1. Bibliographic Information

1.1. Title

Flexible sensors for force detection: A review

1.2. Authors

Xue Zhao, Lili Wang, Yufei Hao, Yan Zhao, Jianhua Zhang

1.3. Journal/Conference

The paper was published in an academic journal. While the full journal name is not explicitly provided in the excerpt, the presence of an ARTICLEINFO section and typical academic publishing formats suggests a peer-reviewed journal. Given the context of the abstract and content, it's likely a reputable journal in materials science, sensors, or robotics.

1.4. Publication Year

The publication year is not explicitly stated in the provided text excerpt. However, the references section contains citations from 2023, 2024, and even 2025 (in press), suggesting the paper was published in late 2024 or early 2025.

1.5. Abstract

This review paper focuses on flexible force sensors, which are gaining increasing attention due to advancements in smart materials, bionics, and robotics. These sensors are crucial for applications such as health monitoring, prosthetics, and robotic manipulation as they can detect both normal and multi-axis forces. The paper highlights that these sensors employ diverse sensing mechanisms, including resistive, magnetic, and optical principles, and must effectively balance competing requirements such as high spatial-temporal resolution, stretchability, and resistance to electromagnetic and thermal disturbances. The review summarizes recent progress in sensor structures, specifically mentioning bionic microstructures, degradable materials, bending-insensitive designs, and sensing arrays. It provides a detailed comparison of different sensing mechanisms, evaluating their performance metrics, strengths, and limitations. The paper also identifies key innovations, discusses current challenges, and proposes future development directions for flexible force sensors.

1.6. Original Source Link

The original source link is provided as /files/papers/690491c39f2f7e6b6c47c53f/paper.pdf. This indicates a direct link to the PDF file, suggesting it is an officially published paper or a pre-print available through an institutional or publisher repository.

2. Executive Summary

2.1. Background & Motivation

The core problem this paper aims to address is the growing demand for flexible force sensors capable of robustly perceiving and measuring external forces under various deformed conditions. Traditional rigid force sensors, while accurate, lack the adaptability needed for modern applications that require seamless integration with complex, curved surfaces and dynamic environments.

This problem is highly important in current fields due to rapid advancements in wearable devices, robotic tactile sensing, medical rehabilitation, electronic skin, and intelligent interaction systems. Specific challenges and gaps in prior research that flexible sensors seek to overcome include:

• The need for high spatial-temporal resolution to accurately capture dynamic force changes.

• The requirement for stretchability and adaptability to conform to non-planar surfaces and undergo significant deformations without compromising performance.

• The necessity for resistance to electromagnetic and thermal disturbances to ensure stable and reliable measurements in diverse operating environments.

• The limitation of many sensors to detecting only unidirectional normal forces, while many advanced applications require multi-axis force detection (e.g., pressure, shear, tensile, torsional forces).

  The paper's entry point is a comprehensive review of the state-of-the-art in flexible force sensors, categorizing them by their underlying sensing mechanisms and structural designs. Its innovative idea is to provide a structured comparison of these diverse approaches, highlighting their strengths, limitations, and potential for future development.

2.2. Main Contributions / Findings

The primary contributions of this paper are:

• Comprehensive Review of Sensing Mechanisms: It systematically reviews and compares various flexible force sensor types based on resistive, capacitive, magnetic, optical, triboelectric, ion electrical double layer (EDL), and hybrid principles.

• Analysis of Structural Innovations: The paper summarizes recent progress in sensor structures, including bionic microstructures, degradable materials, bending-insensitive designs, and sensing arrays, explaining how these enhance performance.

• Detailed Performance Comparison: It provides a comparative analysis of different sensing mechanisms based on key performance metrics such as sensitivity, detection range, response time, linearity, hysteresis, and cross-talk, identifying their unique advantages and limitations.

• Highlighting Key Innovations: The review points out significant advancements and representative sensor designs within each category.

• Exploration of Applications: It delves into the applications of flexible force sensors in critical fields like wearable health monitoring and robotic operations, providing illustrative case studies.

• Discussion of Challenges and Future Directions: The paper identifies current challenges in areas such as sensing accuracy, long-term stability, mechanical durability, and scalable manufacturing. It also outlines future development trends, including innovation in materials and microstructures, integration with deep learning, and organ monitoring with implanted electrodes.

  Key conclusions and findings include:

• No single sensor type is universally optimal; selection depends on specific application requirements and balancing trade-offs between performance metrics (e.g., sensitivity, detection range, cost, complexity).

• Material properties and microstructural design are crucial for determining sensor performance, influencing sensitivity, linearity, and durability.

• Hybrid sensing approaches offer enhanced functionality and address limitations of individual mechanisms but increase design and manufacturing complexity.

• Future advancements will focus on multi-parameter co-design frameworks, integration with AI, and biocompatible implantable systems for advanced medical applications.

3. Prerequisite Knowledge & Related Work

3.1. Foundational Concepts

To understand this paper, a reader should be familiar with the following foundational concepts:

