A Deep Learning Biomimetic Milky Way Compass
Tao, Yiting ; Lucas, Michael ; Perera, Asanka ; Teague, Samuel ; McIntyre, Timothy ; Ogunwa, Titilayo ; Warrant, Eric LU
and Chahl, Javaan
(2024)
In Biomimetics
9(10).
- Abstract
Moving in straight lines is a behaviour that enables organisms to search for food, move away from threats, and ultimately seek suitable environments in which to survive and reproduce. This study explores a vision-based technique for detecting a change in heading direction using the Milky Way (MW), one of the navigational cues that are known to be used by night-active insects. An algorithm is proposed that combines the YOLOv8m-seg model and normalised second central moments to calculate the MW orientation angle. This method addresses many likely scenarios where segmentation of the MW from the background by image thresholding or edge detection is not applicable, such as when the moon is substantial or when anthropogenic light is present.... (More)
Moving in straight lines is a behaviour that enables organisms to search for food, move away from threats, and ultimately seek suitable environments in which to survive and reproduce. This study explores a vision-based technique for detecting a change in heading direction using the Milky Way (MW), one of the navigational cues that are known to be used by night-active insects. An algorithm is proposed that combines the YOLOv8m-seg model and normalised second central moments to calculate the MW orientation angle. This method addresses many likely scenarios where segmentation of the MW from the background by image thresholding or edge detection is not applicable, such as when the moon is substantial or when anthropogenic light is present. The proposed YOLOv8m-seg model achieves a segment mAP@0.5 of (Formula presented.) on the validation dataset using our own training dataset of MW images. To explore its potential role in autonomous system applications, we compare night sky imagery and GPS heading data from a field trial in rural South Australia. The comparison results show that for short-term navigation, the segmented MW image can be used as a reliable orientation cue. There is a difference of roughly 5–10° between the proposed method and GT as the path involves left or right 90° turns at certain locations.
(Less)
- author
- Tao, Yiting
; Lucas, Michael
; Perera, Asanka
; Teague, Samuel
; McIntyre, Timothy
; Ogunwa, Titilayo
; Warrant, Eric
LU
and Chahl, Javaan
- organization
- publishing date
- 2024-10
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- biomimetic, instance segmentation, Milky Way, orientation, YOLOv8
- in
- Biomimetics
- volume
- 9
- issue
- 10
- article number
- 620
- publisher
- MDPI AG
- external identifiers
-
- pmid:39451825
- scopus:85207687038
- ISSN
- 2313-7673
- DOI
- 10.3390/biomimetics9100620
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2024 by the authors.
- id
- 55f3c75e-f26a-4926-8e25-ff37e7f8bf3c
- date added to LUP
- 2024-12-05 12:55:48
- date last changed
- 2025-06-20 04:35:02
@article{55f3c75e-f26a-4926-8e25-ff37e7f8bf3c,
abstract = {{<p>Moving in straight lines is a behaviour that enables organisms to search for food, move away from threats, and ultimately seek suitable environments in which to survive and reproduce. This study explores a vision-based technique for detecting a change in heading direction using the Milky Way (MW), one of the navigational cues that are known to be used by night-active insects. An algorithm is proposed that combines the YOLOv8m-seg model and normalised second central moments to calculate the MW orientation angle. This method addresses many likely scenarios where segmentation of the MW from the background by image thresholding or edge detection is not applicable, such as when the moon is substantial or when anthropogenic light is present. The proposed YOLOv8m-seg model achieves a segment mAP@0.5 of (Formula presented.) on the validation dataset using our own training dataset of MW images. To explore its potential role in autonomous system applications, we compare night sky imagery and GPS heading data from a field trial in rural South Australia. The comparison results show that for short-term navigation, the segmented MW image can be used as a reliable orientation cue. There is a difference of roughly 5–10° between the proposed method and GT as the path involves left or right 90° turns at certain locations.</p>}},
author = {{Tao, Yiting and Lucas, Michael and Perera, Asanka and Teague, Samuel and McIntyre, Timothy and Ogunwa, Titilayo and Warrant, Eric and Chahl, Javaan}},
issn = {{2313-7673}},
keywords = {{biomimetic; instance segmentation; Milky Way; orientation; YOLOv8}},
language = {{eng}},
number = {{10}},
publisher = {{MDPI AG}},
series = {{Biomimetics}},
title = {{A Deep Learning Biomimetic Milky Way Compass}},
url = {{http://dx.doi.org/10.3390/biomimetics9100620}},
doi = {{10.3390/biomimetics9100620}},
volume = {{9}},
year = {{2024}},
}