An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform

Zishen Wan; Yuyang Zhang; Arijit Raychowdhury; Bo Yu; Yanjun Zhang; Shaoshan Liu

doi:10.1109/AICAS51828.2021.9458486

An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform

Zishen Wan
, Yuyang Zhang
, Arijit Raychowdhury
, Bo Yu
, Yanjun Zhang
, Shaoshan Liu

网络空间安全学院

Georgia Institute of Technology
Beijing Institute of Technology
PerceptIn Inc.

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

16 引用（Scopus）

摘要

In our past few years' of commercial deployment experiences, we identify localization as a critical task in autonomous machine applications, and a great acceleration target. In this paper, based on the observation that the visual frontend is a major performance and energy consumption bottleneck, we present our design and implementation of an energy-efficient hardware architecture for ORB (Oriented-Fast and Rotated-BRIEF) based localization system on FPGAs. To support our multi-sensor autonomous machine localization system, we present hardware synchronization, frame-multiplexing, and parallelization techniques, which are integrated in our design. Compared to Nvidia TX1 and Intel i7, our FPGA-based implementation achieves 5.6\times and 3.4\times speedup, as well as 3.0\times and 34.6\times power reduction, respectively.

源语言	英语
主期刊名	2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021
出版商	Institute of Electrical and Electronics Engineers Inc.
ISBN（电子版）	9781665419130
DOI	https://doi.org/10.1109/AICAS51828.2021.9458486
出版状态	已出版 - 6 6月 2021
活动	3rd IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021 - Washington, 美国期限: 6 6月 2021 → 9 6月 2021

出版系列

姓名	2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021

会议

会议	3rd IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021
国家/地区	美国
市	Washington
时期	6/06/21 → 9/06/21

联合国可持续发展目标

此成果有助于实现下列可持续发展目标：

可持续发展目标 7 经济适用的清洁能源

访问文件

10.1109/AICAS51828.2021.9458486

其它文件与链接

链接到 Scopus 的出版物

引用此

Wan, Z., Zhang, Y., Raychowdhury, A., Yu, B., Zhang, Y., & Liu, S. (2021). An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform. 在 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021 文章 9458486 (2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/AICAS51828.2021.9458486

Wan, Zishen ; Zhang, Yuyang ; Raychowdhury, Arijit 等. / An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform. 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021. Institute of Electrical and Electronics Engineers Inc., 2021. (2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021).

@inproceedings{77c10cef714a4f839ac15c67b9433b34,

title = "An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform",

abstract = "In our past few years' of commercial deployment experiences, we identify localization as a critical task in autonomous machine applications, and a great acceleration target. In this paper, based on the observation that the visual frontend is a major performance and energy consumption bottleneck, we present our design and implementation of an energy-efficient hardware architecture for ORB (Oriented-Fast and Rotated-BRIEF) based localization system on FPGAs. To support our multi-sensor autonomous machine localization system, we present hardware synchronization, frame-multiplexing, and parallelization techniques, which are integrated in our design. Compared to Nvidia TX1 and Intel i7, our FPGA-based implementation achieves 5.6\textbackslash{}times and 3.4\textbackslash{}times speedup, as well as 3.0\textbackslash{}times and 34.6\textbackslash{}times power reduction, respectively.",

author = "Zishen Wan and Yuyang Zhang and Arijit Raychowdhury and Bo Yu and Yanjun Zhang and Shaoshan Liu",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 3rd IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021 ; Conference date: 06-06-2021 Through 09-06-2021",

year = "2021",

month = jun,

day = "6",

doi = "10.1109/AICAS51828.2021.9458486",

language = "English",

series = "2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021",

address = "United States",

}

Wan, Z, Zhang, Y, Raychowdhury, A, Yu, B, Zhang, Y & Liu, S 2021, An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform. 在 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021., 9458486, 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021, Institute of Electrical and Electronics Engineers Inc., 3rd IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021, Washington, 美国, 6/06/21. https://doi.org/10.1109/AICAS51828.2021.9458486

An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform. / Wan, Zishen; Zhang, Yuyang; Raychowdhury, Arijit 等.
2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021. Institute of Electrical and Electronics Engineers Inc., 2021. 9458486 (2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021).

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

TY - GEN

T1 - An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform

AU - Wan, Zishen

AU - Zhang, Yuyang

AU - Raychowdhury, Arijit

AU - Yu, Bo

AU - Zhang, Yanjun

AU - Liu, Shaoshan

PY - 2021/6/6

Y1 - 2021/6/6

N2 - In our past few years' of commercial deployment experiences, we identify localization as a critical task in autonomous machine applications, and a great acceleration target. In this paper, based on the observation that the visual frontend is a major performance and energy consumption bottleneck, we present our design and implementation of an energy-efficient hardware architecture for ORB (Oriented-Fast and Rotated-BRIEF) based localization system on FPGAs. To support our multi-sensor autonomous machine localization system, we present hardware synchronization, frame-multiplexing, and parallelization techniques, which are integrated in our design. Compared to Nvidia TX1 and Intel i7, our FPGA-based implementation achieves 5.6\times and 3.4\times speedup, as well as 3.0\times and 34.6\times power reduction, respectively.

AB - In our past few years' of commercial deployment experiences, we identify localization as a critical task in autonomous machine applications, and a great acceleration target. In this paper, based on the observation that the visual frontend is a major performance and energy consumption bottleneck, we present our design and implementation of an energy-efficient hardware architecture for ORB (Oriented-Fast and Rotated-BRIEF) based localization system on FPGAs. To support our multi-sensor autonomous machine localization system, we present hardware synchronization, frame-multiplexing, and parallelization techniques, which are integrated in our design. Compared to Nvidia TX1 and Intel i7, our FPGA-based implementation achieves 5.6\times and 3.4\times speedup, as well as 3.0\times and 34.6\times power reduction, respectively.

UR - https://www.scopus.com/pages/publications/85106075065

U2 - 10.1109/AICAS51828.2021.9458486

DO - 10.1109/AICAS51828.2021.9458486

M3 - Conference contribution

AN - SCOPUS:85106075065

T3 - 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021

BT - 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021

Y2 - 6 June 2021 through 9 June 2021

ER -

Wan Z, Zhang Y, Raychowdhury A, Yu B, Zhang Y, Liu S. An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform. 在 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021. Institute of Electrical and Electronics Engineers Inc. 2021. 9458486. (2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems, AICAS 2021). doi: 10.1109/AICAS51828.2021.9458486

An Energy-Efficient Quad-Camera Visual System for Autonomous Machines on FPGA Platform

摘要

出版系列

会议

联合国可持续发展目标

访问文件

其它文件与链接

指纹

引用此