REVIEW

www.advancedscience.com

Proximity Sensing Electronic Skin: Principles,

Characteristics, and Applications
Bingwei Wu, Ting Jiang, Zhongxiang Yu, Qihui Zhou,* Jian Jiao,* and Ming Liang Jin*

1. Introduction
The research on proximity sensing electronic skin has garnered significant
attention. This electronic skin technology enables detection without physical The skin is the largest sensory organ in
humans, serving multiple functions such
contact and holds vast application prospects in areas such as human-robot
as protection and sensation.[1–3] It func-
collaboration, human-machine interfaces, and remote monitoring. Especially tions as the body’s paramount natural
in the context of the spread of infectious diseases like COVID-19, there is a defense, characterized by exceptional at-
pressing need for non-contact detection to ensure safe and hygienic tributes such as impressive stretchability,
operations. This article comprehensively reviews the significant self-healing capabilities, excellent mechan-
ical toughness, and sensitive tactile percep-
advancements in the field of proximity sensing electronic skin technology in
tion. Structurally, the skin can be roughly di-
recent years. It covers the principles, as well as single-type proximity sensors vided into three layers: the epidermis, der-
with characteristics such as a large area, multifunctionality, strain, and mis, and subcutaneous tissue.[4] In addi-
self-healing capabilities. Additionally, it delves into the research progress of tion to these three layers, there are also
dual-type proximity sensors. Furthermore, the article places a special vellus hairs on the skin’s surface. In daily
emphasis on the widespread applications of flexible proximity sensors in life, each vellus hair serves as numerous
detectors, converting signals from foreign
human-robot collaboration, human-machine interfaces, and remote
objects that have not made physical con-
monitoring, highlighting their importance and potential value across various tact with the skin into stimuli and re-
domains. Finally, the paper provides insights into future advancements in laying them to the nerve cells beneath
flexible proximity sensor technology. the skin, thereby alerting the body. Elec-
tronic skin possesses human-like pressure
sensitivity,[5–10] pain perception,[11–13] and
temperature sensing,[14,15] and additionally,
it encompasses functions that exceed those of human skin, in-
cluding proximity detection[16–18] and chemical sensing.[19,20]
B. Wu, T. Jiang, Z. Yu, Q. Zhou The advent of flexible electronics has ushered in limitless
Heart Center, Qingdao Hiser Hospital Affiliated of Qingdao University prospects in the realm of smart electronics, encompassing flexi-
Qingdao University
Qingdao 266033, China
ble touch screens[21] and implantable devices.[22] However, flexi-
E-mail: [email protected] ble sensors hold greater value in the realization of flexible wear-
B. Wu, M. L. Jin able devices. Flexible sensing represents a crucial research focus
Institute for Future, Shandong Key Laboratory of Industrial Control in the field of flexible electronics. Based on the mode of contact,
Technology, School of Automation sensors can be divided into contact and non-contact sensors. Cur-
Qingdao University rently, the literature mainly reports on contact sensors, but non-
Qingdao 266071, China
E-mail: [email protected] contact sensors actually have advantages in specific situations.
Q. Zhou A proximity sensor device is used to detect the proximity of ob-
School of Rehabilitation Sciences and Engineering jects, enabling non-contact detection. The proximity of an object
University of Health and Rehabilitation Sciences causes a change in the sensor’s signal. Proximity sensing serves
Qingdao 266000, China as a complement to both vision and touch.[23] For touch sen-
J. Jiao sors, proximity sensors provide perception before physical con-
Peng Cheng Laboratory
Shenzhen 518055, China
tact, while for visual sensors, they compensate for limitations re-
E-mail: [email protected] lated to close-range blind spots.
The rapid development of information systems has made the
The ORCID identification number(s) for the author(s) of this article
human-machine relationship increasingly important.[24] Robots
can be found under https://doi.org/10.1002/advs.202308560 play substantial roles in our lives. In conjunction with the in-
© 2024 The Authors. Advanced Science published by Wiley-VCH GmbH. dustrial revolution and the shifting paradigm of manufacturing,
This is an open access article under the terms of the Creative Commons the evolving human-machine relationship is delineated by the 5C
Attribution License, which permits use, distribution and reproduction in journey: Coexistence, Cooperation, Collaboration, Compassion,
any medium, provided the original work is properly cited. and Coevolution.[25] Technological progress empowers robots to
DOI: 10.1002/advs.202308560 engage with humans across various domains.[26,27] Safety is the

Adv. Sci. 2024, 11, 2308560 2308560 (1 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

paramount concern in the hierarchy of human needs within the proximity sensor[35] and mutual capacitance proximity sensor.[36]
industrial context. Ensuring human safety represents the fore- As shown in Figure 2a, self-capacitance refers to the capacitance
most priority in a manufacturing environment. Traditional in- between the sensing electrode and the ground (commonly re-
dustrial robot systems necessitate robust fencing and peripheral ferred to as “GND” in the circuit). The application of an excita-
safety devices,[28] or rely on reactive protection.[29] However, fenc- tion signal to the electrode induces a changing electric field be-
ing diminishes flexibility while inflating costs and space require- tween the electrode and the ground, owing to the presence of self-
ments, and reactive protection poses the risk of harm to either the capacitance. As an object approaches, the parasitic capacitance to
robotic arm or humans. Protective mechanisms should evolve the ground undergoes alterations, enabling the detection of the
toward proactive protection.[30] Moreover, microbial pathogens object. On the other hand, as shown in Figure 2b, mutual capac-
pose a persistent threat to humans. In a span of just a few decades itance entails two electrodes: the transmitting electrode and the
in the 21st century, the world has experienced three signifi- receiving electrode. An electric field is established between the
cant pandemics: SARS, MERS, and the recent COVID-19.[31,32] two electrodes. While the majority of the energy is concentrated
COVID-19 can be transmitted through direct contact with sur- between the capacitor plates, a small portion escapes into the sur-
faces used by infected individuals.[33] In everyday life, numerous rounding space beyond the electrode plates. These escaping elec-
diseases may also spread through contact, including hand, foot, tric fields are commonly termed “fringe fields”.
and mouth disease, and dysentery, among others. Additionally, In the course of approaching the sensor using mutual capaci-
traditional rigid proximity sensors typically cannot achieve large- tance, two capacitors come into play: the mutual capacitance be-
scale and ubiquitous detection, and their compact and rigid ap- tween the two electrodes and the self-capacitance to ground, often
pearance limits their application in these areas.[34] Visual sensors referred to as fringe capacitance. When an object approaches, the
have problems with large data processing and visual blind spots. hand serves as a third electrode capable of partially intercepting
Therefore, flexible proximity sensors hold substantial develop- the fringe electric field and redirecting it to the ground within the
ment potential, whether in the realm of human-robot collabo- sensor’s sensing range. Throughout this process, charges flow
rative safety or in mitigating virus transmission through non- out from both electrodes, resulting in alterations in fringe capac-
contact methods in daily life. itance and mutual capacitance.[37]
The structure of this paper is as follows: First, the principles of Mutual capacitive proximity sensors typically employ a sand-
various types of proximity sensors are introduced, including ca- wich structure comprising electrodes and dielectric layers. The
pacitive, triboelectric, semiconductor, and magnetic sensors. The materials and structure of the device are pivotal in determining
third section focuses on the characteristics such as large area, the strength and distribution of the fringe field, consequently in-
multifunctionality, strain, and self-healing, providing a detailed fluencing the overall performance of the sensor. Research has
overview of the research progress in single-type flexible proximity demonstrated that several factors, including electrode shape and
sensors. Then, the advancements in dual-type proximity sensors parameters, significantly influence performance metrics such as
are discussed, further enriching the development direction of signal strength, sensitivity, and signal-to-noise ratio for capacitive
flexible proximity sensors. In the section highlighting the appli- proximity sensors. Presently, extensive research has been con-
cations of flexible proximity sensors, particular attention is given ducted on electrode shapes for capacitive proximity sensors, in-
to their practical use in areas such as human-robot collaboration, cluding structures like the parallel plate structure,[38] interdigi-
human-machine interfaces, and remote monitoring. These prac- tal electrode,[39–46] innovative complementary Archimedean spi-
tical examples vividly demonstrate the substantial and indispens- ral electrode,[47] circular-circular configuration,[48] and crossed
able role of flexible proximity sensors in diverse fields. Finally, in electrodes,[37,49] among others. These structural variations alter
the conclusion section, a comprehensive summary and summary the effective area between the electrodes, thereby impacting the
of the research results of this article is provided, emphasizing sensor’s sensitivity and other performance aspects. Table 2 pro-
the potential and future prospects of flexible proximity sensors vides an overview of the research results related to flexible prox-
(Figure 1). imity sensors using interdigital electrodes.

2. Transduction Mechanisms
2.2. Triboelectric Proximity Sensor
Categorized by their sensing principles and mechanisms, flexible
proximity sensors encompass various types, including capacitive, The phenomenon of triboelectricity has been a topic of interest
triboelectric, semiconductor, and magnetic field, among others. in the field of tribology for centuries. For over 2000 years, the
Table 1 provides an overview of the advantages and disadvantages universality and intriguing aspects of frictional electricity have
of various types of flexible proximity sensors. Of these, capacitive captured the attention of researchers.[50] The triboelectric nano-
sensing and triboelectric sensing have garnered substantial at- generator (TENG), first pioneered by Zhonglin Wang and his
tention in academic literature. In this section, we primarily focus team in 2012, is designed to convert minute mechanical energy
on flexible proximity sensors based on capacitive and triboelectric into electrical energy through the exploitation of the triboelec-
sensing principle. tric charging effect and electrostatic induction effect.[51] TENG
operates in four distinct modes: vertical contact mode, single-
2.1. Capacitive Proximity Sensor electrode mode, lateral sliding mode, and independent friction
layer mode [Figure 3a].[52,53] While most TENG devices are cur-
According to the capacitance detection principle, capacitive prox- rently utilized as self-powered sensors for various mechanical
imity sensors can be divided into two categories: self-capacitance movements, they can also function as proximity sensors.[54] Tang

Adv. Sci. 2024, 11, 2308560 2308560 (2 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 1. Classification and application of flexible proximity sensors. Middle: By [geralt] via Pixabay. Category identification: Reproduced under terms of
the CC-BY license.[110] Copyright 2023, The Authors, published by Wiley-VCH. Gesture recognition: Reproduced with permission.[127] Copyright 2022,
Wiley-VCH. Action recognition: Reproduced with permission.[137] Copyright 2021, American Chemical Society. Motion monitoring: Reproduced with
permission.[141] Copyright 2020, American Chemical Society. Door lock and Play games: Reproduced under terms of the CC-BY license.[138] Copyright
2022, The Authors, published by Wiley-VCH. Avoiding obstacles: Reproduced under terms of the CC-BY license.[79] Copyright 2022, The Authors, pub-
lished by Wiley-VCH. Contour recognition and Normal interaction: Reproduced under terms of the CC-BY license.[108] Copyright 2022, The Authors,
published by AAAS. Falling detection and Direction determination: Reproduced with permission.[140] Copyright 2021, Elsevier. Location determination:
Reproduced under terms of the CC-BY license.[37] Copyright 2020, American Chemical Society.

et al. first proposed the concept of non-contact induction based picted in Figure 3b. In the initial state, when two objects are sep-
on the participation of the human body in frictional electric self- arated, the lower surface of one object induces a charge, such as
power supply.[55] positive charge, while the upper object induces a negative charge.
The operational principle of triboelectric proximity sensors, During the approach phase, as the upper object nears, the lower
founded on triboelectricity and electrostatic induction, is de- object undergoes electrostatic induction, accumulating extra

Table 1. Comparison of different types of flexible proximity sensors.