• Flexible Force Sensors: These are devices designed to detect and measure external forces (such as pressure, shear force, and tensile force) while maintaining their functionality under various deformations like bending, stretching, and twisting. Unlike traditional rigid sensors, flexible sensors are lightweight, thin, stretchable, and highly adaptable, allowing them to conform to complex curved surfaces.
• Smart Materials: Materials whose properties (e.g., electrical resistance, capacitance, magnetism, optical characteristics) can be significantly changed in a controlled way by external stimuli such as mechanical stress, temperature, electric fields, or magnetic fields. Examples include piezoelectric materials, conductive polymers, and shape memory alloys.
• Bionics: The application of biological methods and systems found in nature to the study and design of engineering systems and modern technology. In the context of sensors, this often involves mimicking biological structures (e.g., human skin, insect hairs) to achieve enhanced sensitivity, conformability, or multi-functionality.
• Robotics: The interdisciplinary field of science and engineering that involves the design, construction, operation, and use of robots. Flexible sensors are crucial for giving robots tactile sensing capabilities, enabling more delicate manipulation, adaptive grasping, and safe human-robot interaction.
• Health Monitoring: The continuous or periodic measurement and analysis of physiological parameters (ee.g., heart rate, respiration, blood pressure, gait, muscle activity) to assess an individual's health status, detect abnormalities, and facilitate early diagnosis or rehabilitation. Flexible sensors are ideal for wearable devices in this field due to their comfort and non-invasiveness.
• Spatial-Temporal Resolution:
  • Spatial Resolution: The smallest distinguishable feature or distance that a sensor can detect. For a force sensor, this relates to how precisely it can localize a force application point or differentiate between closely spaced force inputs.
  • Temporal Resolution: The smallest time interval between successive measurements that a sensor can distinguish. For a force sensor, this indicates how quickly it can respond to and capture rapid changes in applied force.
• Stretchability: The ability of a material or device to undergo significant elongation or deformation without breaking or losing its functional properties. This is critical for flexible electronics that need to conform to dynamic or irregular surfaces.
• Electromagnetic (EM) Disturbances: Unwanted electrical or magnetic signals that can interfere with the operation of electronic devices, including sensors, leading to inaccurate readings. Flexible sensors, particularly capacitive ones, need to be designed to resist such interference.
• Thermal Disturbances: Changes in temperature that can affect the performance of sensors, leading to signal drift or reduced accuracy. Material selection and design can mitigate these effects.
• Sensing Arrays: An arrangement of multiple individual sensor elements in a grid or pattern, allowing for spatially resolved (localized) detection of stimuli over an area. This is essential for applications like electronic skin where a distribution of forces needs to be mapped.
• Key Performance Parameters for Sensors:
  • Sensitivity: The ratio of the change in sensor output to the change in the measured input quantity. High sensitivity means a small change in force produces a large change in signal, allowing detection of minute forces.
  • Detection Range: The span of input values (e.g., force magnitudes) over which a sensor can accurately measure.
  • Hysteresis: The phenomenon where the sensor's output for a given input depends on the history of previous inputs (i.e., whether the force is increasing or decreasing). Low hysteresis is desired for consistent and accurate measurements.
  • Linearity: The degree to which the sensor's output is directly proportional to the input force over its detection range. A linear response simplifies calibration and data interpretation.
  • Response Time: The time it takes for a sensor to react to a change in the input and produce a stable output. A fast response time is crucial for dynamic applications.
  • Cross-talk: The undesirable coupling of signals between different sensing axes or elements in a multi-axis or array sensor. For example, a normal force might inadvertently generate a signal that appears to be a shear force. Minimizing cross-talk is critical for accurate multi-axis force decomposition.

3.2. Previous Works

The paper traces the evolutionary trajectory of force sensors from their early inception to modern flexible systems:

• Early 20th Century (Mechanical Designs): The earliest force sensors were purely mechanical, such as spring scales and mechanical weighing instruments. These measured force by detecting the deformation or displacement of elastic components. They were structurally simple but had limitations in sensitivity and accuracy.
• Mid-20th Century (Electrical Measurement Technology): This era saw significant advancements with the emergence of electrical measurement.
  • Resistance Strain Gauges: Measure force by detecting resistance changes induced by material deformation.
  • Piezoelectric Elements: Capture rapidly changing forces by generating charge signals under mechanical stress.
  • Capacitive Sensors: Sense external forces via changes in electrode spacing or dielectric properties. These developments led to higher precision and enhanced dynamic response.
• End of 20th Century (Micro-Electro-Mechanical Systems - MEMS): Advancements in microelectronics and semiconductor technologies enabled MEMS technology. This facilitated the development of miniaturized and integrated force sensors characterized by small size, low power consumption, high sensitivity, and batch production capability.
• 21st Century (Flexible Electronics and Nanomaterials): The rise of flexible electronics and nanomaterials (e.g., carbon nanotubes, graphene, conductive polymers) led to the creation of flexible force sensors. These materials ensure stable measurements even under complex deformations like bending, stretching, and twisting, making them suitable for wearable devices, electronic skin, and robotic touch.
• Concurrent Development (Bionic Technology): Inspired by biological tactile systems, bionic technology has facilitated the advancement of high-sensitivity and high-resolution force sensors.

3.3. Technological Evolution

The technological evolution of force sensors can be summarized as a progression from macroscopic, rigid, and purely mechanical devices to miniaturized, flexible, and intelligent systems.

1. Mechanical Era: Focused on direct mechanical deformation measurement (e.g., spring compression). Limitations: bulkiness, low accuracy, manual reading.

2. Electromechanical Era: Integrated mechanical deformation with electrical signal conversion. This was a significant leap in accuracy and enabled automated reading. Resistance strain gauges, piezoelectric sensors, and capacitive sensors became prevalent.

3. Miniaturization Era (MEMS): Leveraging microfabrication techniques, sensors became smaller, more integrated, and consumed less power. This opened doors for embedding sensors in more complex systems.

4. Flexible & Nanomaterial Era: The advent of flexible substrates and nanomaterials allowed sensors to be stretchable, conformable, and lightweight. This was crucial for wearable technologies and electronic skin, enabling interaction with non-flat and dynamic surfaces.

5. Intelligent & Multifunctional Era: The current trend integrates artificial intelligence, wireless communication, and cloud computing for intelligent data processing and remote monitoring. Sensors are also becoming multifunctional, detecting multiple physical parameters simultaneously (e.g., force, temperature, humidity).

   This paper's work fits within the current intelligent and multifunctional era, providing a comprehensive overview of the diverse sensing principles and structural innovations that define this stage of development. It serves as a synthesis of the advancements made possible by the preceding technological shifts.

3.4. Differentiation Analysis

As a review paper, its core differentiation lies in its comprehensive scope, structured comparative analysis, and forward-looking perspective on flexible force sensors. Compared to specific research papers focusing on a single sensor type or application, this review stands out by:

• Broad Coverage of Sensing Mechanisms: It systematically covers resistive, capacitive, magnetic, optical, triboelectric, ion-EDL, and hybrid principles, providing a holistic view of the field. Many individual papers might focus on one or two of these.

• Detailed Comparison of Performance Metrics: The paper offers a structured comparison of different sensor types based on key performance metrics (sensitivity, detection range, hysteresis, linearity, response time, cross-talk). This comparative analysis is crucial for researchers and practitioners to select the most appropriate sensor for their specific needs, a feature often lacking in single-study papers.

• Emphasis on Structural Innovations: It highlights how material properties and microstructural designs (e.g., bionic microstructures, bending-insensitive designs) are leveraged to enhance sensor performance, integrating insights from materials science and mechanical engineering.

• Application-Oriented Discussion: The review specifically discusses the relevance and impact of these sensors in wearable health monitoring and robotic operations, grounding the technical discussion in practical use cases.

• Identification of Challenges and Future Directions: It critically assesses the current limitations and proposes clear avenues for future research, including the integration of deep learning and the development of implantable electrodes, which is vital for guiding the field's progression.

  In essence, while individual research papers contribute novel methods or results, this review acts as an indispensable map of the flexible force sensing landscape, synthesizing disparate findings into a coherent narrative and identifying overarching trends and challenges.

