Symposium EQ15—Soft Matter Materials and Mechanics for Haptic Interfaces

Robotic fabrics and stretchable electronics are broad-category technologies that may play a role in next-generation haptic interfaces. In this talk, we will present two recent works and speculate on their applications to wearables.
First, we will present an implementation of robotic fabrics by integrating functional fibers into conventional fabrics using typical textile manufacturing techniques. We will introduce a set of actuating and variable-stiffness fibers, as well as printable in-fabric sensors, which allow for robotic closed-loop control of everyday fabrics while remaining lightweight and maintaining breathability. By designing actuators, sensors, and structural components that conform to a fiber-based geometry, our methodology shows that simple fabrics can be functionalized to create self-deploying structures, morphing airfoils, and damage-responsive tourniquets, among other wearable and haptic devices.
Second, we will present biphasic Ga–In (bGaIn): a printable, solid–liquid biphasic material that maintains a near-constant resistance under extreme strains; maintains direct, consistent, and stretchable electrical connections with conventional electronic components; and is mechanically stable when applied onto numerous soft materials. bGaIn is produced by thermally treating eutectic GaIn nanoparticles to create a mixture of liquid and crystalline solids. It shows a high initial conductivity of 2.06×10^6 S/m and near-constant resistance at strains over 1,000%. We have employed bGaIn as a stretchable interconnect to interface with commercial electronic components---including resistors, capacitors, light-emitting diodes (LEDs), operational amplifiers, and microcontrollers---by simply pressing the electronics into the bGaIn trace. Using bGaIn vertical interconnect accesses (VIAs) to connect traces on the top and bottom layers, we have converted conventional multilayer PCB designs into stretchable circuit board assemblies (SCBAs) that maintain performance while experiencing large strains. Finally, the applicability of bGaIn stretchable circuits to wearables will be shown with a stretchable signal conditioning circuit integrated into a smart sensing garment.

Humans excel at controlling their gripping force during manipulation tasks so that objects don’t slip, while avoiding too-strong pressures that damage soft objects. In this presentation, we develop a soft optical sensor for slipping measurement in gripping applications. The principle is micro-bending of a soft optical fiber by lateral forces and its effect on optical transmission from emitter to receiver. The optical fiber deformation is intensified by a stress-concentrating structure. Based on the data, we describe the empirical relationship between intensity and lateral force with a function.
When it comes to robotic grasping, measuring tactile information is critical for human-like performance. In the case of the human grasping process, cutaneous sensory input determines how grasping forces are modulated. Lateral-force tactile sensors for robotic hands, and sensors that can measure lateral forces experienced in human test cases, will help robots emulate human dexterity in manipulation tasks.
Considering the human case, we need to detect both slipping (lateral forces) and grasping forces (normal forces) in real-time, thereby holding objects with optimum forces--not crushing, but also not dropping the object. In that case, lateral forces and the relative slipping amount provide direct and indirect information for determining slippage and grasping normal force respectively. Therefore, it is natural to identify contact mechanisms, which can provide friction force and normal force information. Furthermore, based on the contact mechanism it can make tactile sensors be designed and analyzed in accordance with the situation.
Because our optic fibers are fabricable and elastic, we can make them available to analyze the contact mechanism assuming under Hooke’s laws. In particular, we measured lateral force by using an elastomeric optical fiber through the stress concentrating structure that deforms the fiber in response to the shear forces of the surface. In other words, we investigate the relationship between light power intensity variations and lateral forces and figure out its viability to grasping applications. Moreover, based on the stress concentrating structure, we investigate how variables related to surface contact dynamics depend on applied forces. In this study, the regions of lateral force sensing are in the 1 – 5 cm² range and the detected displacements are on the 1 – 5 mm scale. Although it could be changed by the stress concentrating structure due to its material and configuration, the lateral forces are low under 20N for this displacement range.