Type Advantages Disadvantages

Capacitive Low cost, flexible structural design, and fast dynamic response Parasitic capacitance and susceptibility to interference from the
surrounding environment (such as electromagnetic interference)
Triboelectric Simple structure, self-power Usually output transient signal, charge attenuation, susceptible to
environmental interference
Semiconductor Strong ability to detect insulators, high accuracy, and simple structure Restricted object detection, usually detecting insulators
Magnetic Not affected by non-magnetic objects, allowing for accurate signal Restricted object detection, detecting magnetic objects
acquisition

Adv. Sci. 2024, 11, 2308560 2308560 (3 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 2. Two types of capacitive proximity sensor principles. a) Self-capacitive sensor. b) Mutual capacitive sensor.

Table 2. Summary of the interdigital electrode in proximity sensors.

Electrode structure Electrode material Tested object Detection distance(mm) Refs.

Interdigital Electrode LM Human hand 50 [46]

Interdigital Electrode ITO Human hand 200 [39]
Annular Interdigital Electrode Cu Human hand NA [45]
Interdigital Electrode Au Acrylic acid 1.4 [44]
Interdigital Electrode EGaIn Human hand 150 [42]
Interdigital Electrode CNTs Force gauge probe 2 [41]
Spider Web Electrode EGaIn Human hand 5 [43]
Interdigital Electrode Cu Human hand 100 [40]

positive charges to neutralize the charges on the upper ob- lower electrode to counterbalance the positive charges, lead-
ject. This process results in the flow of electrons through the ing to a current in the opposite direction to that during the
electrode toward the ground, generating a current signal. In approach phase. In daily life, owing to the triboelectric effect
the phase of electrostatic equilibrium, as the objects draw between footwear and the ground, the human body retains the
nearer, the charges produced by the upper object are entirely electrical charge. Moreover, negative charges may accumulate
neutralized by those generated by the lower object, resulting at the fingertips due to the tip effect. Consequently, via elec-
in electrostatic equilibrium. In the departure phase, as the trostatic induction, flexible sensors based on the triboelectric
upper object retreats from the sensor, the positive charges on effect can detect the proximity of objects, including a human
the electrode diminish, and free electrons are attracted to the hand.[55]

Figure 3. Triboelectric nanogenerators and triboelectric proximity sensors. a) TENG four operating modes. Reproduced with permission.[52] Copyright
2022, Wiley-VCH. b) Principle of triboelectric proximity sensor. Reproduced with permission.[55] Copyright 2019, Wiley-VCH.

Adv. Sci. 2024, 11, 2308560 2308560 (4 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 4. Semiconductor and magnetic proximity sensors. a) Structural diagram of flexible proximity sensor based on ultra-thin rubrene micro nano
single crystal. b) Schematic diagram of flexible proximity sensor based on ultra-thin rubrene micro nano single crystal. Reproduced with permission.[59]
Copyright 2018, American Chemical Society. c) m-MEMS platform. d) Daisy petals decorated with permanent magnets. e) Schematic diagram of the
evolution of magnetic field intensity at the sensor position: the superposition of the builtin magnetic field provided by a compatible permanent magnet
(orange shaded area) and an external magnetic field source (blue shaded area). Reproduced under terms of the CC-BY license.[63] Copyright 2019, The
Authors, published by Springer Nature.

However, proximity sensors based on triboelectricity also face area. When the diameter of the microsphere is 6.536 μm, the out-
challenges similar to those encountered by capacitive proxim- put charge density can reach 128 μC m−2 , and the instantaneous
ity sensors—a decrease in output performance as the distance output power can reach 200 μW cm−2 .[58]
from the object increases. To enhance output performance, a se-
ries of methods are typically employed, including material op-
timization, structural design, and the use of ultra-thin dielec- 2.3. Semiconductor and Magnetic Proximity Sensors
tric layers, among others, to increase surface charge density.
These measures contribute to the improvement of the sensor’s Flexible semiconductor proximity sensors differ from traditional
performance.[56] For example, Chen and colleagues utilized pro- capacitive sensors. These sensors have electrodes located at the
cesses such as oxygen plasma etching, ECR sputtering, and two ends of semiconductor material, creating a two-terminal pla-
through-filter ion etching to establish a secondary micro-nano nar device structure. Wang et al. pioneered the use of a flexible
composite structure on the sensing layer’s surface. This approach rubrene crystal as a sensing element, achieving a sensor with
substantially increased the specific surface area and enhanced proximity sensing capability, as shown in Figure 4a, where gold
the sensor’s output signal by ≈200%.[57] Similarly, Zhang et al. electrodes are located at both ends of the micro-sized organic sin-
harnessed the high specific surface area and structural stabil- gle crystal. When a charged object approaches the sensor, the
ity of spherical multiple physical networks to produce compos- electric field produced by the charges, due to electrostatic in-
ite films of polyvinylidene fluoride (PVDF) @Mxene (Ti3 C2 Tx ) duction, serves as the gate voltage, capable of modulating the
with multiple physical net-work structures for use as triboelec- charge carriers within the organic semiconductor. This modu-
tric materials. Ti3 C2 Tx material and spherical multiple physics lation leads to a current alteration in the organic device, as de-
network structure play a great role in improving TENG output. picted in Figure 4b.[59] Most sensors based on this principle em-
The H and O atoms of Ti3 C2 Tx and the H atoms in the PVDF ploy semiconductor materials as the sensing layer, such as DPP-
chain form dipoles. The capacitor network formed by the dipoles DTT,[60] rGO film,[61] and DNTT.[62] These sensors are primarily
and porous microstructure improves the output capability of the used for detecting non-metallic objects with static electricity.
TENG. The size of the porous microspheres in the spherical mul- Recent research suggests that magnetic-based proximity sen-
tiple physical network structure can change the effective contact sors can address some of the challenges faced by capacitive

Adv. Sci. 2024, 11, 2308560 2308560 (5 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

proximity sensors, such as issues related to interference from un- medium, which were subsequently used in the assembly of both
related objects and the identification of signal sources. Magnetic- single pixel touch sensors and multi-channel sensors. The struc-
based flexible proximity sensors frequently employ principles ture and manufacturing process for these sensors are elucidated
such as the Hall effect, magneto-piezoresistive effect, and giant in Figure 5c, encompassing precursor preparation, laser pattern-
magnetoresistance effect for object proximity detection. The hall ing, and assembly, all of which can be completed within an
effect refers to the phenomenon in a conductor where, when an hour. The definition of patterns and annealing take place dur-
electric current flows, an electric field is generated perpendic- ing the repetitive laser scanning process. While the single-pixel
ular to both the current direction and the magnetic field direc- proximity sensor achieves a detection distance of up to 20 cm
tion, resulting in a lateral voltage. Magnetoresistance effect is the [Figure 5d], the multi-channel sensor exhibits a relatively reduced
phenomenon in which the resistance of certain metals or semi- detection distance and requires further optimization.[67] You et al.
conductors changes with the variation of an applied magnetic applied a coating of graphene oxide (GO)-doped polyurethane
field. Giant magnetoresistance effect refers to the phenomenon (PU) nanofibers to the surface of Ni-coated cotton yarn using
where the resistivity of magnetic materials undergoes a signif- electrospinning. These coated yarns were subsequently wound
icant change when subjected to an external magnetic field as around elastic wires and woven into a vast expanse of elec-
compared to when there is no external magnetic field. Ge et al. tronic skin [Figure 5e]. The spiral structure of the GO-doped
incorporated flexible permanent magnet NdFeB particles into PU nanofiber/Ni-coated cotton yarn elastic composite yarn (GO-
PDMS rubber, forming a magnetic micro-electromechanical sys- PNF/NiCY elastic composite yarn) imparts stretchability to the
tem (m-MEMS) as depicted in Figure 4c. In Figure 4d, mag- sensor. This particular sensor can detect objects at a distance of
netic material opposite to the flexible permanent magnet is ap- ≈10 cm [Figure 5f].[68]
plied to the object to induce specific object detection. Figure 4e Leveraging the benefits of waste reduction and operational
illustrates the interaction process, wherein the external mag- simplicity, Li and his research team opted for inkjet printing
netic field compensates for the built-in magnetic field when to fabricate proximity sensors utilizing silver nanoparticle ink.
in proximity, leading to an increase in GMR (giant magnetore- The sensor structure comprises 100 grids.[69] On the other hand,
sistance) resistance. Upon contact, the magnetic field at the Yang Wei et al. utilized a dispenser printing method to create
GMR sensor’s location increases, causing a reduction in GMR sensors on 100% polyester fabric. Figure 6a illustrates sensors
resistance.[63] with three distinct structures, and simulations were conducted
to analyze the variation of capacitance with distance. Experimen-
tal research revealed that, in the loop design of capacitive prox-
3. Single Type Proximity Sensor imity sensors, when the ratio of track width (L1) to total elec-
A single type of proximity sensor uses a single proximity sens- trode width (L) is 1/16, the usage of conductive ink was reduced
ing principle to detect objects. For example, capacitive proximity by 76%, while still achieving 90% of the maximum detection
sensors and triboelectric proximity sensors. In this section, we distance.[70] The choice of this ratio saves resources without af-
concentrate on the research advancements in single type proxim- fecting the performance of the sensor, achieving a win-win effect.
ity sensors from four dimensions: large area, multifunctionality, Additionally, it provides an idea for designing multimodal flexible
strain, and self-healing. sensors in the future, where the space occupied by proximity sen-
sors can be limited to just the boundaries of the sensor. Dace Gao
et al. used inkjet printing to print lithium chloride ink to form a
3.1. Large Area coplanar interlocking diamond structure and fabricated sensors
with transparent, elastic, and strain-insensitive by 3D printing
One of the key characteristics of the skin is its substantial sur- intermediate cross-linked PDMS spacers [Figure 6b,c]. The sen-
face area. The skin serves as the boundary between the human sor’s detection distance is ≈100 mm.[71] Meanwhile, Leon Yeon
body and the external environment, constituting the largest or- Wei Loh et al. utilized a commercial fused deposition modeling
gan in the human body and making up roughly 15% of the to- (FDM) multi-material 3D printer to 3D print conductive carbon
tal body mass.[64,65] When considering the utilization of flexible black thermoplastic polyurethane (PI-ETPU) electrodes and in-
proximity sensors in applications such as human-robot collab- sulating thermoplastic polyurethane (TPU) [Figure 6d]. During
oration, the imperative of large-scale production becomes evi- the printing process, the dual-nozzle printer (BCN3D) intermit-
dent. Currently, diverse methods are available for the fabrica- tently changed the nozzles. The sensor’s structure is depicted in
tion of electronic skin, encompassing techniques such as pho- Figure 6e. Figure 6f shows the potential for customizing Pois-
tolithography and solution-based processes. In 2017, Sarwar and son’s ratio by printing soft capacitive sensor arrays associated
colleagues implemented a mold-peel-bond technique, similar to with three different metamaterial designs, which are in the aux-
the fabrication process for microfluidic devices, for the devel- etic, neutral, and positive ranges, respectively.[72] Our bodies or
opment of integrated sensors. The manufacturing process, de- humanoid robots possess 3D, irregularly shaped curves. Many
picted in Figure 5a, involves pouring PDMS into a mold, followed existing processes for electronic skin are currently constrained
by peeling and bonding, and ultimately injecting ion-conductive by their planar nature. To address this issue, one solution is to
hydrogel (consisting of acrylamide and NaCl) into the circular create flat electronic skin using conventional or flat printing tech-
disk and ring structure of PDMS. This sensor demonstrates a niques and then apply it to a 3D surface. An alternative approach
detection distance of 12 cm [Figure 5b].[66] Li and his research involves 3D printing technology, while the third method utilizes
team employed laser technology to create copper electrode pat- conventional large-area coating techniques such as brush coating
terns on a polyimide film and a polydimethylsiloxane (PDMS) or spray coating to apply sensing materials to 3D surfaces.[73]