4. Methodology

This review paper's methodology involves categorizing and detailing various types of flexible force sensors based on their underlying physical principles. For each category, the paper elaborates on the working mechanism, highlights representative designs, and discusses their respective strengths and limitations. The core approach is a structured survey and synthesis of existing literature.

4.1. Principles

The fundamental principle of flexible force sensors is to convert mechanical deformation or stress into a measurable electrical or optical signal. This conversion relies on specific physical effects that are sensitive to changes in geometry, material properties, or field interactions caused by an applied force.

4.1.1. Resistive Force Sensors

Core Idea: Resistive flexible force sensors detect force by monitoring changes in their electrical resistance. This change occurs when the sensor material or structure deforms under an applied force.

Theoretical Basis: The electrical resistance of a material is fundamentally related to its geometry and intrinsic properties. The general resistance calculation formula is: $R = \rho l / S$

Symbol Explanation:

• $R$: Denotes the electrical resistance of the material (measured in Ohms, $\Omega$).

• $\rho$: Represents the resistivity of the material (measured in Ohm-meters, $\Omega \cdot m$). Resistivity is an intrinsic property of the material that quantifies how strongly it resists electrical current.

• $l$: Indicates the length of the resistor (measured in meters, $m$).

• $S$: Refers to the cross-sectional area of the resistor (measured in square meters, $m^2$).

  Working Mechanism: When a flexible resistive material is subjected to force, its length ($l$) and/or cross-sectional area ($S$) change. If the material's resistivity ($\rho$) remains constant, these geometric changes directly alter the resistance ($R$). For example, stretching a conductive material increases its length and decreases its cross-sectional area, both leading to an increase in resistance. Compressing it would have the opposite effect. Different structures are designed to enhance sensitivity and enable the detection of various types of forces (tensile, normal pressure, shear force). These structures modulate the change in $l$ or $S$ in response to specific force directions.

Representative Designs:

• Conductive Sensitive Layer Designs: To achieve diverse sensing capabilities, various microstructures are incorporated into the conductive sensitive layer. These include bending and wrinkling structures (for stretchability and small force sensitivity), interlocking microstructures (for improved sensitivity and multi-directional force detection), biomimetic microstructures (mimicking skin textures for conformability), foam-like structures (for compressibility and elasticity), microcolumn array structures, and pyramid array structures.
  • Pyramid Array Sensor: As shown in Figure 2a, this sensor uses a PDMS flexible substrate with a pyramid-shaped structure. A conductive material like PEDOT:PSS/PUD is coated onto the pyramids. Force causes deformation of the pyramids, changing the contact area and thus resistance.
  • Microcolumn Array Sensor: As shown in Figure 2b, it uses a PDMS substrate with a microcolumn array structure. A Ti/Au film is deposited on its surface. Deformation of the microcolumns under force alters the resistance.
  • Interlocking Microstructures: As illustrated in Figure 2c, this sensor consists of two layers with interlocking fibers. Mechanical deformation induces changes in numerous microscopic contacts between adjacent fibers, which are detected as resistance changes, allowing for normal pressure, shear force, and torsional force detection.
• Multi-axis Force Detection:
  • Carbon Nanotubes (CNTs) in Elastic Substrate: A sensor (Figure 2d) embeds CNTs as sensing units within a fully elastic substrate (e.g., PDMS). Shear loads induce distinct deformation in the PDMS, causing resistance changes in different CNT patterns, enabling shear force detection.
  • Liquid Metal in Serpentine Configuration: A design (Figure 2e) incorporates liquid metal in a serpentine pattern within a PDMS substrate. When shear force is applied, one piezoresistor (conductive liquid metal path) is compressed (length reduction), and another is stretched (length increase), leading to opposite resistance changes that allow shear force sensing.
  • Bionic Skin Sensor: As shown in Figure 2f, this sensor uses two layers: one of carbon nanotubes/graphene oxide and another of graphene oxide/PDMS. Under normal force, micro-holes in the lower layer compress, increasing contact area and reducing resistance. Under shear force, the upper layer stretches, increasing its resistance. These opposing changes help distinguish between normal and shear forces.

4.1.2. Capacitive Force Sensors

Core Idea: Capacitive flexible force sensors detect force signals by measuring changes in electrical capacitance.

Theoretical Basis: The capacitance between two parallel plates is determined by the geometric arrangement and the dielectric material between them. The formula for calculating the capacitance of two parallel plates is: $C = \mathfrak { \varepsilon } _ { 0 } \varepsilon _ { r } A / d$

Symbol Explanation:

• $C$: Denotes the capacitance (measured in Farads, F).

• $\mathfrak { \varepsilon } _ { 0 }$: Is the vacuum permittivity (a physical constant, approximately $8.854 \times 10^{-12} \text{ F/m}$).

• $\varepsilon _ { r }$: Is the relative permittivity (or dielectric constant) of the material between the plates (dimensionless). This indicates how easily an electric field can pass through the material.

• $A$: Is the effective area of the plates (measured in square meters, $m^2$).

• $d$: Is the separation distance between the plates (measured in meters, $m$).

  Working Mechanism: When a force is applied to a capacitive sensor, it typically causes either the distance ($d$) between the conductive plates to change, or the effective overlapping area ($A$) of the plates to change, or the relative permittivity ($\varepsilon_r$) of the dielectric material between the plates to change. Any of these variations alters the capacitance, which is then measured to quantify the applied force.

Representative Designs:

• Interlaced Parallel-Plate Electrode Structure: As shown in Figure 3a, a typical design comprises multiple electrodes and a dielectric layer. Normal force changes the distance between plates. Shear force changes the effective overlapping area between a central plate and surrounding plates, allowing for magnitude and direction of shear force detection.

• Shieldable Soft Force Sensor: To mitigate external electromagnetic interference, this design (Figure 3b) incorporates grounded shielding layers above, below, and around the traditional parallel plates. This helps prevent external objects from inducing capacitance changes and distorting readings.

• Dielectric Constant Modulation: Some sensors alter the capacitance by changing the relative dielectric constant between the plates.