Human–machine interfaces are increasingly prevalent in society to enhance productivity and performance (e.g., automated driving), aid people with sensory and motor deficiencies (e.g., tactile displays, assistive exoskeletons), and allow remote tasks (e.g., surgery, bomb deactivation). This field, and particularly haptics and robotics, is gradually switching from hard and bulky materials to softer, lightweight, and conformable ones. As such, new electronic materials are needed to enable finer tasks and sensations, more realistic human–machine interactions, and less intrusive assistive devices. Additionally, ionic and electronic conductivity are both required to transduce biological to electronic signals and vice versa. For this purpose, we have developed a general approach to synthesize electronically and ionically conductive hydrogels with tunable mechanical properties. Instead of using rigid metal nanoparticles or stiff conductive polymers, we use a water-soluble derivative of 3,4-ethylene dioxythiophene (EDOT) functionalized with oligo-ethylene oxide side chains (EDOT-DEG) directly polymerized in a range of hydrogels commonly used for biological interfaces. The resulting conductive hydrogels have low impedance and resistance, and similar mechanical properties as the initial—non-electronically conductive—hydrogels. Importantly, the shape and size of the hydrogel is retained after the polymerization of the conductive PEDOT-DEG polymer. In this presentation, I will share our synthetic approach towards these materials, their characterization, and a vision of how they could address challenges in interfacing electronics with the human body.

Wearable haptic devices use intuitive tactile signals to enable alternative pathways for communication in “noisy” environments and immersive virtual reality experiences. These wearable devices often employ rigid electromechanical components, including onboard motors that require gearboxes for appropriate torques in squeeze cues and rotating eccentric masses for vibration cues [1]; incorporation of these rigid components hinders the construction of unobtrusive, comfortable, and lightweight wearables.

Soft haptic devices overcome this limitation by leveraging intrinsic compliances to apply haptic cues while maintaining flexibility and light weight, with most approaches utilizing either (i) compliant pneumatic systems composed of inflatable bladders or (ii) flexible electronic systems. Focusing on the former, typical pneumatic systems are composed of molded elastomers or bonded thermoplastic sheets which create chambers that can apply kinesthetic or tactile feedback when inflated [2]. However, these materials introduce challenges for seamless and discreet integration into everyday wearable devices, clothing, and other attire.

Here, we present a new approach for soft haptics based on textiles bonded with an adhesive thermoplastic. The adhesive thermoplastic allows tunable actuator geometries and strong bonds between textiles. Textiles are low-profile, ubiquitous, and most importantly, comfortable and conformal to the human body [3], and this approach enables easy and inconspicuous integration into clothing because textiles can be sewn into or bonded onto existing garments.

We selectively bond textiles to each other through heat and pressure to define internal bladders and pathways for flow of pressurized air, driving actuation of the haptic devices. The geometry of the customizable bladders facilitates squeeze, stretch, and vibration cues, while the channels route the pressurized air to the actuators through tailored pathways. The addition of a patterned intermediate layer between textiles permits the formation of these features by preventing adhesion in defined locations. Moreover, fluidic and systematic instabilities can be designed and exploited to provide simpler control [4] than existing pneumatic counterparts that rely on bulky electromechanical regulators for actuation [5].

Soft haptic devices based on textiles and fluidic actuation allow facile fabrication and operation, and the ease of integration into everyday garments and wide array of design possibilities highlight a promising strategy for integration in wearable devices. Our textile-based approach to wearable haptics can supplement the more traditional, yet often saturated, modes of communication (visual, aural, etc.) while delivering salient information to a user in an inconspicuous, lightweight, and comfortable manner.

[1] C. Pacchierotti, S. Sinclair, M. Solazzi, A. Frisoli, V. Hayward, and D. Prattichizzo, “Wearable Haptic Systems for the Fingertip and the Hand: Taxonomy, Review, and Perspectives,” IEEE Transactions on Haptics, vol. 10, no. 4, pp. 580–600, Oct. 2017, doi: 10.1109/TOH.2017.2689006.
[2] R. Niiyama, D. Rus, and S. Kim, “Pouch Motors: Printable/inflatable soft actuators for robotics,” in 2014 IEEE International Conference on Robotics and Automation (ICRA), May 2014, pp. 6332–6337. doi: 10.1109/ICRA.2014.6907793.
[3] V. Sanchez, C. J. Walsh, and R. J. Wood, “Textile Technology for Soft Robotic and Autonomous Garments,” Advanced Functional Materials, vol. 31, no. 6, p. 2008278, 2021, doi: 10.1002/adfm.202008278.
[4] A. Rajappan, B. Jumet, and D. J. Preston, “Pneumatic soft robots take a step toward autonomy,” Science Robotics, vol. 6, no. 51, Feb. 2021, doi: 10.1126/scirobotics.abg6994.
[5] E. M. Young, A. H. Memar, P. Agarwal, and N. Colonnese, “Bellowband: A Pneumatic Wristband for Delivering Local Pressure and Vibration,” in 2019 IEEE World Haptics Conference (WHC), Jul. 2019, pp. 55–60. doi: 10.1109/WHC.2019.8816075.