Adv. Sci. 2024, 11, 2308560 2308560 (6 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 5. Manufacturing process based on microfluidics, laser patterning, and electrospinning for large-area flexible proximity sensors. a) Sensor man-
ufacturing process. b) Relative variation of capacitance near fingers. Reproduced under terms of the CC-BY license.[66] Copyright 2017, published by
AAAS. c) Manufacturing process of microscale components of sensors. d) Variation of relative capacitance of proximity sensors with finger distance.
Reproduced with permission.[67] Copyright 2020, Wiley-VCH. e) Schematic diagram of electronic fabric manufacturing. f) Proximity response of elastic
composite yarn. Reproduced with permission.[68] Copyright 2018, Royal Society of Chemistry.

3.2. Multifunctionality pressure-into a single sensor, as illustrated in Figure 7a. The prox-
imity sensing capability of this sensor relies on self-capacitive de-
One of the remarkable features of human skin is its multifunc- tection, with the range determined by the size of the surround-
tionality, encompassing a diverse array of sensory receptors that ing boundaries. Therefore, silver ink is printed around the bot-
serve to detect pressure, strain, temperature, pain, vibration, and tom sensor as a proximity sensor. This sensor consists of 16
more.[74] square electrodes, forming a pressure sensor, and temperature
Electronic skin, also known as e-skin, is an electronic device sensors composed of thermistors interconnected in a serpen-
or system designed to emulate the functionality of human skin. tine pattern, as depicted in Figure 7b. The utilization of print-
E-skin should be considered as a versatile sensor capable of ing technology lays the groundwork for future large-scale appli-
bridging the gap between humans, robots, and their surround- cations. This sensor can detect the proximity distance of a per-
ing environment. Recently, there has been a significant focus on son’s hand, ≈110 mm, as shown in Figure 7c.[77] Moreover, the
the research and development of versatile flexible sensors with independent design of each sensor within this structure mini-
the capability to detect multiple external stimuli through vari- mizes cross-interference. Typically, pressure sensor electrodes or
ous modes.[75] These flexible, multifunctional sensors hold a piv- electrolytes are connected using materials with a high Young’s
otal role in the domain of electronic skin, facilitating seamless modulus, leading to the propagation of deformation from one
interaction between robots, humans, and their environment.[76] capacitive pressure sensor to another, causing interference.[78]
Luo and his team harnessed the power of printing technology to Li et al. developed a multimodal sensing electronic skin, as de-
integrate three essential functions-proximity, temperature, and picted in Figure 7d,e, which can detect proximity, contact force,

Adv. Sci. 2024, 11, 2308560 2308560 (7 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 6. Large area flexible proximity sensor based on printing. a) Filled, spiral and loop proximity sensors and relative capacitance change with distance.
Reproduced with permission.[70] Copyright 2016, Elsevier. b) Schematic diagram of ink-jet printing process and ion gel formation. c) Manufacturing
process flow of stripe field capacitive touch sensing matrix with interlocking diamond electrode layout. Reproduced under terms of the CC-BY license.[71]
Copyright 2020, The Authors, published by Wiley-VCH. d) FDM 3D printing process diagram of sensor array manufacturing. e) Sensor structure diagram.
f) Negative range, neutral range, positive range and uniform planar sample. Reproduced with permission.[72] Copyright 2021, Wiley-VCH.

contact force position, and temperature. They also proposed an field, and proximity sensing, as shown in Figure 8b.[85] However,
intelligent robot control strategy to enable human-machine in- the fabrication process for this sensor is intricate and time-
teraction. The electronic skin employs parallel plate capacitors consuming, resulting in elevated costs and limitations regarding
for proximity sensing, a pressure-thermal sensor consisting of high transparency and large-scale applications. On the other
a mixture of PDMS, pores, and silver nanoparticles for contact hand, Hong Seok Jo and colleagues engineered a stretchable and
force and position sensing, and a platinum/chromium (Pt/Cr) transparent multifunctional sensor using supersonic-sprayed
sensor for temperature detection. By utilizing a neural network, silver nanowires (AgNW). This sensor is capable of monitoring
the sensor can determine contact force and contact position pressure, proximity, strain, temperature, heating, and offering
coordinates.[79] UV protection, as depicted in Figure 8c. The manufacturing
Typically, multifunctional sensors are designed as pressure- process and the sensor’s structure are delineated in Figure 8d,
proximity sensors[80–82] or pressure-proximity-humidity incorporating a silver nanowire transparent conducting film
sensors,[83,84] offering capabilities limited to two or three (AgNW TCF) to create proximity, pressure, and temperature
functions. However, there have been few reports on single sen- sensors. During production, the AgNW TCF in the central
sors capable of detecting multiple stimuli. Hua and his research segment is wholly severed and employed as a proximity sensor,
team introduced a highly stretchable and con-formable matrix while the AgNW TCF with a single notch on one side serves as
network (SCMN) composed of 100 sensory nodes interconnected a temperature sensor. Experimental findings revealed that as
by meandering wires, as depicted in Figure 8a. This sensor mim- the number of supersonic spray scans increases, the sensor’s
ics the functionalities of the human somatosensory system, transparency gradually diminishes, along with a reduction in
encompassing capabilities to sense strain, pressure, tempera- sensitivity to pressure and proximity.[86] Recently, Hong Seok Jo
ture, and humidity. Remarkably, it incorporates functionalities and colleagues introduced a sensor prepared using the supercrit-
beyond those of the human body, including light, magnetic ical spray method, endowed with the ability to simultaneously

Adv. Sci. 2024, 11, 2308560 2308560 (8 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 7. Multifunctional flexible proximity sensors with four or fewer functions. a) Exploded view of proximity, temperature and pressure sensors. b)
Bottom electrode layer and its circuit schematic diagram under temperature detection mode. c) Changes in self-capacitance when aluminum plates,
wooden hands, and human hands approach. Reproduced with permission.[77] Copyright 2021, Wiley-VCH. d) Structural diagram of multifunctional
electronic skin. e) Experimental response of piezoelectric thermal sensor to roof offset and tilt. Reproduced under terms of the CC-BY license.[79]
Copyright 2022, The Authors, published by Wiley-VCH.

detect proximity, strain, pressure, temperature, humidity, and stretching state when stretching alters the sensing signal and
ultraviolet (UV) light, vividly portrayed in Figure 8e,f. The design then determine the pressure or proximity corresponding to the
incorporates a multi-layer structure to prevent interference output signal of the sensor. Zhao and his colleagues devel-
between different functions. Proximity sensing is achieved oped a hybrid electrode based on a transparent polymer (poly
through fringe capacitance, offering non-contact 3D scanning (3,4-ethylenedioxythiophene): poly (styrene sulfonate) (PEDOT:
capabilities. Notably, the sensor demonstrates an ability to detect PSS)/single-walled carbon nanotubes (SWCNT)), embedded in a
objects at a distance of ≈100 mm.[87] By amalgamating the bottom PEDOT: PSS/SWCNT hybrid electrode array of PDMS
detection of six external stimuli within a single sensor, this packaging layer, and a sensor composed of a PDMS dielectric
advancement simplifies the layout of sensors in subsequent layer [Figure 9a]. Under no strain, charge transport is dependent
multi-modal sensor systems. on PEDOT: PSS, whereas under significant strains, charge trans-
port relies on SWCNT [Figure 9b]. As shown in Figure 9c, the
resistance signal is influenced only by stretching and not by pres-
3.3. Strain sure or proximity. The sensor can initially ascertain the bending
state through changes in resistance, and subsequently accurately
One of the typical characteristics of human skin is its excel- determine the pressure and distance based on the capacitance’s
lent adaptability to strain. Skin exhibits remarkable flexibility, dependence on object pressure and proximity in a fixed bending
allowing it to bend, stretch, or compress under external pres- state.[89] Ding et al. synthesized hydrogel fibers using acrylamide
sures or deformations. Although stretchable flexible proximity (AAm) and sodium alginate as raw materials. These fibers were
sensors have been developed,[88] most of the reported responses subsequently encapsulated in an Ecoflex layer, with two fibers
of flexible proximity sensors are still measured under flat and vertically intersecting to create a sensor [Figure 9d]. This sen-
strain-free conditions, which, to some extent, limits their per- sor displays outstanding attributes, with virtually no alteration
formance in real-world applications. In order to better enhance in proximity sensing performance under strain conditions of 0–
the application potential of flexible proximity sensors, measure- 100% and minimal influence on proximity sensing output signal
ment experiments should not only be conducted under flat con- under stretching conditions of 0–90% [Figure 9e]. In the case of
ditions, but also in bending and other states, and even achieve stretching, the proximity response of the sensor remains essen-
object positioning function in these states. This advancement tially unchanged, and this can be attributed to the essentially un-
will enhance the performance and applicability of flexible sen- changed values of the sensor’s initial capacitance, the capacitance
sors in various scenarios and pave the way for their broader of the object, and the distance between the object and the sensor.
use in practical applications. One approach is to determine the As the distance (d) and the contact area (s) of the capacitor both