  • A sensor (Figure 3c) with five small coplanar electrodes. When force is applied to convex points, the flexible PDMS substrate deforms, compressing the air layer. This changes the relative dielectric constant above the plates. Normal and shear forces are determined by changes in capacitance between the central and surrounding electrodes.
  • For coplanar electrodes, the capacitance calculation is more complex than the parallel plate formula. The paper cites a specific formula: $C = \frac { 7 } { 40 } \times \varepsilon \times b \times \left( \frac { a } { d } \right) ^ { \frac { 1 } { 3 } + \frac { 2 } { \pi } \arctan \left( \frac { d } { b } \right) } \times \left( 1 + \frac { b } { a } \right) ^ { \frac { 1 } { 3 } } \times \left( \frac { d } { a } \right) ^ { - \frac { 1 } { 3 } }$

  Symbol Explanation:

  • $C$: Denotes the capacitance.
  • $\varepsilon$: Represents the permittivity of the dielectric material.
  • $a$: Represents the length of the plates.
  • $b$: Represents the width of the plates.
  • $d$: Denotes the distance between the plates.

• Bioinspired Electronic Skin: A sensor (Figure 3d) inspired by human skin, uses carbon nanotube (CNT) electrodes embedded in a polyurethane (PU) matrix. The top layer has molded square pyramids. Applied pressure causes elastic deformation of the PU, minimizing viscoelastic effects and enhancing sensitivity. An array of 25 capacitors allows determining the form of force application.

• Bending-Insensitive Design: To address performance degradation under bending, a sensor (Figure 3e) employs a pyramid-shaped structure with truncated pyramids on both sides. This design concentrates bending strain in the regions between the truncated pyramids, preventing it from localizing within the sensing elements and thus achieving bending insensitivity.

4.1.3. Magnetic Force Sensors

Core Idea: Flexible magnetic force sensors rely on magneto-sensitive effects to transform variations in a magnetic field into electrical signals, which are then processed to deduce force readings. Their structure typically integrates magnetic materials within a soft substrate.

Theoretical Basis: Various quantum mechanical and physical phenomena underpin magnetic sensing:

• Hall Effect: Discovered by Edwin Hall, this occurs when a current flows through a conductor in a magnetic field perpendicular to the current. The Lorentz force deflects charge carriers, creating a potential difference (Hall voltage) perpendicular to both the current and the magnetic field. The Hall voltage can be calculated using the formula: $V _ { H } = I B /$ qnd

  Symbol Explanation:

  • V _ { H }: Is the Hall voltage (measured in Volts, V).

  • $I$: Is the current flowing through the material (measured in Amperes, A).

  • $B$: Is the magnetic flux density (measured in Teslas, T).

  • $q$: Is the elementary charge (approximately $1.602 \times 10^{-19}$ Coulombs, C).

  • $n$: Is the carrier concentration (number of charge carriers per unit volume, measured in $m^{-3}$).

  • $d$: Is the thickness of the material (measured in meters, m).

    Working Mechanism: In a Hall effect sensor (Figure 4a), a flexible columnar structure embeds an Nd magnet. A planar Hall magnetoresistor (PHMR) detects the magnet's movement caused by shear forces, translating magnetic field changes into Hall voltage variations.

• Giant Magnetoresistance (GMR) Effect: Discovered by Fert and Grünberg, this quantum mechanical phenomenon occurs in multilayered structures (alternating ferromagnetic and non-magnetic conductive layers). An external magnetic field changes the relative magnetization directions of the ferromagnetic layers, altering electron spin scattering and thus the electrical resistance. The resistance change can be quantified using the formula: R $= R _ { 0 } ( 1 - \Delta R / R _ { 0 } )$

  Symbol Explanation:

  • R: Represents the resistance in the presence of a magnetic field.

  • R _ { 0 }: Represents the resistance in the absence of a magnetic field.

  • $\Delta R$: Denotes the change in resistance due to the applied magnetic field.

    Working Mechanism: When magnetic moments of ferromagnetic layers are parallel, electron scattering is minimized, resulting in lower resistance. When antiparallel, scattering increases, leading to higher resistance. A flexible tactile sensor based on GMR (Figure 4b) uses a hollow PDMS membrane and a pyramid-structured magnetic film with a GMR sensor. When a magnet approaches, the magnetic field intensity decreases, reducing GMR resistance. Under compression, the field strengthens, increasing resistance. This inverse relationship differentiates contact and proximity signals.

• Tunnel Magnetoresistance (TMR) Effect: Similar to GMR but involves a very thin insulating barrier (e.g., MgO or $\mathrm { { A l } } _ { 2 } \mathrm { { O } } _ { 3 }$) between two ferromagnetic layers, forming a magnetic tunnel junction (MTJ). Electrons tunnel through the insulating layer. The resistance change ratio of TMR can be calculated using the formula: TMR $= ( R _ { A P } - R _ { P } ) \times 1 0 0 \% / R _ { P }$

  Symbol Explanation:

  • TMR: Represents the Tunnel Magnetoresistance ratio (expressed as a percentage).

  • R _ { A P }: Represents the resistance when the magnetization directions of the ferromagnetic layers are in antiparallel alignment.

  • R _ { P }: Represents the resistance when the magnetization directions of the ferromagnetic layers are in parallel alignment.

    Working Mechanism: TMR exhibits a significantly higher magnetoresistance ratio than GMR. A tactile sensor array based on TMR (Figure 4c) can be integrated into robotic hands for grasping objects of varying hardness by detecting resistance changes in the MTJ array.

• Anisotropic Magnetoresistance (AMR) Effect: This effect depends on the relative orientation between the material's magnetization direction and the current flow direction. Resistance changes as the angle between these directions varies due to differing electron scattering. The formula for resistance change is expressed as: $R = R _ { 0 } + \Delta R c o s ^ { 2 } \theta$

  Symbol Explanation:

  • $R$: Represents the resistance.

  • R _ { 0 }: Represents the reference resistance.

  • $\Delta R$: Denotes the resistance variation.

  • $\theta$: Is the angle between the magnetization direction and the current direction.

    Working Mechanism: A flexible magnetic film leveraging the AMR effect (Figure 4d) can be used for skin monitoring, such as detecting joint movements and page-turning control in wearable sensor technology.

4.1.4. Optical Waveguide Force Sensors

Core Idea: Flexible optical force sensors measure external forces by detecting changes in optical signals transmitted through optical fibers embedded within a flexible substrate.

Theoretical Basis: The underlying working principles primarily involve optical phenomena such as intensity modulation, interference effects, wavelength shifts, scattering variations, or resonance effects. When a force deforms the fiber or its surrounding medium, these optical properties change, which can then be measured.

Working Mechanism: The general architecture includes optical fibers, light source detectors, and flexible substrates. The optical fibers transmit light. When the sensor is subjected to external force, the deformation of the flexible substrate (and thus the embedded fibers) causes a measurable change in light characteristics (e.g., intensity, phase, or wavelength). This change is detected by the light source detectors and quantified to determine the applied force.

