Multimodal and Multidimensional Tactile Perception for Universal Robotic Sophisticated Tasks
Research · 01.01.2026 · 9
DOI: 10.34133/research.1477 ↗Kurz gefasst
Fachartikel in der Zeitschrift Research, online erschienen am 06.11.2026. Verfasst von Qian Mao und 3 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Precision Measurement Technology and Instruments, Department of Precision Instrument, Tsinghua University, Beijing, China. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Beijing Natural Science Foundation, National Natural Science Foundation of China, Tsinghua-Toyota Joint Research Fund.
Abstract (Originalsprache)
Human hands have rich tactile senses with multimodal and multidimensional perceptions for hand dexterous manipulation. However, current robotic tactile sensory modality and dimensionality are deficient and hardly meet the harsh demands for diverse manipulation tasks. Here, we report a multimodal and multidimensional tactile sensor based on unified thermo-sensation, enabling comprehensive perceptions of contact, matter type, slip/slide, texture/scratch, temperature, and multidimensional force to meet the general-purpose perceptual requirement across diverse sophisticated robotic tasks. The sensor is compact, flexible, ultralight (0.33 g), and integrated on the robot’s fingertip. Ultrafast contact and slip detections (~5 ms) of the sensor enable instantaneous action triggering of robots to ensure agile manipulations, while its precise texture/scratch detection (at the ~20 μm scale) surpasses human touch sensitivity. The sensor also provides decoupled and high-precision measurements of 3-axis forces with fast response, low hysteresis, and low crosstalk, ensuring dexterous action control of robots. Furthermore, we propose a framework of tactile perception–control synergy, endowing robots with self-adaptive action-on-feeling like humans. Leveraging the multimodal and multidimensional tactile perception as well as the synergy with adaptive control, robots autonomously accomplish various sophisticated tasks of precise scratch inspection, fine assembly, and plug-socket insertion, etc.
Autor:innen und Institutionen
- State Key Laboratory of Precision Measurement Technology and Instruments, Department of Precision Instrument, Tsinghua University, Beijing, China.
ORCID-Profile: Yichen Qin, Zijian Liao, Rong Zhu
Förderung
- Beijing Natural Science Foundation
- National Natural Science Foundation of China
- Tsinghua-Toyota Joint Research Fund
Zitieren
APA
Mao, Q., Qin, Y., Liao, Z., & Zhu, R. (2026). Multimodal and Multidimensional Tactile Perception for Universal Robotic Sophisticated Tasks. Research, 9. https://doi.org/10.34133/research.1477
Vancouver
Mao Q, Qin Y, Liao Z, Zhu R. Multimodal and Multidimensional Tactile Perception for Universal Robotic Sophisticated Tasks. Research. 2026;9. doi:10.34133/research.1477
BibTeX
@article{Mao2026,
author = {Mao, Qian and Qin, Yichen and Liao, Zijian and Zhu, Rong},
title = {Multimodal and Multidimensional Tactile Perception for Universal Robotic Sophisticated Tasks},
journal = {Research},
year = {2026},
volume = {9},
doi = {10.34133/research.1477}
}
EUVIMED