Adv. Sci. 2024, 11, 2308560 2308560 (9 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 8. Multifunctional flexible proximity sensors with more than four functions. a) Layout schematic of SCMN-an integrated sensor array with eight
functionalities. b) Multifunctional sensing. Reproduced under terms of the CC-BY license.[85] Copyright 2018, The Authors, published by Springer Nature.
c) Figure displaying a transparent flexible multifunctional sensor for detecting environmental stimuli. d) Multifunctional sensor assembly diagram.
Reproduced with permission.[86] Copyright 2019, American Chemical Society. e) Manufacturing process of multi-functional soft sensors with pressure,
bending, proximity, heat, light and humidity sensing functions. f) Overall structure of the sensor. Reproduced under terms of the CC-BY license.[87]
Copyright 2022, The Authors, published by Springer Nature.

decrease to a similar extent during the stretching process, this may even fail due to factors such as fatigue, corrosion, or physical
may result in the sensor’s initial capacitance remaining essen- damage.[91] Introducing self-healing materials can provide sen-
tially unchanged. Additionally, the sensor features a broad detec- sors with self-healing properties, thereby extending their lifespan
tion range of 18 cm and is devoid of crosstalk between pressure and offering important advantages in the field of sensors.[92,93]
and proximity sensing.[90] Chen and colleagues used inorganic nanoclay and lithium chlo-
ride to synthesize a PSBMA-Clay-LiCl (PSCL) hydrogel through
in situ polymerization of the polyelectrolyte. They then sand-
3.4. Self-Healing wiched this hydrogel between VHB sheets to create a sensor with
both mechanical and electrical self-healing capabilities, as illus-
One of the skin’s most remarkable features is its self-healing abil- trated in Figure 10a. The self-healing mechanism of this sen-
ity, which enables it to repair damage. Human skin possesses suf- sor is delineated in Figure 10b. When two freshly cut hydrogel
ficient strength to self-heal when subjected to injuries. In con- pieces come into contact, an initial physical connection is estab-
trast, artificial electronic devices can deteriorate over time and lished, followed by an accelerated healing process facilitated by

Adv. Sci. 2024, 11, 2308560 2308560 (10 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 9. Strain performance of proximity sensor. a) Based on PEDOT: PSS/SWCNT proximity sensor structure. b) Electrical conductivity diagram before
and after stretching. c) Pressure/proximity sensing under distinguishable strain. Reproduced with permission.[89] Copyright 2020, American Chemical
Society. d) Schematic diagram of a capacitive fiber optic bimodal sensor and its manufacturing process. e) Proximity response during tensile strain
process. Reproduced with permission.[90] Copyright 2022, The Royal Society of Chemistry.

ion interactions. This sensor is capable of detecting objects at a 4. Dual Type Proximity Sensor
range of ≈160 mm, as depicted in Figure 10c.[94] Guo et al. and
their collaborators proposed a new structure. They embedded Currently, many studies are focused on single types of proximity
3D electrodes into the new self-healing foam material (AiFoam) sensors, but they often fail to meet the requirements or achieve
[Figure 10d]. The robust dipole-dipole interaction between sur- the desired results in certain situations. The integration of di-
factant molecules and the cross-linked polymer network effec- verse proximity sensor types offers the potential to attain a spec-
tively traps the surfactant within the underlying elastomer, en- trum of intricate functions and further enhance sensor perfor-
abling the foam material to swiftly self-heal and regain function- mance. While existing systems combine flexible proximity sen-
ality after sustaining mechanical damage [Figure 10e]. By em- sors with various sensor types such as triboelectric and piezore-
bedding a 3D electrode into a foam containing μNi particles, it sistive sensors,[97] or triboelectric and piezoelectric sensors,[98]
can form a capacitive sensor and detect objects within ≈12 cm they predominantly rely on a singular proximity sensing princi-
[Figure 10f].[95] Yuan and colleagues introduced a self-powered ple. However, by effectively harnessing the principles and charac-
non-contact triboelectric nanogenerator (NTENG) based on elec- teristics of various flexible proximity sensors, there lies an oppor-
trostatic induction and the triboelectric effect. This device in- tunity for substantial improvement in proximity sensing perfor-
corporates graphene/shear stiffening gel (SSG) electrodes and a mance. This integration and optimization harbor the potential
shear stiffening elastomer (SSE) shell. Notably, the sensor pos- to drive notable advancements in proximity sensor technology,
sesses the capability to self-heal, dissipate, and absorb impact thereby bolstering its capabilities across a multitude of applica-
energy. It achieves self-healing by leveraging cross-links and tions. The inductive mode exhibits the capability to distinguish
dynamic hydrogen bonds within the polymer. These dynamic metal materials based on their distinct permeabilities, while the
bonds can break during cutting but re-solidify during the heal- capacitive mode discriminates non-metallic materials grounded
ing process, all without the need for external stimuli. When in their dielectric constants. In daily life, insulating materials like
the NTENG is cut and subsequently healed, its output voltage, hair, paper, and plastic sheets often accumulate charges imper-
current, and charge remain consistent with the pre-cut state, ceptibly, and flexible semiconductor proximity sensors excel at
with the healing process taking just one minute, as depicted in detecting these insulating materials. By integrating flexible ca-
Figure 10g,h. Additionally, this sensor can detect FEP films of pacitive sensors with semiconductor sensors, it is possible to ad-
≈8 cm in size.[96] dress the limitations associated with the tested materials and

Adv. Sci. 2024, 11, 2308560 2308560 (11 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 10. Self-healing performance of proximity sensor. a) Schematic diagram of synthesis, properties and applications of PSCL hydrogel. b) Self-healing
mechanism of PSCL hydrogel. c) Relative capacitance changes with finger proximity distance. Reproduced with permission.[94] Copyright 2022, Elsevier.
d) Sensor structure diagram based on AiFoam. e) Self-healing performance of AiFoam. f) Capacitance change response of distance between finger and
sensor (with 25 μ Volume percentage of Ni). Reproduced under terms of the CC-BY license.[95] Copyright 2020, The Authors, published by Springer
Nature. g) The assembly and self-healing mechanism of NTENG. h) Digital photos of cutting and repairing units, as well as corresponding non-contact
electrical performance. Reproduced with permission.[96] Copyright 2021, Elsevier.

simultaneously detect conductors and insulators.[99,100] Zhao tion of capacitance and resistance engenders more efficient col-
et al. used andalusite crystal material as the sensing layer of or- laborative object proximity detection [Figure 11d].[101] Addition-
ganic semiconductor proximity sensors and gold electrode as the ally, sensors grounded in other principles can complement capac-
electrode of the sensor [Figure 11a]. Experimental results show itive proximity sensors, for instance, the integration of capacitive
that the capacitive signal and the current signal have different sensors with inductive sensors enabling the detection of various
detection capabilities and response characteristics. The capaci- materials.[102]
tive signal has a detection range of 0.5 to 6 centimeters, while the
current signal’s detection range extends from 0.5 to 10 centime- 5. Application
ters. These differences are due to the different detection princi-
ples. Figure 11b illustrates the response of the capacitive proxim- In recent years, proximity sensors have seen widespread adop-
ity sensor to a steel ruler, while the organic semiconductor sen- tion across various domains of life. For instance, capacitive prox-
sor responds to a charged rubber rod.[100] Furthermore, combin- imity sensors have been utilized in tasks such as liquid level
ing capacitive and resistive sensing units allows for the full uti- detection,[103] assessing the aging of composite insulators,[104]
lization of the high sensitivity and rapid response characteristics material identification,[105] and measuring automotive strain.
of capacitive proximity sensors at close distances. This approach The rapid advancements in artificial intelligence technology have
also lever-ages the advantages of resistive temperature sensors driven the development of functional sensors, particularly within
in detecting objects over larger distances. Qiu et al. adopted the the realm of smart sensing technology. This technology empow-
approach of embedding a capacitive sensor with coplanar elec- ers both humans and robots to swiftly and accurately gather
trodes and an interdigital temperature sensor with graphene information from their external environments, reshaping peo-
nanosheets as the temperature sensing material into polyimide ple’s lives and providing convenience across various domains.
to create a dual-type proximity sensor [Figure 11c]. This combina- Research into electronic skin not only contributes to enhancing

Adv. Sci. 2024, 11, 2308560 2308560 (12 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 11. Dual type proximity sensor. a) Structural diagram of flexible conformal composite proximity sensor. b) Proximity response of dual types
of sensors to steel rulers and charged rubber rods. Reproduced with permission.[100] Copyright 2021, Elsevier. c) Structure of capacitive and resistive
proximity sensors. d) Changes in capacitance (C-mode) and resistance (R-mode) of a 60 °C constant temperature plastic block at different distances
from the sensor during proximity induction testing. Reproduced with permission.[101] Copyright 2015, IOP Publishing Ltd.

the intelligence of robots but also ensures safety in human-robot motion tracking.[107] In the depths of the ocean, sharks employ
collaborations. The integration of flexible proximity sensors into electrical reception strategies for remote sensing, and machine
smart environments has introduced a captivating dimension to learning holds the promise of extending functionality with min-
life. For instance, in everyday life and entertainment, these sen- imal sensor deployment. As depicted in Figure 12a, based on
sors find application in gesture detection and interaction. these principles, Guo et al. developed an artificial electrorecep-
tor that incorporates electret-embedded inorganic nanoparticles
SiO2 , ionic hydrogel, and PDMS elastomer. Figure 12b illustrates
5.1. Human–Robot Collaboration this sensor attached to a robot’s surface, enabling interaction
through the electrical charge naturally carried by humans. Utiliz-
With the relentless progress of science, technology, and the rapid ing the data-to-color approach, the potential distribution is trans-
advancements in artificial intelligence, humanity has ushered lated into a 2D image, facilitating enhanced feature extraction by
in the era of intelligence. In recent times, the swift advance- the convolutional neural network. Ultimately, the VGG-16 archi-
ment of mobile internet and intelligent gadgets has substan- tecture, known as the Visual Geometry Group, provided the foun-
tially stimulated the investigation of smart sensing technolo- dational framework for the convolutional neural network, achiev-
gies in areas like human-robot interaction, machine learning, ing recognition of spheres, cones, ellipses, and human faces with
and portable devices. Cutting-edge flexible sensor technology for an impressive accuracy rate of 97% [Figure 12c].[108] Gege Ma
robots is achieving remarkable breakthroughs, contributing sig- and colleagues have pioneered a 4D sensing technique in con-
nificantly to the emergence of the intelligent era. While the pre- junction with a spatio-temporal fully variational algorithm. This
cise origin of the term “Fifth Industrial Revolution” (5IR) re- innovation employs 4D mapping to enable object tracking, sub-
mains somewhat un-clear, it emphasizes human-robot collabo- sequently facilitating the development of predictive models to
ration over competition.[106] Security is a paramount concern in control robot movement based on obstacle trajectories. This is
the future of human-robot collaboration. Furthermore, vision- particularly valuable for robots operating in unpredictable envi-
based sensors may not function optimally in certain scenarios. ronments and seeking to avoid collisions.[109] Ge et al. integrated
To address these challenges, the application of proximity sen- their multifunctional electronic skin into a commercial robotic
sors assumes particular significance. Enhancing the safety of arm and found that insulating objects did not trigger an immedi-
human-computer interactions in industrial environments can ate response when approaching the robotic arm. This is because
be achieved through distance detection, surface contouring, and insulators cannot modify capacitance as effectively as conductors,