Representative Designs:

• Multi-point Optical Triaxial Force Sensor: As depicted in Figure 5a, this sensor has a flexible backlight layer on top, a semi-transparent porous rubber pressure-sensitive layer in the middle, and a thin-film flexible imaging device at the bottom. Light passing through the porous rubber scatters at pore interfaces. When pressure is applied, the rubber deforms, changing pore spacing and light source positions, which alters the light spot patterns captured by the imaging device. Analyzing these patterns allows multi-point 3-axis pressure distribution reconstruction.
• Bioinspired Flexible Optical Force Sensor (Human Finger Skin): Inspired by human finger skin, this sensor (Figure 5b) features a fingerprint-like surface with parallel ridges and a multi-layered structure of resins and PDMS with varying stiffness. Force and slip stimuli deform optical microfibers, leading to variations in output light intensity, thus providing responsive reaction to external stimuli.
• Flexible Multi-axis Force Sensor with U-shaped Micro-nano Fibers: Illustrated in Figure 5c, this sensor embeds U-shaped micro-nano fibers within an elastomer membrane that has a dome-shaped protrusion. Force applied to the dome deforms the fibers, inducing changes in the temporal resolution of light transmission, enabling external force detection.
• Flexible Optical Tactile Sensor with Plasmonic Optical Fiber: As shown in Figure 5d, this sensor incorporates a soft plasmonic optical fiber made of gold nanoparticles and elastomer composites. Embedded in a sandwich structure, it transforms external mechanical stimuli into interpretable optical signal changes for high-precision force detection.

4.1.5. Other Types of Force Sensors

Beyond the main categories, other principles are utilized, often for specialized advantages.

• Triboelectric Force Sensors (TENG Sensors): Core Idea: These are self-powered sensors that measure force by generating charge transfer through contact-separation or sliding friction between two dissimilar materials (the triboelectric effect). The concept was introduced by Wang in 2012, exploiting the coupling of electrostatic induction and contact electrification to convert mechanical energy into electrical energy. Working Mechanism: When two materials with different electron affinities come into contact and then separate, or slide against each other, electrons are transferred between their surfaces. This generates static charges, which, through electrostatic induction, can create an electrical signal (voltage or current) that correlates with the applied mechanical force or movement. Representative Designs:

  • Triboelectric Array Sensor: As seen in Figure 6a, this sensor is fabricated on a polyimide (PI) film using laser direct writing technology, with conductive patterned laser-induced graphene (LIG) as electrodes. This creates a flexible, high-resolution tactile sensor array (TSA) for self-powered real-time tactile sensing.
  • Bioinspired Wearable Sensor: A flexible wearable sensor (Figure 7c) based on a self-powered solid-liquid triboelectric nanogenerator mimics a bionic shark skin-like microstructure. It is used for gait monitoring and real-time rehabilitation tracking. Strengths: Self-powered (no external power supply needed), high sensitivity, integrability, versatile configurations, compact size, low cost. Limitations: Prone to performance degradation due to environmental humidity and material aging. Currently often limited to normal force detection.

• Ion-Electronic Double Layer (EDL) Force Sensors: Core Idea: These devices operate based on the electric double layer (EDL) mechanism, which involves the redistribution of mobile ions at the interface between an ionic material (e.g., hydrogels, ionogels, ionic liquids) and an electrode surface. Working Mechanism: When subjected to mechanical deformation, the change in pressure or strain alters the interface between the ionic material and the electrode. This causes variations in interfacial ion concentration, which in turn leads to measurable changes in capacitance or electrical potential. These changes serve as the pressure signals. Representative Designs:

  • Hydrogel-based EDL Sensor: Bai et al. (2020) demonstrated a hydrogel-based EDL sensor with sub-kPa sensitivity and long-term mechanical stability for epidermal pressure monitoring.
  • Ion-electronic Skin: Shi et al. (2023) developed an ion-electronic skin by embedding microstructured ion gels in an elastomer matrix. This showed promise for robot manipulation and object recognition (Figure 8b). Strengths: Ultra-high sensitivity under low-pressure conditions, superior conformability, favorable biocompatibility with biological interfaces. Limitations: Signal instability under environmental fluctuations, difficulties in encapsulating ionic media, concerns regarding long-term operational reliability, material toxicity, poor environmental stability, and inability to detect shear forces.

• Hybrid Force Sensors: Core Idea: These sensors integrate multiple sensing principles to enhance functionality, overcome individual limitations, and achieve multi-modal detection. The most common combination is resistive and capacitive sensing. Working Mechanism: By combining different transduction mechanisms, hybrid sensors can simultaneously measure various parameters or detect forces in multiple axes with improved accuracy or robustness. Each component contributes its specific advantage. Representative Designs:

  • Multi-axis Resistive-Capacitive Sensor: A sensor (Figure 6b) employs a multi-layer structure. Normal force is detected by a central parallel capacitor, while the magnitude and direction of shear force are determined using a potentiometer-like resistive structure. It requires only three wires for operation.
  • Multi-axis Capacitive-Resistive Sensor: Another design (Figure 6c) merges capacitive and resistive principles for three-axis force detection. A resistive sensor is used for normal force, and an interlaced parallel plate electrode structure for tangential force (shear force).
  • Hybrid Sensor for Force and Proximity: A sensor (Figure 6d) combines a capacitive sensor for proximity detection and a resistive sensor for normal force measurement. This enables applications in robot collaboration by sensing both contact and near-field presence. Strengths: Enhanced functionality, multi-modal detection (e.g., force and temperature, force and proximity), ability to overcome limitations of individual sensor types. Limitations: Higher design and manufacturing costs, more complex signal processing requirements.

5. Experimental Setup

As a review paper, this article does not present new experimental setups or collect new data. Instead, it synthesizes and discusses the experimental findings and methodologies presented in the cited literature. Therefore, this section will discuss the general characteristics of "datasets" and "evaluation metrics" as they apply to flexible force sensor research, based on the paper's overview.

5.1. Datasets

In the context of flexible force sensor research, "datasets" refer to the types of input stimuli and output signals typically collected to characterize sensor performance and validate applications. The paper implicitly discusses the following types of data generated and collected by these sensors:

• Force Data: Measurements of normal pressure, shear force, and tensile force applied to the sensor. This can range from static loads to dynamic, rapidly changing forces.

• Deformation Data: Information on bending, stretching, and twisting conditions, often correlated with the applied force.

