Analysis of Cushioned Landing Strategies of Cats Based on Posture Estimation DOI Creative Commons
Li Zhang,

Liangliang Han,

Haohang Liu

et al.

Biomimetics, Journal Year: 2024, Volume and Issue: 9(11), P. 691 - 691

Published: Nov. 13, 2024

This article addresses the challenge of minimizing landing impacts for legged space robots during on-orbit operations. Inspired by agility cats, we investigate role forelimbs in process. By identifying kinematic chain cat skeleton and tracking it using animal posture estimation, derive cushioning strategy that cats use to handle impacts. The results indicate effectively transforms high-intensity into prolonged low-intensity impacts, thereby safeguarding brain internal organs. We adapt this robotic platforms through reasonable assumptions simplifications. Simulations are conducted both gravitational zero gravity environments, demonstrating optimized not only reduces ground impact prolongs duration but also suppresses robot's rebound. In gravity, enhances stable attachment target surfaces. research introduces a novel biomimetic control operations robots.

Language: Английский

Application of deep learning for evaluation of the growth rate of Daphnia magna DOI
Shinkichi Inagaki,

Yohei Kondo,

Pijar Religia

et al.

Journal of Bioscience and Bioengineering, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Language: Английский

Citations

0
Quantifying Facial Gestures Using Deep Learning in a New World Monkey DOI Creative Commons
Filippo Carugati,

Dayanna Curagi Gorio,

Chiara De Gregorio

et al.

American Journal of Primatology, Journal Year: 2025, Volume and Issue: 87(3)

Published: Feb. 28, 2025

ABSTRACT Facial gestures are a crucial component of primate multimodal communication. However, current methodologies for extracting facial data from video recordings labor‐intensive and prone to human subjectivity. Although automatic tools this task still in their infancy, deep learning techniques revolutionizing animal behavior research. This study explores the distinctiveness cotton‐top tamarins, quantified using markerless pose estimation algorithms. From footage captive individuals, we extracted manually labeled frames develop model that can recognize custom set landmarks positioned on face target species. The trained predicted landmark positions subsequently transformed them into distance matrices representing landmarks' spatial distributions within each frame. We employed three competitive machine classifiers assess ability automatically discriminate configurations cooccur with vocal emissions associated different behavioral contexts. Initial analysis showed correct classification rates exceeding 80%, suggesting voiced highly distinctive unvoiced ones. Our findings also demonstrated varying context specificity gestures, highest accuracy observed during yawning, social activity, resting. highlights potential advancing communication, even challenging species such as tamarins. distinguish contexts represents critical step developing automated cues raw data.

Language: Английский

Citations

0
Discrimination between the facial gestures of vocalising and non-vocalising lemurs and small apes using deep learning DOI Creative Commons
Filippo Carugati, Olivier Friard,

Elisa Protopapa

et al.

Ecological Informatics, Journal Year: 2024, Volume and Issue: unknown, P. 102847 - 102847

Published: Oct. 1, 2024

Language: Английский

Citations

1
Analysis of Cushioned Landing Strategies of Cats Based on Posture Estimation DOI Creative Commons
Li Zhang,

Liangliang Han,

Haohang Liu

et al.

Biomimetics, Journal Year: 2024, Volume and Issue: 9(11), P. 691 - 691

Published: Nov. 13, 2024

This article addresses the challenge of minimizing landing impacts for legged space robots during on-orbit operations. Inspired by agility cats, we investigate role forelimbs in process. By identifying kinematic chain cat skeleton and tracking it using animal posture estimation, derive cushioning strategy that cats use to handle impacts. The results indicate effectively transforms high-intensity into prolonged low-intensity impacts, thereby safeguarding brain internal organs. We adapt this robotic platforms through reasonable assumptions simplifications. Simulations are conducted both gravitational zero gravity environments, demonstrating optimized not only reduces ground impact prolongs duration but also suppresses robot's rebound. In gravity, enhances stable attachment target surfaces. research introduces a novel biomimetic control operations robots.

Language: Английский

Citations

0