Page 30 - Read Online

Yang et al. Soft Sci. 2025, 5, 46 https://dx.doi.org/10.20517/ss.2025.44 Page 27 of 39

[218]
physiological signals such as respiration of the human body in different postures .

Interactive soft robots
Interactive soft robotics represents a rapidly evolving field with significant implications for healthcare and
assistive technologies. These systems, often composed of flexible materials integrated with optical
waveguides, are designed to interact with human users safely and adaptively. In this context, researchers
have investigated various configurations and materials to enhance the functionality and integration of soft
robotics in medical and assistive devices. Notable progress has been made in developing interactive soft
robotic systems, such as robotic prosthetics and assistive robotic hands. These advancements enable more
intuitive and responsive interfaces between robots and users, improving the effectiveness of rehabilitation
and daily assistance. This section reviews recent advancements in interactive soft robotics, focusing on their
development and application in enhancing human–machine interactions (HMIs).

Unlike devices used for physiological information monitoring, some wearable devices have already realized
human–computer interactions. Various devices based on flexible optical waveguides for recognizing gesture
information and reconstructing 3D models have been developed [52,114,219-221] , providing new ideas for bionic
mechanical manufacturing. Khan et al. developed a novel strain sensor based on an optomechanical
concept, in which the device senses the magnitude of the strain it experiences through a change in the
[221]
transmittance of a flexible optical waveguide prepared from an Eco-flex/MoS material . Additionally, the
2
sensor was applied to develop a glove to achieve effective control of different parts of the bionic robot NAO.
In another bionic stretchable optical fiber sensor proposed by Li et al. [Figure 11A], optical fibers were
embedded into Eco-flex films to form Lindernia nummularifolia (LN) structures . The sensor has an
[52]
excellent response to strain and the bending angle and can translate real-time human motion signals into
control commands for HMI applications. Researchers have further developed immersive rehabilitation
training systems and remote robot-assisted motion stacking games for disabled and sick people, which have
rich application prospects. In addition to HMIs, devices that realize human–machine–environment
interactions exist. For example, to realize real-time dynamic monitoring of exposure to viruses or bacteria in
[159]
the environment, a wearable freeze-dried, cell-free (wFDCF) platform was developed by Nguyen et al. .
The reported wFDCF POF system apparel involves attaching a fabric-based module to a wearable POF
spectrometer with wireless connectivity, which is then integrated with a wireless mobile app. The app
enables continuous cloud-based data logging, signal processing, geolocation tracking, and real-time control
of various detector components from a smartphone or other connected digital device.

Another application of interactive soft robots is bionic mechanical grippers, the development of which has
exploded. Owing to the advantages of the materials and sensing ability of flexible optical waveguides, many
well-performing soft grippers and even bionic robots have been reported in recent years [222,223] . As shown in
Figure 11B, Guo et al. used AuNPs and elastomer composites to make flexible plasma fibers and proposed a
flexible optical tactile sensor . The hardness, roughness, and shape of an object can be sensed via this
[54]
sensor integrated into a robotic hand. The tactile perception of the human hand when grasping includes the
[224]
perception of static friction or slippage in addition to the perception of a positive pressure . Therefore,
artificial tactile sensing, especially friction measurement and slip detection, plays a crucial role in robot
operation [225,226] . The optical microfiber-based flexible tactile sensors inspired by finger skin designed by
[227]
Jiang et al. have force sensing and slip detection capabilities . When connected to a robotic gripper, the
soft sensor successfully distinguishes between soft and hard objects, measures the grasping force, and
detects object slip, making it suitable for robotic grasping and manipulation. In some recent studies,
inspired by topological mechanics, Pan et al. developed optical fiber knot (OFN) sensors that can be used
for slip detection and friction measurements [Figure 11C] . The twisted structure of the knot allows the
[53]
sensor to sense not only loads along the fiber direction but also slippage and triaxial forces through