• Physiological Data: In health monitoring applications, flexible sensors collect data such as heart rate, respiration patterns, blood pressure, gait patterns, joint movements, grip strength, and muscle activity. For instance, Figure 7a and 7c show sensors monitoring human kinematic functions and gait, respectively. Figure 7d shows a system for monitoring penile expansion and stiffness.

• Robotic Interaction Data: In robotic operations, sensors collect data on grasping force, object slippage, contact pressure distributions, temperature (for multi-modal sensors), and proximity to objects or humans. For example, Figure 8b shows a sensor used for object recognition based on tactile data, and Figure 8d illustrates multimodal perception of pressure and temperature for robotic interaction.

• Time-series Data: Many applications involve continuous monitoring, generating time-series data of sensor responses over periods, which is crucial for dynamic response analysis and long-term stability assessment.

  These "datasets" are chosen to demonstrate a sensor's ability to accurately and reliably perceive the intended physical stimuli in a given application domain. The effectiveness is judged by how well the sensor's output correlates with the actual physical phenomenon and its ability to distinguish between different stimuli.

5.2. Evaluation Metrics

The paper explicitly lists key performance parameters used to assess flexible force sensors. For a beginner-friendly explanation, here are the conceptual definitions for each, as they are standard in sensor evaluation and their formulas are not provided in the paper's main text.

• Sensitivity:

  • Conceptual Definition: Sensitivity quantifies how much the sensor's output signal changes in response to a change in the measured force input. A higher sensitivity means the sensor can detect smaller variations in force, which is crucial for high-resolution force mapping or acquiring subtle physiological signals.
  • Mathematical Formula: In its simplest form, for a linear sensor, sensitivity ($S_e$) can be defined as: $S_e = \frac{\Delta \text{Output}}{\Delta \text{Input}}$
  • Symbol Explanation:
    • $\Delta \text{Output}$: The change in the sensor's output signal (e.g., change in resistance, capacitance, voltage, or current).
    • $\Delta \text{Input}$: The change in the applied force (e.g., in kPa or N). (Note: The units for sensitivity often reflect the output change per unit of force, e.g., $\text{kPa}^{-1}$ for resistive/capacitive sensors or mV/kPa for potentiometric).

• Detection Range:

  • Conceptual Definition: The detection range (or operating range) specifies the minimum and maximum values of the force that the sensor can accurately and reliably measure. Sensors are designed for different ranges depending on the application, from very low pressures (e.g., for delicate touch) to high pressures (e.g., for heavy robotic manipulation).
  • Mathematical Formula: There is no single formula for detection range; it is defined by the interval $[F_{min}, F_{max}]$, where $F_{min}$ is the minimum detectable force and $F_{max}$ is the maximum measurable force.
  • Symbol Explanation:
    • $F_{min}$: Minimum detectable force.
    • $F_{max}$: Maximum measurable force.

• Hysteresis:

  • Conceptual Definition: Hysteresis refers to the phenomenon where a sensor's output for a given input force differs depending on whether the force is increasing or decreasing. It represents the "memory" effect of the sensor, often caused by viscoelastic properties of materials. Low hysteresis is desirable for consistent and reproducible measurements.
  • Mathematical Formula: Hysteresis (H) is often quantified as the maximum difference between the increasing and decreasing output values for a given input, expressed as a percentage of the full-scale output (FSO). $H = \frac{\text{Max } | \text{Output}_{increasing} - \text{Output}_{decreasing} |}{\text{FSO}} \times 100\%$
  • Symbol Explanation:
    • $\text{Output}_{increasing}$: Sensor output when force is increasing.
    • $\text{Output}_{decreasing}$: Sensor output when force is decreasing.
    • FSO: Full-scale output, the maximum output change over the sensor's range.

• Linearity:

  • Conceptual Definition: Linearity describes how closely the sensor's output signal is directly proportional to the applied force input over its detection range. A perfectly linear sensor would have a straight-line relationship between input and output. Good linearity simplifies calibration and makes the sensor's readings easier to interpret and use in calculations.
  • Mathematical Formula: Linearity (L) is often expressed as the maximum deviation from a best-fit straight line, as a percentage of the full-scale output (FSO). $L = \frac{\text{Max } | \text{Output}_{actual} - \text{Output}_{ideal\_linear} |}{\text{FSO}} \times 100\%$
  • Symbol Explanation:
    • $\text{Output}_{actual}$: Actual sensor output.
    • $\text{Output}_{ideal\_linear}$: Output predicted by an ideal linear relationship.
    • FSO: Full-scale output.

• Response Time:

  • Conceptual Definition: Response time measures how quickly a sensor can react and settle to a new, stable output reading after a sudden change in the applied force. A fast response time is critical for applications involving dynamic forces or rapid interactions, such as robotic manipulation or real-time physiological monitoring.
  • Mathematical Formula: Response time is typically defined as the time taken for the sensor's output to reach a certain percentage (e.g., 90% or 95%) of its final stable value after a step input change.
  • Symbol Explanation: Not typically represented by a single formula but rather a measured duration in milliseconds (ms) or seconds (s).

• Cross-talk:

  • Conceptual Definition: Cross-talk refers to the undesirable phenomenon in multi-axis or array sensors where a force applied along one axis (e.g., normal force) inadvertently produces a significant signal in another axis (e.g., shear force), or where an input to one sensor element affects the reading of an adjacent element. Minimizing cross-talk is essential for accurate multi-dimensional force decomposition and spatial resolution.
  • Mathematical Formula: Cross-talk (CT) can be quantified as the ratio of the unintended output signal (on an un-stimulated axis) to the intended output signal (on the stimulated axis), often expressed as a percentage. $CT = \frac{\text{Output}_{unintended\_axis}}{\text{Output}_{intended\_axis}} \times 100\%$
  • Symbol Explanation:
    • $\text{Output}_{unintended\_axis}$: Signal registered on an axis that was not directly stimulated.
    • $\text{Output}_{intended\_axis}$: Signal registered on the directly stimulated axis.

5.3. Baselines

In a review paper, "baselines" typically refer to existing or alternative technologies against which new developments are implicitly or explicitly compared. For this review of flexible force sensors, the "baselines" are:

• Traditional Rigid Force Sensors: These form the primary baseline, characterized by their lack of flexibility, bulkiness, and inability to conform to complex surfaces. The paper highlights flexible sensors' advantages over these.

• Different Sensing Mechanisms: Within flexible sensors, each sensing principle (resistive, capacitive, magnetic, optical, etc.) serves as a baseline for the others. The paper's comparative tables (Table 1 and Table 2) directly compare these different mechanisms against each other based on various performance metrics. For example, a resistive sensor might be compared to a capacitive sensor in terms of sensitivity or cost.