Immersive systems for virtual and augmented reality (VR/AR) will transform the way that we interact with computer-generated environments and, by extension, with one another. Interfaces that add spatio-temporally controlled physical sensations to well-developed audio and video components of VR/AR systems will qualitatively enhance the user experience. These technologies have myriad potential uses in remote health care, in prosthetic control, in physical rehabilitation, in social media, in entertainment and in training. This talk summarizes our most recent work in this area, with a focus on concepts in materials science and electrical engineering for thin, soft, lightweight sheets that embed wirelessly powered and addressed arrays of high-speed, millimeter-scale mechanical actuators, capable of gently laminating onto the skin. We highlight the capabilities through several example application areas, and we outline frontier opportunities for continued research.

Soft robotic garments represent a promising approach to provide a wearer with mobile assistance for haptic communication, grasping and reaching, and dynamic thermoregulation [1]. For these devices to be as wearable as everyday clothing, it is necessary for them to be created from compliant, lightweight, breathable, and stretchable materials, and textiles satisfy all of these needs. Textile-based soft robotic components, including sensors, actuators, and integration components (which connect to power and control elements) are gaining traction to enable these properties; however, they are traditionally developed separately using different manufacturing processes due to their diverse requirements in terms of mechanical and electrical properties. In order to build a full robotic garment with distributed sensing and actuation, subcomponents must be manufactured discretely and subsequently connected together using cut-and-sew processing. Such manufacturing strategies increase complexity and also reduce durability by adding potential failure points.

To overcome these challenges, we leverage weft knitting as a strategy to create integrated robotic textile materials. Weft knitting is a single additive manufacturing process that can form monolithic 3D shell shapes out of multiple materials [2,3]. Simultaneously, the ability to vary the knit stitch pattern (including changing stitch symmetry and geometry) and material choice enables mechanical diversity, which is necessary to optimize for separate component needs [4,5]. Here, we investigate the ability of weft knitting to create integrated textile actuators and sensors, and we show how the knit architecture, dictated by stitch pattern, leads to tailorable properties required for these components. Specifically, we explore how the stitch pattern changes knit mechanical properties, including stress-strain behavior and Poisson’s ratio, for use in fluidic actuators. We also study how varied knit architectures change the sensitivity, linearity, and range of knit strain sensors. We demonstrate fully integrated robotic textile components (sensors, actuators, and conductive signal lines) based on these strategies.

[1] Sanchez, V., Walsh, C. J., & Wood, R. J. Advanced Functional Materials, 31(6), 2021.
[2] Underwood, J. "The Design of 3D Shape Knitted Preforms." RMIT University, 2009.
[3] Hodgins, J., McCann, J., Grow, A., Albaugh, L., Mankoff, J., Matusik, W., & Narayanan, V. ACM Transactions on Graphics, 35(4), 2016.
[4] Poincloux, S., Adda-Bedia, M., & Lechenault, F. Physical Review X, 8(2), 2018.
[5] Bueno, M. A., & Camillieri, B. Structure and mechanics of knitted fabrics. In Structure and Mechanics of Textile Fibre Assemblies (2nd ed.). Elsevier Ltd. 2019

For the sense of sight, we can purchase HD screens to recreate nearly any image or movie. For touch, we are not yet able to recreate the variety of sensations from our everyday experiences. However, an accurate and rich recreation of tactile sensations could have broad implications in human machine interfaces, soft robotics, and disability rehabilitation. While most haptic devices rely on reconfigurable bumps or electrical stimulation, one possible avenue we explore for is creating tactile sensations through materials chemistry.

