Robotic Cell Micromanipulation for Posture Adjustment of Zebrafish Embryonic Cell

Precise posture adjustment of biological cells is a critical prerequisite for downstream micromanipulation tasks such as microinjection and cell surgery. In this study, we present a novel robotic micromanipulation strategy for in-plane orientation adjustment of zebrafish embryonic cells using an imi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of physics. Conference series Jg. 3101; H. 1; S. 12014 - 12022
Hauptverfasser: Zhang, Youchao, Zhao, Antian, Wang, Chuhan, Wang, Fanghao, Zhou, Tong, Lyu, Yining, Liu, Chuanjie, Ying, Yibin, Zhou, Mingchuan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Bristol IOP Publishing 01.09.2025
Schlagworte:
Biomedical materials
Machine learning
Micromanipulation
Robotics
Suction
Zebrafish
ISSN:1742-6588, 1742-6596
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Precise posture adjustment of biological cells is a critical prerequisite for downstream micromanipulation tasks such as microinjection and cell surgery. In this study, we present a novel robotic micromanipulation strategy for in-plane orientation adjustment of zebrafish embryonic cells using an imitation learning framework with a mixture of experts strategy (MEIL). Unlike conventional methods that rely on suction-based mechanical constraints or time-consuming manual programming, our approach leverages multimodal expert demonstration data and cross-modal alignment of visual and haptic features to enable efficient and safe cell manipulation. We constructed a primary-secondary robotic simulation environment based on PyBullet to systematically compare MEIL with traditional control, reinforcement learning (RL), and state-of-the-art imitation learning algorithms. Experimental results in simulation demonstrate that MEIL outperforms other strategies in terms of success rate, efficiency, and safety. Specifically, MEIL achieves the highest success rate, the shortest manipulation distance, and the lowest variance across tasks. This study highlights the potential of expert-driven learning frameworks for enhancing the dexterity, adaptability, and robustness of robotic cell micromanipulation in biomedical applications.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/3101/1/012014