THE COMPARATION WITH YOLO AND PNP METHOD OF VISION NAVIGATION FOR HELICOPTER LANDING

Jiayu She; Yanfang Liu; Rui Zhou; Naiming Qi

doi:10.1049/icp.2021.0331

THE COMPARATION WITH YOLO AND PNP METHOD OF VISION NAVIGATION FOR HELICOPTER LANDING

Jiayu She
, Yanfang Liu^*
, Rui Zhou
, Naiming Qi

^*Corresponding author for this work

School of Astronautics, Harbin Institute of Technology

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

1 Scopus citations

Abstract

Vision navigation is increasingly used in helicopter landing, where the position and attitude of the helicopter with respect to the landing target is measured by the camera. YOLO-6D pose algorithm is proposed in this paper, which is based on You Only Look Once (YOLO) network. The original structure is modified by reducing the convolution layers and adding a fully connected layer. These modification enables the YOLO-6D network to output the 6-demensional position directly and reduce the time consumed in the training and executing. Comparisons are carried out between the YOLO-6D and the PnP algorithms in accuracy and cost time under different working conditions. Experiment results show that YOLO-6D has a more robust performance with respect to the working distance, image noise, and image vibration than PnP method.

Original language	English
Title of host publication	IET Conference Proceedings
Publisher	Institution of Engineering and Technology
Pages	279-284
Number of pages	6
Volume	2020
Edition	3
ISBN (Electronic)	9781839534195
DOIs	https://doi.org/10.1049/icp.2021.0331
State	Published - 2020
Externally published	Yes
Event	2020 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2020 - Virtual, Online Duration: 18 Sep 2020 → 21 Sep 2020

Conference

Conference	2020 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2020
City	Virtual, Online
Period	18/09/20 → 21/09/20

Keywords

Helicopter Landing
PnP
Pose Estimation
YOLO-6D

Access to Document

10.1049/icp.2021.0331

Cite this

@inproceedings{3c24f1c9bae54b35a5d64adbced1ef8f,

title = "THE COMPARATION WITH YOLO AND PNP METHOD OF VISION NAVIGATION FOR HELICOPTER LANDING",

abstract = "Vision navigation is increasingly used in helicopter landing, where the position and attitude of the helicopter with respect to the landing target is measured by the camera. YOLO-6D pose algorithm is proposed in this paper, which is based on You Only Look Once (YOLO) network. The original structure is modified by reducing the convolution layers and adding a fully connected layer. These modification enables the YOLO-6D network to output the 6-demensional position directly and reduce the time consumed in the training and executing. Comparisons are carried out between the YOLO-6D and the PnP algorithms in accuracy and cost time under different working conditions. Experiment results show that YOLO-6D has a more robust performance with respect to the working distance, image noise, and image vibration than PnP method.",

keywords = "Helicopter Landing, PnP, Pose Estimation, YOLO-6D",

author = "Jiayu She and Yanfang Liu and Rui Zhou and Naiming Qi",

note = "Publisher Copyright: {\textcopyright} 2020 The Institution of Engineering and Technology.; 2020 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2020 ; Conference date: 18-09-2020 Through 21-09-2020",

year = "2020",

doi = "10.1049/icp.2021.0331",

language = "英语",

volume = "2020",

pages = "279--284",

booktitle = "IET Conference Proceedings",

publisher = "Institution of Engineering and Technology",

address = "英国",

edition = "3",

}

She, J, Liu, Y, Zhou, R & Qi, N 2020, THE COMPARATION WITH YOLO AND PNP METHOD OF VISION NAVIGATION FOR HELICOPTER LANDING. in IET Conference Proceedings. 3 edn, vol. 2020, Institution of Engineering and Technology, pp. 279-284, 2020 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2020, Virtual, Online, 18/09/20. https://doi.org/10.1049/icp.2021.0331

TY - GEN

T1 - THE COMPARATION WITH YOLO AND PNP METHOD OF VISION NAVIGATION FOR HELICOPTER LANDING

AU - She, Jiayu

AU - Liu, Yanfang

AU - Zhou, Rui

AU - Qi, Naiming

PY - 2020

Y1 - 2020

N2 - Vision navigation is increasingly used in helicopter landing, where the position and attitude of the helicopter with respect to the landing target is measured by the camera. YOLO-6D pose algorithm is proposed in this paper, which is based on You Only Look Once (YOLO) network. The original structure is modified by reducing the convolution layers and adding a fully connected layer. These modification enables the YOLO-6D network to output the 6-demensional position directly and reduce the time consumed in the training and executing. Comparisons are carried out between the YOLO-6D and the PnP algorithms in accuracy and cost time under different working conditions. Experiment results show that YOLO-6D has a more robust performance with respect to the working distance, image noise, and image vibration than PnP method.

AB - Vision navigation is increasingly used in helicopter landing, where the position and attitude of the helicopter with respect to the landing target is measured by the camera. YOLO-6D pose algorithm is proposed in this paper, which is based on You Only Look Once (YOLO) network. The original structure is modified by reducing the convolution layers and adding a fully connected layer. These modification enables the YOLO-6D network to output the 6-demensional position directly and reduce the time consumed in the training and executing. Comparisons are carried out between the YOLO-6D and the PnP algorithms in accuracy and cost time under different working conditions. Experiment results show that YOLO-6D has a more robust performance with respect to the working distance, image noise, and image vibration than PnP method.

KW - Helicopter Landing

KW - PnP

KW - Pose Estimation

KW - YOLO-6D

UR - https://www.scopus.com/pages/publications/85174656794

U2 - 10.1049/icp.2021.0331

DO - 10.1049/icp.2021.0331

M3 - 会议稿件

AN - SCOPUS:85174656794

VL - 2020

SP - 279

EP - 284

BT - IET Conference Proceedings

PB - Institution of Engineering and Technology

T2 - 2020 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2020

Y2 - 18 September 2020 through 21 September 2020

ER -

THE COMPARATION WITH YOLO AND PNP METHOD OF VISION NAVIGATION FOR HELICOPTER LANDING

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this