A challenge behind controlling touch is that touch is a complex mechanical event between the finger—a heterogenous and deformable interface—and an object. During touch, friction generated between the finger and object produces a variety of mechanical vibrations, which ultimately forms a tactile perception. In this talk, we describe our efforts to use materials coating to both control touch and to study touch. In the first part, we will discuss our efforts in using silane-derived monolayers and our findings that humans can perceive single atom substitutions within silane-coated silicon wafers, thus opening the possibility for molecular-scale control over touch. In the second part, we use tools from materials chemistry to enable psychophysical investigations on the perception of softness. By creating samples which react to touch in a known manner, we can perform our investigations with high “ecological validity”, i.e., perform studies in a manner similar to everyday human exploration. We will discuss our key findings that the perception of softness is universal between subjects, and thus is a repeatable and controllable tactile percept.

Since the modern concepts for virtual reality and augmented reality were first introduced in the second half of the twentieth century, the field has strived to develop technologies that could create realistic experience for the user to immerse themselves in a fully or partially virtual environment. While there has been great progress in visual and auditory technologies, haptics has seen much slower technological advances. The challenge is because skin has densely packed mechanoreceptors distributed over a very large area with complex topography; devising something as targeted as an audio speaker or television where sensory input is made into something as specific as an ear canal or iris seems to be a more difficult task. To compound the difficulty of a somewhat universal haptic solution, the soft and sensitive nature of the organ makes it difficult to apply solid state electronic solutions that cannot address large areas without causing discomfort. The emerging field of soft robotics, where many actuators and sensors are built from materials science innovations, offers potential solutions towards this challenge. In this article, we first provide an overview of haptic output and input technologies to identify opportunities for soft robotics to provide solutions for haptics. We then introduce mechanisms of intrinsically soft actuators and sensors and discuss soft haptic output and input devices categorized by device form of wearable, graspable, or touchable interfaces. We also review the definition and history of virtual and augmented reality and show examples of soft haptic devices demonstrated in VR/AR environments. Finally, we will introduce an innovative solution to the multiplexing vs. sensory density problem facing haptics.

The existing familiarity of textiles makes them ideal for lightweight, breathable, wearable soft robots and shape changing deployable garments. These garments are often manufactured by weft-knitting. The weft-knitting process allows us to alter the topology of a manufactured textile through an infinite number of stitch patterns, which allows for precise control of both the tactile and mechanical properties. We harness the existing symmetry in a knit stitch to manipulate the natural curvature of the material into a self-folding pattern. While traditional knit textiles are monostable and return to a single rest state when a load is removed, we leverage this folding to design a textile with multiple stable configurations. This multistable textile is evaluated by mechanical testing and predicted by a finite element model. We investigate how the distribution of stitches affects both the mechanical properties and the energy landscape of the knit. Understanding the relationship between the stitch pattern and the mechanical properties allows us to tune the behavior of these structures; we can select the switching load and design the shape of the structure in each stable state. By fine tuning the geometric parameters in the structure, we exploit the bistability to realize shape changing garments and deployable structures.

The ability to decode hand gestures is critical for natural human-machine interaction, immersive virtual reality, and creation of effective prosthetics. Although technologies such as video tracking have been applied for hand gesture recognition, they restrict the user’s mobility as they need to capture images of the body with several optical devices surrounding the user, and the cameras must maintain clear line-of-sight with all movements. Therefore, bioelectronics that are able to directly acquire and communicate the electrical activity of the human peripheral nervous system to machines have the potential to overcome these limitations. However, it is challenging to non-invasively acquire high-resolution electrophysiology signals that allow representation of ongoing muscle activity in the body. Effective contact and adhesion between conformable high-density electrodes and human skin are essential for the success of this kind of interface. Here we present a novel device that combines conformable electronics and organic mixed-conducting particulate composites (MCPs) to acquire reliable and high-spatiotemporal resolution muscle activity at the level of individual motor neuron action potentials. We demonstrated that MCP can establish anisotropic conduction with the tissue allowing the creation of high-density conformable electronics for the peripheral nervous system. We developed a 10x12 high density conformable array of electrodes combined with mixed conducting particulate composites with similar particle size as skin roughness to cover approximately 500 mm² of muscle area. We performed electrophysiological recordings in humans and non-human primates while subjects performed predefined tasks and movements. We are able to cluster the acquired action potentials into a large number of putative single motor units (MUs) using a template matching-based clustering algorithm. The action potential firing pattern was then correlated with the corresponding movement to establish a decoding model to allow construction of gestures. As such, our approach facilitates investigation into the neural mechanics of our sensorimotor system and enables creation of effective bioelectronic interfaces.