Adv. Sci. 2024, 11, 2308560 2308560 (13 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 12. Flexible proximity sensors applied in human-robot collaboration. a) Shark-inspired electroreceptor and artificial electroreceptor. b) Demon-
stration of manipulating robot arm when an adult is approaching. c) Target recognition of cones, cylinders, ellipses, and other objects based on electric
receptor systems. Reproduced under terms of the CC-BY license.[108] Copyright 2022, The Authors, published by AAAS. d) Improving the approaching
process performance of the multi-functional electronic skin system through machine learning. Reproduced under terms of the CC-BY license.[110] Copy-
right 2023, The Authors, published by Wiley-VCH. e) Schematic diagram of robot feedback control system for collision detection and safety control. f)
Applying electronic skin on robot arms to achieve human-machine interaction and teaching. Reproduced under terms of the CC-BY license.[79] Copyright
2022, The Authors, published by Wiley-VCH.

causing the proximity sensor’s output to fall short of the active robot will trigger its safety control to reduce damage. The au-
safety control threshold. To address this issue, a Long Short Term thor assumes that the robotic arm follows the spring damping
Memory (LSTM) network for object classification was used. The dynamic mechanism and proposes the following interactive con-
network achieved exceptional accuracy rates of 99.5%, 95.5%, and trol laws to achieve human-robot interaction of the robotic arm:
91.5% [Figure 12d]. Through this network, the threshold was dy-
namically optimized in real-time, thereby enhancing the effec-
tiveness and precision of active safety control.[110] Li et al. in- Δ𝜃 = −Kk e𝜏 − Kd 𝜃̇ (1)
stalled their designed electronic skin on the six degrees of free-
dom manipulator to achieve collision detection of robots. When where Δ𝜽 is the angle increment of the robot, kk and kd are
the detected signal exceeds the predefined safety threshold, the the coefficients associated with the stiffness and damping of the

Adv. Sci. 2024, 11, 2308560 2308560 (14 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

kinetic system, respectively, e𝝉 = 𝝉 − 𝝉 des is torque deviation, achieving the recognition of 16 different hand gestures with
𝝉 is the interaction torque on the robot detected by the e-skin, an impressive average accuracy rate of 96.5% [Figure 13b,c].[127]
𝝉 des is a desired acting torque, and 𝜃̇ is the robot angular velocity. Wearable systems find extensive use across various applications,
When the human hand is close to the robot arm, the robot arm notably in health monitoring, virtual reality,[128] and augmented
will immediately stop moving and enter the interaction mode. reality.[129,130] Wearable gesture detection has been extensively re-
When the human hand touch is detected, the robot immedi- searched, primarily based on resistive sensors,[131,132] electromyo-
ately enters the human-robot interaction mode, and the control graphy (EMG),[133–135] and optical devices.[136] However, the ma-
flowchart is shown in Figure 12e. By mounting it on the robotic jority of these studies focus on single-modal systems. Pan et al.
arm, it can avoid obstacles and teach the robot to handwrite words developed a flexible wearable system with a hybrid design to cap-
and perform other operations [Figure 12f].[79] Seung Jae Moon ture both spatial and temporal data from finger movements and
et al. employed dual-type proximity sensors developed by Aidin hand positions. As shown in Figure 13d, peaks represent sud-
Robotics to handle obstacles through admittance control and dis- den tactile movements, while more gradual changes indicate ap-
tance measurement techniques.[111] proaching movements. The constructed bimodal sensor platform
In the enterprise setting, MRK System utilizes capacitive achieves high-precision recognition of static and dynamic ges-
skin to implement collision avoidance functionality. Addition- tures (such as “Thank you”, “I’ll say”, and “Everyday”) through
ally, Aidin Robotics has developed dual-type proximity sensors machine learning and LSTM neural networks, respectively.[137]
for compliant control in robotic arms. Fogale Robotics and Yue- In everyday life, door locks or entering passwords in bank are
jiang Technology have successfully commercialized proximity- often accomplished using touch keypads. However, this method
sensitive safety skins, enabling obstacle avoidance in robotic poses the risk of leaving fingerprints or viruses on the keypad,
arms. Proximity sensors can also be employed for object local- potentially resulting in the exposure of personal information and
ization during the robotic grasping process.[112] hygiene concerns. Touchless input technology can address this
As future robotic work environments grow increasingly com- issue. Le et al. accomplished touchless password input by com-
plex, there is a heightened demand for enhanced robot flexibility, bining a piezoelectric resonant humidity sensor with a triboelec-
safety, and other aspects. Single-modal sensing technology has tric proximity sensor. The humidity sensor consists of a two-port
become inadequate to meet these contemporary requirements. aluminum nitride (AlN) body resonator and a uniformly thick
Single-modal perception can offer only partial insights into ob- graphene oxide film. The two-port piezoelectric body wave res-
jects, rendering inferences about object images susceptible to onator is fabricated on a layered structure of AlN piezoelectric
misinterpretation.[113] Consequently, it is imperative to introduce layer and Si structure layer, with the Si layer highly doped and
multimodal information in the future to comprehensively cap- serving as the bottom electrode. The triboelectric proximity sen-
ture objectattributes and fulfill the operational demands of future sor consists of two parts, including a point base and an Ecoflex
robots in intricate environments. ring on the fingertip. As shown in Figure 13e, the input display
interface for non-contact passwords is composed of three inter-
faces that do not interfere with each other. There are four control
5.2. Human Machine Interface points included in each input interface. The triboelectric-based
sensor is responsible for activating the panels and determining
The human-machine interface (HMI) plays a pivotal role in the position of the password. The humidity sensor controls the
human-computer interaction.[114] Amid the pandemic, touch- switching of the password panels and increases the number of
less HMI has garnered heightened interest due to its flexibility panels. This approach enables touchless input of 3D passwords.
and hygiene advantages. It demonstrates substantial potential Furthermore, flexible proximity sensors can also enable non-
in various domains, encompassing gesture recognition[115–118] contact control of games, providing players with a more immer-
and touchless door locks.[119] Vision-based gesture recognition sive gaming experience.[138]
is an effective method for non-contact HMI. Its system is gen-
erally based on color cameras or depth sensors to collect ges-
ture data.[120] This information is then processed and analyzed, 5.3. Remote Monitoring
such as image segmentation, extraction and classification of
gesture actions, etc.[121] However, vision-based gesture recogni- Monitoring human movement and its interaction with the sur-
tion has problems such as being affected by lighting (such as rounding environment is of particular importance for medical
changing lighting conditions),[122–125] time-consuming process- applications.[139] Proximity sensors are extensively employed in
ing of large amounts of image data, and occlusion of complex safety prevention and healthcare systems, facilitating real-time
gestures.[123–126] Proximity sensors used for gesture recognition and precise monitoring of movement without direct contact with
have the advantages of not being affected by light and having the human body. This provides favorable conditions for remote
a small amount of data, so using proximity sensors for gesture monitoring. In medical facilities or elderly care environments,
recognition has certain advantages. Zhou et al. combined tribo- individuals’ memory impairment or resistance to wearable de-
electric proximity sensors with deep learning to create an intelli- vices may affect the effectiveness of monitoring based on such
gent touchless gesture recognition system. The sensor comprises devices. Therefore, an ideal medical monitoring device should
liquid metal, polymer fibers (SBS), and branched starch-based not only fulfill its monitoring function but also respect the privacy
hydrogel [Figure 13a]. It can easily and securely adhere to vari- of the individuals being monitored. Anaya et al. have engineered
ous common substrates. They integrated the sensor with a deep a triboelectric-based sensor by connecting a negatively charged
learning-based multilayer perceptron (MLP) neural network, precharged PDMS film to aluminum (Al) foil. This sensor

Adv. Sci. 2024, 11, 2308560 2308560 (15 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 13. Flexible proximity sensors applied in human-machine interface. a) Structural design of TTS (touchless tracking system). b) Photographs of
16 different non-contact gestures and electrical signals from the TTS. c) Confusion matrix for verifying the structure. Reproduced with permission.[127]
Copyright 2022, Wiley-VCH. d) Display images of “Thank you”, “I’ll say”, and “Everyday” dynamic gestures in sequence. Reproduced with permission.[137]
Copyright 2021, American Chemical Society. e) Login to the system’s non-contact 3D password interface. Reproduced under terms of the CC-BY
license.[138] Copyright 2022, The Authors, published by Wiley-VCH.

enables near-field remote behavior monitoring of the human sensor response. In Figure 14c, sensors are placed at specific loca-
body within a range of 1.5 meters. It can estimate the relative tions, such as walls. When a blind person approaches, the voltage
positions of two individuals and distinguish between various increases, and the positive and negative peak values reach the set
human activities, such as jumping and walking. As shown in threshold. The data is transmitted to the wristband alarm through
Figure 14a, the walking direction of a person can be identified Bluetooth, triggering the alarm to achieve a collision prevention
using two sensors combined with a new signal power discrimi- function. The sensor can also be placed in key dangerous places,
nation algorithm. In Figure 14b, four sensors are placed at fixed such as shower doors. When someone falls in front of the shower
positions on the wall, allowing the walking path of individuals door, the sensor signal will suddenly have a negative voltage
in the house to be obtained based on the time sequence of each peak, followed by two positive voltage peaks. This will be judged

Adv. Sci. 2024, 11, 2308560 2308560 (16 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Figure 14. Flexible proximity sensors applied in remote monitoring. a) Walking and running direction monitoring. b) Indoor positioning system moni-
toring. c) Blind navigation assistance, indoor positioning and fall sensor detection. Reproduced with permission.[140] Copyright 2021, Elsevier. d) Prepa-
ration and self-healing properties of PAM-dc-fGO thin films. e) Remote monitoring of human motion through film-based proximity sensor. Reproduced
with permission.[141] Copyright 2020, American Chemical Society.

