Page 21 - Read Online

Jin et al. Soft Sci 2023;3:8 https://dx.doi.org/10.20517/ss.2022.34 Page 19 of 26

(2) Multi-modal tactile sensors play an essential role in robotic tactile perception. Multi-modal tactile
devices have long been reported, but a trade-off exists between structural complexity and modality amount.
It is of great significance to develop simple structured tactile sensors sensitive to multiple physical or
[119]
chemical parameters, and multi-modal tri-axis force sensor is also less reported before . In addition, since
there exists crosstalk between different modalities, decoupling technology selecting various materials and
structures or signal revising can greatly contribute to precise measuring.

(3) Advanced data processing algorithms, especially machine learning, can help analyze numerous tactile
signals. Data-drive signal processing methods such as NN, SVM or PCA can extract high-dimensional
features from large-scale datasets through intensive model training, which renders the robotics to be
intelligent enough for advanced applications, but most data processing processes are executed offline. To
achieve immediate interaction, it is necessary to do online or even in-site data processing to improve the
analyzing speed. Besides, the advanced algorithm for multi-data fusion shows great advantages in extra
feature extraction, and it can achieve more accurate sensing when blending various data of tactile, vision or
voice if the problems of data source mismatch can be addressed.

(4) High-level human-machine interactions are always the target application for robotics. Advanced data
processing methods, especially machine learning, can help extract high-dimensional and meaningful data
from simple interactive movements, which enhance the intelligence of robotics to make autonomous
decisions. Although various feedback principles such as mechanical, thermal or electrical stimulation have
been reported, it is not easy to precisely simulate the interactive parameters on site. Feedback combing
tactile, vision, voice and other information shows considerable research value. In addition, remote control
of robotics in extreme environments, namely underwater, aerospace or some disaster sites, is probably
promising HMIs applications in the future.

DECLARATIONS
Authors’ contributions
Wrote the manuscript: Jin J, Wang S, Zhang Z
Revised the manuscript: Jin J, Wang S, Mei D, Wang Y

Availability of data and materials
Not applicable.

Financial support and sponsorship
This work is supported in part by the National Natural Science Foundation of China (52175522), Key
Research and Development Program of Zhejiang Province (2022C01041), Fundamental Research Funds for
the Central Universities (2022FZZX01-06) and Science Foundation of Donghai Laboratory (DH-
2022KF01002).

Conflicts of interest
All authors declared that there are no conflicts of interest.

Ethical approval and consent to participate
Not applicable.

Consent for publication
Not applicable.