Soft pressure sensors with high sensitivity over a wide pressure range are required for various applications such as electronic skins for human-mimetic robotics and electronic tattoos for mechanophysiology measurement. In the last decade, most research aiming at increasing the sensitivity of capacitive pressure sensors focused on developing dielectric materials with added air gaps and/or higher dielectric constants. After extensive research, sensitivity has been significantly improved at low pressure range, e.g. 1 kPa, but still decays drastically as the pressure increases. To overcome this challenge, we engineered a soft capacitive pressure sensor based on an electrically conductive porous nanocomposite. The porous nanocomposite was fabricated by molding the mixture of functionalized carbon nanotubes (f-CNT) and Ecoflex out of a nickel (Ni) foam. The nanocomposite was 600-µm thick and 86% porous with open cells and tubular ligaments. With enough f-CNT doping (e.g. 0.25 wt%), the nanocomposite became electrically conductive with both piezoresistive and piezocapacitive characteristics. An ultrathin dielectric layer was added between the conductive nanocomposite and the electrode to ensure the whole device was still capacitive. This hybrid response pressure sensor (HRPS) was measured to have a sensitivity of 3.13 /kPa within 0-1 kPa, 1.65 /kPa within 1-5 kPa, 1.16 /kPa within 5-10 kPa, 0.68 /kPa within 10-30 kPa, and 0.43 /kPa within 30-50 kPa of pressure ranges. The hybrid response was fully understood through a simplified circuit model, which has been validated by experimental measurements and could be used to determine the optimal f-CNT loading. We have successfully applied this sensor to measure very subtle surface pulse waves of the radial artery, the common carotid artery, and the temporal artery, even under large preloads such as a VR (virtual reality) headset.

Carbon nanotube (CNT) is a promising material for multifunctional wearable electronic applications due to their excellent electrical conductivity, low density, and superior strength. However, the shorter length of CNTs and nanoscale dimension restricts them for macro assembly and bulk scale processing. Therefore, different synthesis methods emerged for the macro-assembly of the CNTs. Here we report a solid-state and straightforward process of producing CNT-wrapped textile yarns for multifunctional wearable electronic applications. Highly aligned and millimeter-tall CNTs were synthesized on a quartz substrate by chemical vapor deposition. Two CNTs arrays were placed on a custom-made spinning device, and the textile yarn was inserted as a core. Both the arrays were wrapped over the textile yarns for producing highly conductive and strong yarn.
The produced yarn has a diameter similar to the conventional textile yarns essential for the bulk production of electrodes. Scanning electron microscope (SEM) showed successful wrapping of CNTs over different textile yarns. The CNT-wrapped yarns are suitable for textile processing such as weaving and knitting, sewing, and embroidering into apparel and have excellent electrical conductivity. Both the strength and conductivity of the CNT-wrapped yarn can be customized for various applications. A 1% thermoplastic polyurethane nanocoating increased packing density by aligning the fibers along the yarn axis and increasing the strength two folds. We have demonstrated the use of the yarns in different multifunctional applications such as motion sensors, powering LEDs and heating garments. The yarn was seamlessly integrated into 3D knitted gloves and used for finger movement detection. The yarn exhibited a large resistance change when the finger was bent and returned to the initial position in relaxing conditions. When the yarn is connected to a power source, the yarn can transfer the power lighting 12V LED. A knitted wrist band was produced for wearable thermotherapy, and when connected to 3V, it can generate heat higher than skin temperature and reach about 46 degrees Celcius. The yarn can also be used as a flexible electrode for different physiological health monitoring such as electrocardiography and electromyography.