• Earlier Generations of Flexible Sensors: The paper implicitly compares current state-of-the-art flexible sensors with their predecessors, demonstrating technological evolution (e.g., from simple designs to bionic microstructures or bending-insensitive designs).

  These comparisons are representative because they showcase the progress and trade-offs inherent in flexible sensor development. They help identify which sensing mechanism or design is best suited for particular application requirements by highlighting the relative strengths and weaknesses of each approach.

6. Results & Analysis

6.1. Core Results Analysis

The paper's core results are presented through a comprehensive comparative analysis of various flexible force sensor types based on their sensing mechanisms, performance metrics, and suitability for different applications. The key takeaway is that each sensor type has distinct advantages and limitations, necessitating application-specific design and selection.

• Resistive Sensors: Are structurally simple and easy to manufacture, making them cost-effective. However, they typically exhibit relatively low sensitivity compared to capacitive or EDL sensors and are prone to temperature-induced errors. For multi-axis detection, they often face challenges, being susceptible to crosstalk (as indicated in Table 2).
• Capacitive Sensors: Are characterized by high sensitivity and low power consumption, making them suitable for multi-point detection. Their linearity is generally good, and hysteresis is low. However, they are vulnerable to electromagnetic interference and environmental humidity, and often require sophisticated signal processing. They are easier for multi-axis force detection, where algorithm decoupling can be possible.
• Magnetic Sensors: Offer non-contact measurement and strong resistance to environmental interference, making them suitable for long-life applications. Their main drawbacks include the necessity of an additional magnetic field source and susceptibility to external magnetic field disturbances. Their system complexity is high. For multi-axis detection, they have medium cross-talk but allow for decoupling through algorithms or structural design.
• Optical Sensors: Provide anti-electromagnetic interference capabilities and high sensitivity by detecting changes in light signals. However, they suffer from high manufacturing and integration costs and complex systems. They generally exhibit excellent linearity and lower cross-talk for multi-axis detection, though signal reading can be complex.
• Triboelectric Sensors: A significant advantage is their self-powered detection capability, eliminating the need for an external power supply. However, they are prone to signal drift and humidity sensitivity, and currently are limited to normal force detection (cannot measure shear or multi-axis forces).
• EDL-based (Ionic-Electronic) Sensors: These show ultra-high sensitivity at low pressure and excellent flexibility compatibility with slight hysteresis. However, they face challenges with material toxicity, poor environmental stability, and are currently limited to normal force detection (no shear force capability).
• Hybrid Sensors: These integrate multiple sensing principles to enhance performance and broaden functionality. They can achieve multi-modal detection (e.g., force and temperature) and effectively detect multi-axis forces. The trade-off is higher design and manufacturing costs, more complex signal processing, and difficult signal fusion. Cross-talk can be negligible to medium depending on the design.

Key Trade-offs: The paper emphasizes that sensor design inherently involves trade-offs. For example, increasing the porosity of a sensing layer might improve sensitivity but reduce durability and introduce nonlinearity. Achieving low hysteresis might require specific elastomers, potentially compromising stretchability. Therefore, multi-parameter co-design frameworks are proposed to jointly optimize sensing principles, materials, microstructures, and packaging based on specific application requirements.

The role of material properties and microstructural design is highlighted across all types. Soft, compressible, or porous materials (e.g., PDMS foam, aerogels, ionic gels) generally offer high sensitivity due to pronounced deformation under light loads. In contrast, high-modulus materials provide broader detection ranges. Regular and symmetrical microstructures (e.g., pyramid arrays, microdomes) enhance linearity and repeatability, while viscoelastic materials can lead to hysteresis and signal delay.

6.2. Data Presentation (Tables)

The following are the results from Table 1 and Table 2 of the original paper:

The following are the results from Table 1 of the original paper:

The following are the results from Table 2 of the original paper:

6.3. Ablation Studies / Parameter Analysis

This section typically describes how researchers dissect their proposed models or systems to understand the contribution of individual components or the impact of different parameters. As this paper is a review, it does not present its own ablation studies. However, it implicitly performs a "parameter analysis" by highlighting how different material properties and microstructural designs (analogous to "parameters") influence sensor performance across the various sensing mechanisms:

• Material Properties: The choice of conductive sensitive materials (e.g., PEDOT:PSS/PUD, Ti/Au film, carbon nanotubes, liquid metal, graphene) in resistive sensors, dielectric layers (e.g., PDMS, air gaps, PU) in capacitive sensors, magnetic materials in magnetic sensors, optical fibers in optical sensors, and ionic materials (e.g., hydrogels, ionogels) in EDL sensors directly dictates sensitivity, detection range, flexibility, and environmental stability. For example, soft, compressible, or porous materials generally lead to higher sensitivity but might compromise durability.

• Microstructural Design: The geometric configurations (e.g., pyramid arrays, microcolumn arrays, interlocking microstructures, serpentine patterns, truncated pyramids, U-shaped fibers, fingerprint-like surfaces, microstructured ion gels) are critical "design parameters" that significantly affect how mechanical stress is distributed and converted into a measurable signal. These designs are engineered to enhance sensitivity, linearity, stretchability, bending insensitivity, and the ability to detect multi-axis forces or proximity. For instance, specific microstructures can concentrate stress, increase contact area, or modulate optical path changes more effectively.

• Integration Strategies: For hybrid sensors, the "parameter analysis" involves understanding how the combination of different sensing principles (e.g., resistive + capacitive) leads to enhanced functionality (multi-axis, multi-modal detection) compared to individual mechanisms.

  In essence, the entire review serves as a broad-scale "parameter analysis" of the design space for flexible force sensors, showing how researchers have optimized various aspects to achieve diverse performance characteristics for different applications.

7. Conclusion & Reflections

7.1. Conclusion Summary

This review paper provides a comprehensive overview of flexible force sensors, categorizing them by their underlying sensing mechanisms (resistive, capacitive, magnetic, optical, triboelectric, ion-EDL, and hybrid). It thoroughly explains their working principles, highlights design improvements and key innovations, and discusses their applications in fields such as wearable electronics, human-machine interaction, robotics, and biomedical monitoring. The paper concludes that while significant progress has been made in terms of sensitivity, speed, flexibility, and integration, substantial challenges persist. These include issues with signal stability, scalable manufacturing, application-specific adaptation, and the inherent trade-off between performance and fabrication complexity, which currently limit widespread real-world adoption. The paper advocates for future research to focus on developing multifunctional, self-powered, and biocompatible sensors with long-term reliability, and emphasizes the essential role of integrating these sensors with artificial intelligence (AI), edge computing, and wireless communication for the next generation of smart sensing systems.