as a fall by the code, and then the alarm will sound an alarm, self-healing. Finally, the significance of flexible proximity sen-
thereby achieving fall detection.[140] Furthermore, Zhang et al. de- sors in applications such as human-robot collaboration, human-
veloped a conductive film (PAM-dc-fGO) by cross-linking intrin- machine interface, and remote monitoring was summarized.
sic self-healing polyazome-thine (PAM) with ethylenediamine- With the development of flexible electronic technology, these sen-
functionalized graphene oxide (fGO) through dynamic covalent sors can achieve more intelligent and adaptive human-machine
bonds (imine bonds, −CH = N−) [Figure 14d]. After 24 h, the interaction, providing key support for the future development of
severed conductive film can self-heal, and its stress-strain curve human-machine collaboration.
matches the initial state. This conductive film can be integrated Capacitive proximity sensors are deployable in non-conductive
with the human body and air to create an organic field-effect shielded environments and offer a straightforward and cost-
transistor (OFET), contributing to remote monitoring of the hu- effective sensor solution. However, factors such as the humidity
man body. While the human body is in motion, the conductive level in the surrounding environment and the presence of inter-
film translates a variety of movements into unique electrical sig- nal objects can easily affect their detection accuracy. To mitigate
nals, encompassing actions such as stomping, jumping, hand this issue, one approach involves integrating capacitive proximity
gestures, and walking back and forth in close proximity to the de- sensors with other detection methods, such as ultrasonic or in-
vice. This capability enables the sensor to achieve remote mon- ductive sensors. Alternatively, data processing techniques can be
itoring of the human body [Figure 14e]. The sensor’s detection utilized to decrease data variability and minimize noise interfer-
range surpasses 50 cm, rendering it especially valuable in the do- ence. Furthermore, specific sensor chips, such as the FDC2214
main of long-distance monitoring.[141] Moreover, flexible proxim- series chips, can aid in mitigating environmental interference.
ity sensors also have applications in remote positioning and other On the other hand, triboelectric proximity sensors operate
fields.[37] without requiring external power sources and demonstrate sen-
sitivity to dynamic changes. However, their signal detection is
6. Conclusion instantaneous, which may have limitations in specific applica-
tions. Flexible proximity sensors hold great promise in the field
In this review, we explore the most recent research findings on of human-robot collaboration. As ro-bots increasingly take on di-
flexible proximity sensors. We place a strong emphasis on under- verse and complex tasks, the demand for high-performance and
standing the operational principles of capacitive proximity sen- adaptable flexible proximity sensors will continue to rise. How-
sors and triboelectric proximity sensors. Additionally, the devel- ever, along with opportunities come challenges.
opment of single type and dual type proximity sensors is summa- 1. Multifunctionality: Developing sensors with multiple in-
rized. The research progress of single-type proximity sensors is tegrated capabilities within a single device poses significant
described concerning large area, multifunctionality, strain, and challenges. As mentioned previously, addressing inter-ference

Adv. Sci. 2024, 11, 2308560 2308560 (17 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

between sensors and external stimuli and optimizing the distri- [8] K. Dong, Z. Wu, J. Deng, A. C. Wang, H. Zou, C. Chen, D. Hu, B. Gu,
bution of sensor structures are critical factors. B. Sun, Z. L. Wang, Adv. Mater. 2018, 30, 1804944.
2. Self-healing: Research on the self-healing properties of flex- [9] Y. Xie, X. Wu, X. Huang, Q. Liang, S. Deng, Z. Wu, Y. Yao, L. Lu,
ible proximity sensors remains relatively limited. Although self- Research 2023, 6, 0157.
[10] Y. Cheng, L. Li, Z. Liu, S. Yan, F. Cheng, Y. Yue, S. Jia, J. Wang, Y. Gao,
healing capabilities have been investigated in hydrogels or ion
L. Li, Research 2022.
gels within electronic skins, a universally applicable implementa-
[11] E. J. Markvicka, R. Tutika, M. D. Bartlett, C. Majidi, Adv. Funct. Mater.
tion method is still lacking. Additionally, the practical implemen- 2019, 29, 1900160.
tation of self-healing abili-ties in real devices is still constrained. [12] L. E. Osborn, A. Dragomir, J. L. Betthauser, C. L. Hunt, H. H.
3. Response Time: In applications such as robot safety, the re- Nguyen, R. R. Kaliki, N. V. Thakor, Sci. Robot. 2018, 3, eaat3818.
sponse time of flexible proximity sensors plays a critical role. [13] F. Li, S. Gao, Y. Lu, W. Asghar, J. Cao, C. Hu, H. Yang, Y. Wu, S. Li, J.
However, many research papers do not explicitly mention the re- Shang, Adv. Sci. 2021, 8, 2004208.
sponse time related to proximity effects. [14] S. Han, X. Zhi, Y. Xia, W. Guo, Q. Li, D. Chen, K. Liu, X. Wang, Small
In addition to these challenges, addressing issues related to 2023, 2301593.
detection range, measurement accuracy, durability, and stability [15] Y. Yang, G. Zhao, X. Cheng, H. Deng, Q. Fu, ACS Appl. Mater. Inter-
faces 2021, 13, 14599.
is essential to ensure the reliability and long-term use of flexi-
[16] R. Moheimani, N. Aliahmad, N. Aliheidari, M. Agarwal, H. Dalir, Sci.
ble proximity sensors. Furthermore, the effective integration of
Rep. 2021, 11, 1104.
these sensors with robot systems for real-time data transmission [17] Z. Yuan, X. Du, H. Niu, N. Li, G. Shen, C. Li, Z. L. Wang, Nanoscale
and processing, along with the extraction of valuable information 2019, 11, 495.
from abundant sensor data, are areas of focus deserving careful [18] P. Zhao, Y. Song, P. Xie, F. Zhang, T. Xie, G. Liu, J. Zhao, S. T. Han,
consideration. Y. Zhou, Adv. Funct. Mater. 2023, 2301816.
[19] M. L. Jin, S. Park, J. S. Kim, S. H. Kwon, S. Zhang, M. S. Yoo, S. Jang,
H. J. Koh, S. Y. Cho, S. Y. Kim, C. W. Ahn, K. Cho, S. G. Lee, D. H.
Acknowledgements Kim, H. T. Jung, Adv. Mater. 2018, 30, 1706851.
[20] M. L. Jin, S. Park, H. Kweon, H. J. Koh, M. Gao, C. Tang, S. Y. Cho,
The current work was financially supported by the National Natural Sci- Y. Kim, S. Zhang, X. Li, K. Shin, A. Fu, H. T. Jung, C. W. Ahn, D. H.
ence Foundation of China (Grand nos. 52003134) to M.L.J., the Major Kim, Adv. Mater. 2021, 33, 2007605.
Key Projects of PCL (PCL2023A07, PCL2023A10) to J.J., Young Taishan [21] T. Mochizuki, Y. Takigami, T. Kondo, H. Okuzaki, J. Appl. Polym. Sci.
Scholars Program of Shandong Province (Grand nos. tsqn201909099, 2018, 135, 45972.
tsqn202306272) to M.L.J. and Q. Z., and the National Key Research and [22] Y. Gu, T. Zhang, H. Chen, F. Wang, Y. Pu, C. Gao, S. Li, Nanoscale
Development Project of China (Grant nos. 2023YFFO715101) to Q.Z. Res. Lett. 2019, 14, 263.
[23] S. E. Navarro, S. Muhlbacher-Karrer, H. Alagi, H. Zangl, K. Koyama,
B. Hein, C. Duriez, J. R. Smith, IEEE Trans Robot 2022, 38, 1599.
Conflict of Interest [24] F. Semeraro, A. Griffiths, A. Cangelosi, Robot Comput Integr Manuf
2023, 79, 102432.
The authors declare no conflict of interest. [25] Y. Lu, H. Zheng, S. Chand, W. Xia, Z. Liu, X. Xu, L. Wang, Z. Qin, J.
Bao, J. Manuf. Syst. 2022, 62, 612.
[26] J. Inga, M. Ruess, J. H. Robens, T. Nelius, S. Rothfuß, S. Kille,
Keywords P. Dahlinger, A. Lindenmann, R. Thomaschke, G. Neumann, S.
Matthiesen, S. Hohmann, A. Kiesel, Int. J. Hum. Comput. Stud. 2023,
proximity sensing electronic skin, human-robot collaboration, human- 170, 102926.
machine interface, remote monitoring
[27] J. M. Hoc, Ergonomics 2000, 43, 833.
[28] E. Matheson, R. Minto, E. G. G. Zampieri, M. Faccio, G. Rosati,
Received: November 9, 2023 Robotics 2019, 8, 100.
Revised: December 27, 2023 [29] S. Haddadin, A. AlbuSchäffer, A. D. Luca, G. Hirzinger, presented at
Published online: January 28, 2024 IEEE/RSJ International Conference on Intelligent Robots & Systems
2008.
[30] T. Schlegl, T. Kröger, A. Gaschler, O. Khatib, H. Zangl, presented at
2013 IEEE/RSJ International Conference on Intelligent Robots and
[1] Y. Chen, Y. Sun, Y. Wei, J. Qiu, Adv. Mater. Technol. 2023, 8, 2201352. Systems 2013.
[2] H. H. Chou, A. Nguyen, A. Chortos, J. W. To, C. Lu, J. Mei, T. [31] J. Begum, N. A. Mir, K. Dev, B. Buyamayum, M. Y. Wani, M. Raza,
Kurosawa, W. G. Bae, J. B. Tok, Z. Bao, Nat. Commun. 2015, 6, 8011. Transbound Emerg. Dis. 2021, 68, 1111.
[3] Y. Zhou, D. Dai, Y. Gao, Z. Zhang, N. Sun, H. Tan, X. Cai, J. Cai, [32] Z. Zhu, X. Lian, X. Su, W. Wu, G. A. Marraro, Y. Zeng, Respir. Res.
ChemNanoMat 2021, 7, 982. 2020, 21, 224.
[4] A. K. Dabrowska, G. M. Rotaru, S. Derler, F. Spano, M. Camenzind, [33] M. Lotfi, M. R. Hamblin, N. Rezaei, Clin. Chim. Acta 2020, 508, 254.
S. Annaheim, R. Stampfli, M. Schmid, R. M. Rossi, Skin Res Technol [34] Y. Luo, M. R. Abidian, J. H. Ahn, D. Akinwande, A. M. Andrews, M.
2016, 22, 3. Antonietti, Z. Bao, M. Berggren, C. A. Berkey, C. J. Bettinger, J. Chen,
[5] G. Y. Bae, J. T. Han, G. Lee, S. Lee, S. W. Kim, S. Park, J. Kwon, S. P. Chen, W. Cheng, X. Cheng, S. J. Choi, A. Chortos, C. Dagdeviren,
Jung, K. Cho, Adv. Mater. 2018, 30, 1803388. R. H. Dauskardt, C. A. Di, M. D. Dickey, X. Duan, A. Facchetti, Z.
[6] M. L. Jin, S. Park, Y. Lee, J. H. Lee, J. Chung, J. S. Kim, J. S. Kim, S. Y. Fan, Y. Fang, J. Feng, X. Feng, H. Gao, W. Gao, X. Gong, et al., ACS
Kim, E. Jee, D. W. Kim, J. W. Chung, S. G. Lee, D. Choi, H. T. Jung, Nano 2023, 17, 5211.
D. H. Kim, Adv. Mater. 2017, 29, 1605973. [35] D. Fonseca, M. Safeea, P. Neto, IEEE Transactions on Industrial Infor-
[7] G. Lee, J. Son, D. Kim, H. J. Ko, S. G. Lee, K. Cho, Small 2022, 18, matics 2022, 19, 2485.
e2205643. [36] Y. Cheng, R. Wang, H. Zhai, J. Sun, Nanoscale 2017, 9, 3834.

