchanging perceived object containers (POCs), european telecommunications standards institute (ETSI) introduced several redundancy mitigation schemes; however, there are several limitations in application to the vehicular environment. In this paper, we propose a deep reinforcement learning (DRL)-based context-Aware redundancy mitigation (DRL-CARM) scheme where various vehicular contexts (i.e., location, speed, heading, and perception area) are employed for redundancy mitigation. To derive the optimal policy on redundancy mitigation, the DRL-CARM scheme employs a deep Q-network (DQN) with a reward function on the usefulness of POC. Evaluation results demonstrate that the DRL-CARM scheme can improve the average usefulness of POC by 254% and reduce the network load by 49.4%, compared with conventional redundancy mitigation schemes.

AB - Collective perception service (CPS) is one of the most fundamental services in intelligent transportation systems. Since it can incur significant overhead in exchanging perceived object containers (POCs), european telecommunications standards institute (ETSI) introduced several redundancy mitigation schemes; however, there are several limitations in application to the vehicular environment. In this paper, we propose a deep reinforcement learning (DRL)-based context-Aware redundancy mitigation (DRL-CARM) scheme where various vehicular contexts (i.e., location, speed, heading, and perception area) are employed for redundancy mitigation. To derive the optimal policy on redundancy mitigation, the DRL-CARM scheme employs a deep Q-network (DQN) with a reward function on the usefulness of POC. Evaluation results demonstrate that the DRL-CARM scheme can improve the average usefulness of POC by 254% and reduce the network load by 49.4%, compared with conventional redundancy mitigation schemes.

KW - Collective Perception Service

KW - Deep Reinforcement Learning

KW - ETSI Redundancy Mitigation Scheme

KW - Intelligent Transportation System

UR - http://www.scopus.com/inward/record.url?scp=85125652269&partnerID=8YFLogxK

U2 - 10.1109/ICOIN53446.2022.9687254

DO - 10.1109/ICOIN53446.2022.9687254

M3 - Conference contribution

AN - SCOPUS:85125652269

T3 - International Conference on Information Networking

SP - 276

EP - 279

BT - 36th International Conference on Information Networking, ICOIN 2022

PB - IEEE Computer Society

T2 - 36th International Conference on Information Networking, ICOIN 2022

Y2 - 12 January 2022 through 15 January 2022

ER -

Deep Reinforcement Learning-based Context-Aware Redundancy Mitigation for Vehicular Collective Perception Services

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this