7.2. Limitations & Future Work

The authors explicitly point out several limitations and suggest future research directions:

Current Challenges (Limitations):

• Sensing Accuracy: Ensuring high precision and reliability across diverse operating conditions remains a challenge.
• Long-term Stability: Maintaining consistent performance over extended periods, especially under repeated use and exposure to various environmental factors (e.g., humidity, temperature), is critical.
• Mechanical Durability: Sensors need to withstand significant and repeated mechanical deformations without degradation or failure.
• Scalable Manufacturing: Developing cost-effective and efficient methods for large-scale production of complex flexible sensor structures is crucial for widespread adoption.
• Application-Specific Adaptation: Tailoring sensors precisely to the unique demands of different applications often requires considerable customization.
• Performance vs. Fabrication Complexity Trade-off: There is often a difficult balance between achieving high performance and maintaining a manageable fabrication cost and complexity.
• Environmental Sensitivity: Specific sensor types (e.g., triboelectric, capacitive) are vulnerable to environmental humidity or electromagnetic interference.
• Limited Multi-axis Detection: Some emerging sensor types (e.g., triboelectric, EDL-based) are currently limited to normal force detection and lack shear force or multi-axis sensing capabilities.
• Material Toxicity: Concerns regarding material toxicity (e.g., in some EDL sensors) and biocompatibility are critical for medical applications.

Future Research Directions:

• Innovation of Materials and Microstructures:
  • Development of high-performance flexible materials with superior conductivity, mechanical flexibility, and environmental stability (e.g., advanced conductive polymers, nanocomposites, 2D materials like graphene, MXene).
  • Incorporation of smart materials (e.g., shape memory alloys, self-healing materials) to enhance reliability and extend service life.
  • Implementation of rational geometric configurations and multi-scale micro-nano patterning techniques (e.g., porous architectures, wrinkled surface designs, biomimetic microstructures) to modulate stress distribution and improve sensitivity, dynamic response, and stretchability.
  • Development of multifunctional heterogeneous integration structures (e.g., combining resistive-capacitive, triboelectric, or optical mechanisms) to broaden applications and enhance response to diverse stimuli.
• Combination with Deep Learning:
  • Application of deep learning for signal denoising, feature extraction, and nonlinear compensation.
  • Utilization of CNNs and RNNs to analyze multi-dimensional spatiotemporal data for improved signal interpretation.
  • Employment of GANs for data augmentation to address insufficient data.
  • Integration for high-level information fusion and pattern recognition (e.g., precise contact force, slip detection, material property identification in robotics; physiological signal analysis for disease diagnosis in health monitoring).
  • Development of efficient, low-power edge computing frameworks for real-time deployment on resource-limited devices.
• Organ Monitoring with Implanted Electrodes:
  • Development of implantable electrodes integrated with flexible force sensors for real-time monitoring of organ mechanical activities (e.g., heart, muscular system, digestive system).
  • Ensuring long-term biocompatibility, stability, and signal reliability within complex physiological environments.
  • Miniaturization of implantable systems.
  • Advancements in wireless data transmission and sustainable energy supply for implantable devices.
  • Development of novel biocompatible materials, refinement of sensor packaging techniques, and incorporation of low-power electronic technologies.

7.3. Personal Insights & Critique

This review paper provides an excellent, structured, and comprehensive overview of the rapidly evolving field of flexible force sensors. Its classification based on sensing mechanisms and the detailed comparison tables are particularly valuable for researchers, offering a quick reference for the strengths and weaknesses of each approach. The emphasis on material science and microstructural design as critical drivers of performance is well-placed, reflecting the interdisciplinary nature of the field.

The discussion of future directions, especially the integration with deep learning and the potential for implantable electrodes, highlights the transformative potential of these sensors. The concept of a multi-parameter co-design framework is a crucial insight, acknowledging that real-world applications require careful balancing of often conflicting performance metrics rather than optimizing a single parameter in isolation. This holistic perspective is a significant strength of the review.

Potential issues or areas for improvement (Critique):

• Lack of Unified Performance Metrics: While the paper provides tables for comparison, the units and metrics themselves sometimes vary across different studies (e.g., sensitivity in kPa⁻¹ vs. mV/kPa vs. dB/N), making direct quantitative comparison challenging, a common issue in review papers that aggregate diverse research. This points to a broader need in the field for standardization of reporting protocols.
• Material Characterization Depth: While material importance is emphasized, a deeper dive into the specific properties of commonly used flexible substrates (e.g., PDMS, PUD, elastomers) and conductive elements (e.g., CNTs, graphene, liquid metals) and their impact on long-term mechanical stability, fatigue resistance, and creep could further enhance the review.
• Manufacturing Techniques: The review briefly mentions fabrication methods but could benefit from a dedicated section detailing common manufacturing techniques (e.g., soft lithography, 3D printing, inkjet printing, laser direct writing) and their respective advantages and limitations for producing these complex microstructures and integrating diverse materials. This is a critical aspect for scalable manufacturing, a challenge identified by the authors.
• Quantitative Comparison of Hybrid Sensors: While hybrid sensors are discussed, their "performance metrics" in the tables are less detailed than for single-mechanism sensors, often stating "Depends on combined mechanisms." A more rigorous quantitative comparison of how hybrid approaches quantifiably improve specific metrics (e.g., cross-talk reduction, enhanced detection range) would be beneficial, perhaps by presenting ranges or representative values.
• Environmental Robustness Beyond Temperature/EM: Beyond electromagnetic and thermal disturbances, flexible sensors, especially in wearable and implantable contexts, face challenges from sweat, pH changes, biofouling, and mechanical abrasion. A more explicit discussion of these environmental factors and proposed solutions would add value.

Inspirations and Applications: The paper strongly inspires thinking about context-aware robotics, where robots can not only perceive forces but also interpret them intelligently using AI to perform delicate tasks or safely interact with humans. In personalized medicine, the vision of implantable flexible sensors monitoring organ activities in real-time opens vast possibilities for early disease detection, targeted therapy, and closed-loop biofeedback systems. The principles discussed could also be applied to structural health monitoring of flexible infrastructure, smart packaging that indicates handling stress, or even next-generation human-computer interfaces that go beyond simple touchscreens to capture nuanced physical interactions. The detailed comparison encourages a more informed selection of sensing modalities for novel applications, moving beyond a "one-size-fits-all" approach.