Adv. Sci. 2024, 11, 2308560 2308560 (18 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

[37] F. Guan, Y. Xie, H. Wu, Y. Meng, Y. Shi, M. Gao, Z. Zhang, S. Chen, [70] Y. Wei, R. Torah, Y. Li, J. Tudor, Sens. Actuators, A 2016, 247,
Y. Chen, H. Wang, Q. Pei, ACS Nano 2020, 14, 15428. 239.
[38] P. Zhao, R. Zhang, Y. Tong, X. Zhao, T. Zhang, X. Wang, Q. Tang, Y. [71] D. Gao, J. Wang, K. Ai, J. Xiong, S. Li, P. S. Lee, Adv. Intell. Syst. 2020,
Liu, ACS Appl. Mater. Interfaces 2021, 13, 49085. 2, 2000088.
[39] D. Bu, S. Q. Li, Y. M. Sang, C. J. Qiu, Mod. Phys. Lett. B 2018, 32. [72] L. Y. W. Loh, U. Gupta, Y. Wang, C. C. Foo, J. Zhu, W. F. Lu, Adv. Eng.
[40] C. Ge, Z. Wang, Z. Liu, T. Wu, S. Wang, X. Ren, D. Chen, J. Zhao, Mater. 2021, 23, 2001082.
P.-A. Hu, J. Zhang, Adv. Intell. Syst. 2022, 4, 2100213. [73] J. C. Yang, J. Mun, S. Y. Kwon, S. Park, Z. Bao, S. Park, Adv. Mater.
[41] C. F. Hu, J. Y. Wang, Y. C. Liu, M. H. Tsai, W. Fang, Nanotechnology 2019, 31, 1904765.
2013, 24, 444006. [74] M. Wang, Y. Luo, T. Wang, C. Wan, L. Pan, S. Pan, K. He, A. Neo, X.
[42] M. g. Kim, H. Alrowais, O. Brand, Adv. Electron. Mater. 2018, 4, Chen, Adv. Mater. 2021, 33, 2003014.
1700434. [75] M. Xie, K. Hisano, M. Zhu, T. Toyoshi, M. Pan, S. Okada, O.
[43] S. Kim, S. Lee, P. Bhuyan, Y. Wei, S. Kim, K. Shimizu, J. Shintake, S. Tsutsumi, S. Kawamura, C. Bowen, Adv. Mater. Technol. 2019, 4,
Park, Chem. Eng. J. 2023, 456. 1800626.
[44] S. R. A. Ruth, V. R. Feig, M.-g. Kim, Y. Khan, J. K. Phong, Z. Bao, Small [76] Y. Guo, X. Wei, S. Gao, W. Yue, Y. Li, G. Shen, Adv. Funct. Mater. 2021,
Struct 2021, 2, 2000079 31, 2104288.
[45] Z. Zhang, J. Li, B. Yu, D. Huang, Q. Wang, Z. Li, presented at 2022 [77] H. Luo, G. Pang, K. Xu, Z. Ye, H. Yang, G. Yang, Adv. Mater. Technol.
IEEE 35th International Conference on Micro Electro Mechanical 2021, 6, 2100616.
Systems Conference (MEMS) 2022. [78] S. Chen, M. Li, Y. Huang, H. Xu, G. Gu, X. Guo, IEEE Trans. Electron
[46] K. Zheng, F. Gu, H. Wei, L. Zhang, X. a. Chen, H. Jin, S. Pan, Y. Chen, Devices 2020, 67, 2940.
S. Wang, Small Methods 2023, 7, 2201534. [79] G. Li, S. Liu, Q. Mao, R. Zhu, Adv. Sci. 2022, 9, 2104969.
[47] J. Huang, H. Wang, J.-a. Li, S. Zhang, H. Li, Z. Ma, M. Xin, K. Yan, [80] Y. Luo, X. Chen, X. Li, H. Tian, L. Wang, J. Shao, Nano Res. 2023, 1.
W. Cheng, D. He, X. Wang, Y. Shi, L. Pan, ACS Mater. Lett. 2022, 4, [81] W. Liu, F. Xiang, D. Mei, Y. Wang, Adv. Mater. Technol. 2023, 2301685.
2261. [82] H. Niu, H. Li, Q. Zhang, E. S. Kim, N. Y. Kim, Y. Li, Small 2023,
[48] C. Zhang, S. Liu, X. Huang, W. Guo, Y. Li, H. Wu, Nano Energy 2019, 2308127.
62, 164. [83] Y. Dai, S. Gao, IEEE Sens. J. 2021, 21, 26355.
[49] J. Chen, G. Zhu, J. Wang, X. Chang, Y. Zhu, ACS Appl. Mater. Inter- [84] D. C. Wang, H. Y. Yu, L. Jiang, D. Qi, X. Zhang, L. Chen, W. Lv, W.
faces 2023, 15, 7485. Xu, K. C. Tam, Nano Res. 2022, 15, 2616.
[50] S. Pan, Z. Zhang, Friction 2019, 7, 2. [85] Q. Hua, J. Sun, H. Liu, R. Bao, R. Yu, J. Zhai, C. Pan, Z. L. Wang, Nat.
[51] J. Zhu, M. Zhu, Q. Shi, F. Wen, L. Liu, B. Dong, A. Haroun, Y. Yang, Commun. 2018, 9, 244.
P. Vachon, X. Guo, T. He, EcoMat 2020, 2, e12058. [86] H. S. Jo, S. An, C.-W. Park, D.-Y. Woo, A. L. Yarin, S. S. Yoon, ACS
[52] R. Mondal, M. A. M. Hasan, R. Zhang, H. Olin, Y. Yang, Adv. Mater. Appl. Mater. Interfaces 2019, 11, 40232.
Technol. 2022, 7. [87] H. S. Jo, C.-W. Park, S. An, A. Aldalbahi, M. El-Newehy, S. S. Park, A.
[53] Y. Xiong, J. Han, Y. Wang, Z. L. Wang, Q. Sun, Research 2022. L. Yarin, S. S. Yoon, NPG Asia Mater 2022, 14.
[54] J. Ma, J. Zhu, P. Ma, Y. Jie, Z. L. Wang, X. Cao, ACS Energy Lett. 2020, [88] Q. Liang, D. Zhang, Y. Wu, X. Qu, Y. Jia, S. Chen, H. Wang, C. Lee,
5, 3005. Nano Energy 2023, 113.
[55] Y. Tang, H. Zhou, X. Sun, N. Diao, J. Wang, B. Zhang, C. Qin, E. [89] P. Zhao, R. Zhang, Y. Tong, X. Zhao, T. Zhang, Q. Tang, Y. Liu, ACS
Liang, Y. Mao, Adv. Funct. Mater. 2019, 30, 1907893. Appl. Mater. Interfaces 2020, 12, 55083.
[56] D. Liu, Y. Gao, L. Zhou, J. Wang, Z. L. Wang, Nano Res. 2023. [90] H. Ding, Z. Wu, H. Wang, Z. Zhou, Y. Wei, K. Tao, X. Xie, J. Wu, Mater.
[57] S. Chen, Y. Wang, L. Yang, Y. Guo, M. Wang, K. Sun, Nano Energy Horiz. 2022, 9, 1935.
2021, 82, 105719. [91] T. P. Huynh, P. Sonar, H. Haick, Adv. Mater. 2017, 29, 1604973.
[58] W. Zhang, Y. Lu, T. Liu, J. Zhao, Y. Liu, Q. Fu, J. Mo, C. Cai, S. Nie, [92] M. Khatib, O. Zohar, H. Haick, Adv. Mater. 2021, 33, 2004190.
Small 2022, 18, e2200577. [93] M. Qi, R. Yang, Z. Wang, Y. Liu, Q. Zhang, B. He, K. Li, Q. Yang, L.
[59] H. Wang, Q. Tang, X. Zhao, Y. Tong, Y. Liu, ACS Appl. Mater. Interfaces Wei, C. Pan, M. Chen, Adv. Funct. Mater. 2023, 33.
2018, 10, 2785. [94] Y. Chen, C. Zhang, R. Yin, A. Yin, Q. Feng, F. Liu, J. Shao, T. Su, H.
[60] W. Liu, Y. Niu, Q. Chen, H. Jiang, F. Xu, G. Zhu, Adv. Electron. Mater. Wang, G. Chen, W. Zhao, Chem. Eng. J. 2022, 449, 137907.
2019, 5, 1900586. [95] H. Guo, Y. J. Tan, G. Chen, Z. Wang, G. J. Susanto, H. H. See, Z.
[61] V. Kedambaimoole, N. Kumar, V. Shirhatti, S. Nuthalapati, S. Kumar, Yang, Z. W. Lim, L. Yang, B. C. K. Tee, Nat. Commun. 2020, 11,
M. M. Nayak, P. Sen, D. Akinwande, K. Rajanna, Adv. Electron. Mater. 5747.
2021, 7, 2001214. [96] F. Yuan, S. Liu, J. Zhou, S. Wang, Y. Wang, S. Xuan, X. Gong, Nano
[62] G. Lv, H. Wang, Y. Tong, L. Dong, X. Zhao, P. Zhao, Q. Tang, Y. Liu, Energy 2021, 86.
Adv. Mater. Interfaces 2020, 7, 2000306. [97] W. Liu, Y. Duo, J. Liu, F. Yuan, L. Li, L. Li, G. Wang, B. Chen, S. Wang,
[63] J. Ge, X. Wang, M. Drack, O. Volkov, M. Liang, G. S. Canon H. Yang, Y. Liu, Y. Mo, Y. Wang, B. Fang, F. Sun, X. Ding, C. Zhang,
Bermudez, R. Illing, C. Wang, S. Zhou, J. Fassbender, M. L. Wen, Nat. Commun. 2022, 13, 5030.
Kaltenbrunner, D. Makarov, Nat. Commun. 2019, 10, 4405. [98] M. Zhu, M. Lou, J. Yu, Z. Li, B. Ding, Nano Energy 2020, 78.
[64] R. Dahiya, N. Yogeswaran, F. Liu, L. Manjakkal, E. Burdet, V. [99] J. Liu, Y. Tong, D. Xian, X. Chen, C. Yang, R. Zou, X. Liu, Z. Liang, X.
Hayward, H. Jorntell, Proc. IEEE 2016, 107, 2019. Zhao, Q. Tang, Adv. Funct. Mater. 2023, 2306704.
[65] A. Zimmerman, L. Bai, D. D. Ginty, Sci 2014, 346, 950. [100] Z. H. A. O. Guo-Dong, Z. H. A. O. Lei, W. A. N. G. Hai-Ting, W. E. I.
[66] M. S. Sarwar, Y. Dobashi, C. Preston, J. K. M. Wyss, S. Mirabbasi, J. Zhan, T. O. N. G. Yan-Hong, Z. H. A. O. Xiao-Li, T. A. N. G. Qing-Xin,
D. W. Madden, Sci. Adv. 2017, 3, e1602200. L. I. U. Yi-Chun, Chinese J. Anal. Chem. 2022, 50, 20.
[67] Y. Li, X. Zhou, J. Chen, W. Guo, S. He, S. Gao, P. Peng, Adv. Mater. [101] S. Qiu, Y. Huang, X. He, Z. Sun, P. Liu, C. Liu, Meas. Sci. Technol.
Interfaces 2020, 7, 1901845. 2015, 26, 105101.
[68] X. You, J. He, N. Nan, X. Sun, K. Qi, Y. Zhou, W. Shao, F. Liu, S. Cui, [102] T. Kim, J. Noh, T. D. Nguyen, H. R. Choi, presented at 2019 IEEE/RSJ
J. Mater. Chem. C 2018, 6, 12981. International Conference on Intelligent Robots and Systems (IROS)
[69] T. Li, S. Zhang, Microsyst. Technol. 2018, 24, 4875. 2019.

