Page 35 - Read Online

Page 22 of 38 Zhu et al. Soft Sci 2024;4:17 https://dx.doi.org/10.20517/ss.2024.05

Multimodal e-skin systems
Integration of multiple sensing capabilities (multimodal e-skins) is necessary to achieve a high level of
sensing capability [37,185,186] . Scholars have fabricated excellent multimodal e-skin systems through exquisite
material and structural design [187-190] .

Some scholars prepare multimodal e-skins with multifunctional materials, which embody two or more
[191]
sensing properties . However, it is difficult for such multimodal e-skins to achieve excellent
comprehensive performance, and it is difficult to avoid interference between various sensing characteristics.
Various sensing characteristics often cannot reflect outstanding performance at the same time.

In contrast, integrating multiple types of sensing devices with superior performance into the same e-skin
device can achieve better overall performance. In such multimodal e-skins, dealing with the interference
problem between different sensing characteristics remains an important issue , which is a major shackle
[192]
restricting the performance of the multimodal e-skin system [112,117,123,193] .

A common solution is to independently integrate multiple sensors into a single e-skin device, with each
[194]
sensor having its own dedicated independent signal transmission line . As shown in [Figure 11A], Zhao
et al. reported a flexible pressure-strain multimodal sensor capable of sensing both signals of external force
magnitude and direction simultaneously . Three-dimensional tubular graphene sponge (3D-TGS) and
[195]
spider web-like electrodes were used as pressure-sensitive and strain-sensitive modules, respectively. By
comparing the output signals, the magnitude and direction of the force can be monitored simultaneously.

Hua et al. prepared a highly stretchable conformal matrix network (SCMN) on a polyimide network
[Figure 11B], which can be fabricated into a multisensory e-skin able to simultaneously sense multiple
stimulus signals through the selection of different sensing components, including temperature (Pt-sensitive
layer), humidity (Al/PI-sensitive layer), plane strain (Constantan-sensitive layer), ultraviolet (UV, Al/ZnO-
sensitive layer), magnetism (Co/Cu multilayer-sensitive layer), pressure, and proximity, thereby realizing
[196]
multimodal sensing with a tunable range of sensing and large scalable area .
Such structures can avoid interference problems, but they also bring about the problem of complex
structure, which is disadvantageous in the development trend of high density and miniaturization. Some
scholars avoid this interference by carefully designing the structure of the e-skin systems. Lin et al. reported
[197]
a piezoelectric tactile sensor array with both pressure sensing and bending sensing capabilities
[Figure 11C]. The specially designed insulating layer and structural design of the row + column electrodes
can greatly alleviate the crosstalk problem in other sensors. A signal processor and logic algorithms were
also integrated to enable real-time sensing and differentiation of the magnitude, location, and pattern of
various external stimuli, including gentle sliding, touching, and bending. Pressure sensing and bending
sensing tests show that the proposed haptic sensor array has high sensitivity (7.7 mV·kPa ), long-term
-1
durability (80,000 cycles), and faster response time (10 ms) than human skin.

IoT-integrated and ML-enabled e-skin systems
In recent years, the integration of e-skins in the IoT has become a clear trend, as it allows for the
rationalization of e-skins with other cutting-edge technologies, such as advanced mechanics and ML
methods, to create intelligent systems, which is the key to allowing e-skin to be truly integrated into human-
machine systems. Advances in this field are expected to open up entirely new applications in areas such as
[198]
health monitoring, wearable electronics, and robotics .