Low-Latency Edge Video Analytics for On-Road Perception of Autonomous Ground Vehicles

Jie Lin; Peng Yang; Ning Zhang; Feng Lyu; Xianfu Chen; Li Yu

doi:10.1109/TII.2022.3181986

Low-Latency Edge Video Analytics for On-Road Perception of Autonomous Ground Vehicles

Jie Lin
, Peng Yang^*
, Ning Zhang
, Feng Lyu
, Xianfu Chen
, Li Yu

^*Corresponding author for this work

BA6401 Connectivity solutions

Research output: Contribution to journal › Article › Scientific › peer-review

38 Citations (Scopus)

Abstract

To improve the transportation efficiency of advanced manufacturing, cameras have been extensively deployed to enhance the on-road perception of autonomous ground vehicles in smart industrial parks. Considering the informative yet substantial data volume of contents generated by those cameras, we employ vehicle-to-everything links to deliver the captured video frames to neighboring vehicles, road-side units, or base stations, in order to respond to vehicle-control-related video queries. To help vehicles obtain low-latency and high-accuracy on-road information for autonomous driving, an optimization problem is formulated, taking into account the impact of vehicle mobility and diverse resource demands of different video queries. Then, a two-stage algorithm is proposed to determine the frame rate of video cameras, as well as the destination of the associated video frames based on matching theory. Extensive simulation results show that this approach can improve the accuracy by up to 18% and reduce the response delay by 13.2%.

Original language	English
Pages (from-to)	1512-1523
Journal	IEEE Transactions on Industrial Informatics
Volume	19
Issue number	2
DOIs	https://doi.org/10.1109/TII.2022.3181986
Publication status	Published - 1 Feb 2023
MoE publication type	A1 Journal article-refereed

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Keywords

Autonomous ground vehicle
Cameras
Collaboration
edge computing
Low latency communication
low latency communication
Roads
Throughput
Vehicle-to-everything
video analytics
Visual analytics

Access to Document

10.1109/TII.2022.3181986

Cite this

@article{711a508462d9439fa41fa807de451864,

title = "Low-Latency Edge Video Analytics for On-Road Perception of Autonomous Ground Vehicles",

abstract = "To improve the transportation efficiency of advanced manufacturing, cameras have been extensively deployed to enhance the on-road perception of autonomous ground vehicles in smart industrial parks. Considering the informative yet substantial data volume of contents generated by those cameras, we employ vehicle-to-everything links to deliver the captured video frames to neighboring vehicles, road-side units, or base stations, in order to respond to vehicle-control-related video queries. To help vehicles obtain low-latency and high-accuracy on-road information for autonomous driving, an optimization problem is formulated, taking into account the impact of vehicle mobility and diverse resource demands of different video queries. Then, a two-stage algorithm is proposed to determine the frame rate of video cameras, as well as the destination of the associated video frames based on matching theory. Extensive simulation results show that this approach can improve the accuracy by up to 18\% and reduce the response delay by 13.2\%.",

keywords = "Autonomous ground vehicle, Cameras, Collaboration, edge computing, Low latency communication, low latency communication, Roads, Throughput, Vehicle-to-everything, video analytics, Visual analytics",

author = "Jie Lin and Peng Yang and Ning Zhang and Feng Lyu and Xianfu Chen and Li Yu",

note = "Funding Information: This work was supported in part by the Natural Science Foundation of China under Grant 62001180, Grant 62002389, and Grant 61871437, in part by the Fundamental Research Funds for the Central Universities, HUST, under Grant 2021XXJS014, in part by the Natural Science Foundation of Hubei Province of China under Grant 2021CFB338 and Grant 2019CFA022, and in part by the Natural Science Foundation of Hunan Province of China under Grant 2021JJ20079 Publisher Copyright: {\textcopyright} 2005-2012 IEEE.",

year = "2023",

month = feb,

day = "1",

doi = "10.1109/TII.2022.3181986",

language = "English",

volume = "19",

pages = "1512--1523",

journal = "IEEE Transactions on Industrial Informatics",

issn = "1551-3203",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

number = "2",

}

TY - JOUR

T1 - Low-Latency Edge Video Analytics for On-Road Perception of Autonomous Ground Vehicles

AU - Lin, Jie

AU - Yang, Peng

AU - Zhang, Ning

AU - Lyu, Feng

AU - Chen, Xianfu

AU - Yu, Li

N1 - Funding Information: This work was supported in part by the Natural Science Foundation of China under Grant 62001180, Grant 62002389, and Grant 61871437, in part by the Fundamental Research Funds for the Central Universities, HUST, under Grant 2021XXJS014, in part by the Natural Science Foundation of Hubei Province of China under Grant 2021CFB338 and Grant 2019CFA022, and in part by the Natural Science Foundation of Hunan Province of China under Grant 2021JJ20079 Publisher Copyright: © 2005-2012 IEEE.

PY - 2023/2/1

Y1 - 2023/2/1

N2 - To improve the transportation efficiency of advanced manufacturing, cameras have been extensively deployed to enhance the on-road perception of autonomous ground vehicles in smart industrial parks. Considering the informative yet substantial data volume of contents generated by those cameras, we employ vehicle-to-everything links to deliver the captured video frames to neighboring vehicles, road-side units, or base stations, in order to respond to vehicle-control-related video queries. To help vehicles obtain low-latency and high-accuracy on-road information for autonomous driving, an optimization problem is formulated, taking into account the impact of vehicle mobility and diverse resource demands of different video queries. Then, a two-stage algorithm is proposed to determine the frame rate of video cameras, as well as the destination of the associated video frames based on matching theory. Extensive simulation results show that this approach can improve the accuracy by up to 18% and reduce the response delay by 13.2%.

AB - To improve the transportation efficiency of advanced manufacturing, cameras have been extensively deployed to enhance the on-road perception of autonomous ground vehicles in smart industrial parks. Considering the informative yet substantial data volume of contents generated by those cameras, we employ vehicle-to-everything links to deliver the captured video frames to neighboring vehicles, road-side units, or base stations, in order to respond to vehicle-control-related video queries. To help vehicles obtain low-latency and high-accuracy on-road information for autonomous driving, an optimization problem is formulated, taking into account the impact of vehicle mobility and diverse resource demands of different video queries. Then, a two-stage algorithm is proposed to determine the frame rate of video cameras, as well as the destination of the associated video frames based on matching theory. Extensive simulation results show that this approach can improve the accuracy by up to 18% and reduce the response delay by 13.2%.

KW - Autonomous ground vehicle

KW - Cameras

KW - Collaboration

KW - edge computing

KW - Low latency communication

KW - low latency communication

KW - Roads

KW - Throughput

KW - Vehicle-to-everything

KW - video analytics

KW - Visual analytics

UR - https://www.scopus.com/pages/publications/85132718474

U2 - 10.1109/TII.2022.3181986

DO - 10.1109/TII.2022.3181986

M3 - Article

AN - SCOPUS:85132718474

SN - 1551-3203

VL - 19

SP - 1512

EP - 1523

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

IS - 2

ER -

Low-Latency Edge Video Analytics for On-Road Perception of Autonomous Ground Vehicles

Abstract

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this