Adv. Sci. 2024, 11, 2308560 2308560 (19 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

[103] T. Islam, O. P. Maurya, A. U. Khan, IEEE Sens. J. 2021, 21, 24812. [122] M. Wang, Z. Yan, T. Wang, P. Cai, S. Gao, Y. Zeng, C. Wan, H. Wang,
[104] J. Jiao, L. Li, B. Wu, C. He, Sens. Actuators, A 2017, 253, 75. L. Pan, J. Yu, S. Pan, K. He, J. Lu, X. Chen, Nat. Electron 2020, 3, 563.
[105] H. Alagi, A. Heiligl, S. E. Navarro, T. Kroegerl, B. Hein, presented at [123] S. Wang, A. Wang, M. Ran, L. Liu, Y. Peng, M. Liu, G. Su, A. Alhudhaif,
2018 IEEE/RSJ International Conference on Intelligent Robots and F. Alenezi, N. Alnaim, Inform. Sci. 2022, 606, 722.
Systems (IROS) 2018. [124] Z. Zhou, K. Chen, X. Li, S. Zhang, Y. Wu, Y. Zhou, K. Meng, C. Sun,
[106] S. M. Noble, M. Mende, D. Grewal, A. Parasuraman, Journal of Re- Q. He, W. Fan, Nat. Electron 2020, 3, 571.
tailing 2022, 98, 199. [125] H. Liu, L. Wang, Int. J. Ind. Ergonom. 2018, 68, 355.
[107] F. Xia, B. Bahreyni, F. Campi, presented at IEEE SENSORS IEEE [126] S. An, H. Zhu, C. Guo, B. Fu, C. Song, P. Tao, W. Shang, T. Deng,
2016, 1. Nat. Commun. 2022, 13, 1446.
[108] Z. H. Guo, H. L. Wang, J. Shao, Y. Shao, L. Jia, L. Li, X. Pu, Z. L. Wang, [127] H. Zhou, W. Huang, Z. Xiao, S. Zhang, W. Li, J. Hu, T. Feng, J. Wu,
Sci. Adv. 2022, 8, eabo5201. P. Zhu, Y. Mao, Adv. Funct. Mater. 2022, 32.
[109] G. Ma, M. Soleimani, Sci. Rep. 2020, 10, 11525. [128] M. Zhu, Z. Sun, Z. Zhang, Q. Shi, T. He, H. Liu, T. Chen, C. Lee, Sci.
[110] C. Ge, X. An, X. He, Z. Duan, J. Chen, P. Hu, J. Zhao, Z. Wang, J. Adv. 2020, 6, eaaz8693.
Zhang, Adv. Sci. 2023, 2301341. [129] F. Wen, Z. Sun, T. He, Q. Shi, M. Zhu, Z. Zhang, L. Li, T. Zhang, C.
[111] S. J. Moon, J. Kim, H. Yim, Y. Kim, H. R. Choi, IEEE Robot Autom Lee, Adv. Sci. 2020, 7, 2000261.
Lett 2021, 6, 8021. [130] B. Dong, Y. Yang, Q. Shi, S. Xu, Z. Sun, S. Zhu, Z. Zhang, D. L.
[112] H. Guo, H. Wu, Y. Song, L. Miao, X. Chen, H. Chen, Z. Su, M. Han, Kwong, G. Zhou, K. W. Ang, C. Lee, ACS Nano 2020, 14, 8915.
H. Zhang, Nano Energy 2019, 58, 121. [131] J. DelPreto, J. Hughes, M. D’Aria, M. de Fazio, D. Rus, IEEE Robot
[113] S. Duan, Q. Shi, J. Wu, Adv. Intell. Syst. 2022, 4, 2200213. Autom Lett 2022, 7, 10589.
[114] Z. Lin, G. Zhang, X. Xiao, C. Au, Y. Zhou, C. Sun, Z. Zhou, R. Yan, [132] Y. Su, K. Ma, X. Mao, M. Liu, X. Zhang, Nanomaterials 2022, 12.
E. Fan, S. Si, L. Weng, S. Mathur, J. Yang, J. Chen, Adv. Funct. Mater. [133] S. Jiang, Q. Gao, H. Liu, P. B. Shull, Sens. Actuators, A 2020, 301,
2021, 32. 111738.
[115] H. Lee, J. K. Mandivarapu, N. Ogbazghi, Y. Li, arXiv preprint 2022. [134] X. Li, Y. Ma, C. Zhang, Y. Lin, H. Chen, presented at 2022 2nd In-
[116] R. Liu, Y. Lai, S. Li, F. Wu, J. Shao, D. Liu, X. Dong, J. Wang, Z. L. ternational Conference on Computation, Communication and Engi-
Wang, Nano Energy 2022, 95. neering (ICCCE) 2022.
[117] M. Virone, P. Lopes, R. P. Rocha, A. T. de Almeida, M. Tavakoli, [135] L. Zhu, G. Mao, H. Su, Z. Zhou, W. Li, X. Lu, Z. Wang, IEEE Sens. J.
presented at 2021 IEEE/RSJ International Conference on Intelligent 2021, 21, 9937.
Robots and Systems (IROS) 2021. [136] F. Chen, H. Lv, Z. Pang, J. Zhang, Y. Hou, Y. Gu, H. Yang, G. Yang,
[118] A. Braun, T. Dutz, F. Kamieth, presented at Proceedings of the 6th IEEE Sens. J. 2019, 19, 8441.
International Conference on PErvasive Technologies Related to As- [137] J. Pan, Y. Li, Y. Luo, X. Zhang, X. Wang, D. L. T. Wong, C. H. Heng,
sistive Environments 2013. C. K. Tham, A. V. Thean, ACS Sens. 2021, 6, 4156.
[119] S. Zeiß, A. Marinc, A. Braun, T. Große-Puppendahl, S. Beck, pre- [138] X. Le, Q. Shi, Z. Sun, J. Xie, C. Lee, Adv. Sci. 2022, 9, 2201056.
sented at Distributed, Ambient, and Pervasive Interactions: Second [139] X. Fu, X. Pan, Y. Liu, J. Li, Z. Zhang, H. Liu, M. Gao, Adv. Funct. Mater.
International Conference, DAPI 2014, Held as Part of HCI Intera- 2023, 2306749.
tional Heraklion, Crete, Greece 2014, Proceedings 2. [140] D. V. Anaya, K. Zhan, L. Tao, C. Lee, M. R. Yuce, T. Alan, Nano Energy
[120] H. P. Wang, Y. X. Zhou, H. Li, G. D. Liu, S. M. Yin, P. J. Li, S. Y. Dong, 2021, 90.
C. Y. Gong, S. Y. Wang, Y. B. Li, Adv. Sci. 2022, 9, 2105056. [141] Y. Du, G. Yu, X. Dai, X. Wang, B. Yao, J. Kong, ACS Appl. Mater. In-
[121] Y. Dong, J. Liu, W. Yan, IEEE Trans. Instrum. Meas. 2021, 70, 1. terfaces 2020, 12, 51987.

Bingwei Wu is currently a master’s student at the Heart Center, Qingdao Hiser Hospital Affiliated with
Qingdao University and the Institute for Future, School of Automation, Qingdao University. His main
research areas include the design and application of flexible proximity sensors, flexible tactile sensors,
as well as embedded design and development.

Adv. Sci. 2024, 11, 2308560 2308560 (20 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH
 21983844, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202308560 by University Of Exeter, Wiley Online Library on [17/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advancedscience.com

Qihui Zhou received his Ph.D. degree from the University Medical Center Groningen in 2018. He is
currently a professor/PI of the University of Health and Rehabilitation Sciences and the Head of the
Institute of Biomaterials for Tissue Repair and Rehabilitation and Qingdao Key Laboratory. He is
also a Young Taishan Scholar and leader of the “Materiobiology” project under the “Youth Innova-
tion Science and Technology Plan” in Shandong. His research focuses on the development of bio-
multifunctional materials based on the bioadaptability concept to carry out basic and applied research
on tissue repair and rehabilitation (heart/blood vessel, nerve, oral, motor system, skin).

Jian Jiao received his Ph.D. degree in Mechanical Engineering from Beihang University, Beijing, China,
in 2019. He is currently an Associate Professor and Doctoral Supervisor, working at the Department of
Mathematics and Theories, Peng Cheng Laboratory, Shenzhen, China. He has been a reviewer of IEEE
Transactions on Industrial Informatics since 2023. His research interests include soft robotics, flexible
electronics, haptics, and virtual reality.

Ming Liang Jin received his Ph.D. degree in Chemical and Biomolecular Engineering from the Korea
Advanced Institute of Science and Technology in 2017. He worked as a visiting scholar at Stanford
University in 2012. He participated in postdoctoral research in Electrical and Chemical Engineering
at the University of Minnesota in 2018. He is currently a Professor at the Institute for Future, School of
Automation, Qingdao University. Additionally, he is a Young Taishan Scholar in Shandong, China. His
current research interests include intelligent materials, nano-micro structure, and electronic skins.

Adv. Sci. 2024, 11, 2308560 2308560 (21 of 21) © 2024 The Authors. Advanced Science published by Wiley-VCH GmbH