Liquid crystal elastomers (LCEs) are stimuli-responsive materials. Programming the local orientation within LCEs is a promising route to realize 3-D deformations. Recent demonstrations have shown these materials can exhibit exceptional work capacity of upwards of 40 J/kg. However, the inefficient and slow responses of LCEs to heat and light limit the functional integration of these materials in many applications. Electrical control of LCEs continues to be a preferred stimuli for use in actuation. Recently, electromechanical actuation of LCEs at ambient conditions was realized by exploiting the unique mechanical anisotropy (e.g., association of modulus with orientation) of aligned LCEs. By coating both sides of an aligned LCE with compliant electrodes and applying voltage, the LCE film is compressed in its thickness, leading to deformation in the planar area. However, due to the mechanical anisotropy, the LCE deforms preferentially in the lower modulus direction, leading to directional mechanical response in these actuating elements. Here, we are concerned with understanding the role of materials properties on the electromechanical response of LCEs. Using a newly developed materials chemistry and employing established processing methods, reliable electromechanical actuation is realized in LCEs with high mechanical anisotropy and uniaxial actuation strains up to 17.5%. The uniaxial actuation behavior is closely predicted using previously developed theory and the physical materials properties of the LCEs. Response times of 1 Hz are demonstrated. Furthermore, 3-D deformations from flat sheets to cones are demonstrated.

Human culture is replete with artifacts that interact with the senses of sight, hearing, taste, and smell. Material objects whose purpose is to produce a thoughtful or emotional response through the sense of touch, however, are rare. In this talk, I present my group’s recent work on the intersection between the science of soft materials and the science of touch. This field, which we have named “organic haptics,” combines active polymers, contact mechanics, and psychophysics. We are beginning to understand the ways in which stick slip friction, adhesion, and capillary forces between planar surfaces and human skin affect the ways materials produce tactile objects in consciousness as mediated by the sense of touch. This work, which combines human subject experiments, laboratory mockups of human skin, and analytical models accounting for friction, has led to several important observations. In particular, we have elucidated the mechanism by which humans can differentiate hydrophilic from hydrophobic surfaces when bulk parameters such as hardness, roughness, and thermal conductivity are held constant. We examined the role of relief structures in the skin—i.e., fingerprints—in determining the human ability to differentiate between surfaces. We have taken the insights from these psychophysical experiments to design new electroactive and ionically conductive materials to produce “actuator skins” whose goal is to produce realistic sensations for applications in tactile therapy, instrumented prostheses, education and training, and virtual and augmented reality.

Soft optical pressure-mapping elastomeric skins are an intrinsically stretchable and deformable platform for human-like tactile sensing. In this work, we investigate how the microstructure of an elastomer layer controls its light transmission as a function of pressure. Under pressure, voids in a soft, optically-transmissive material will preferentially compress, making the microstructured material densify. Not only do the voids control the tactile feel of the layer, but shrinking surface voids can increase the solid contact area that the elastomer makes with an underlying waveguide core, causing more light to exit the waveguide at the pressure location. Whether the voids are randomly distributed or deterministically placed by molding, the surface takes on new optical and mechanical properties thanks to its microstructure. Developments in this area will lead to all-polymer surfaces that measure and map deformation on a force scale controlled by the material’s design.

Optical waveguiding in a core-clad structure requires a cladding that has a lower index of refraction than the core. The refractive index (RI) describes the speed of light in the material, with higher RI materials having higher optical density, shorter wavelengths for a given frequency, and slower light velocity than in a low RI material. The ideal high-index core, low-index cladding structure enables rays with a sufficiently shallow incidence angle to achieve total internal reflection (TIR) at the core-cladding interface, the principle behind long-distance optical fiber communication. Normally the cladding is a solid material. Here, the cladding is an elastomer with RI between 1.4 and 1.6, containing voids: air pockets with RI close to 1. Voids are created using two methods: molding liquid-cure elastomers on micromachined substrates, and incorporating porogens into liquid-cure elastomers.

Two principles apply to these porous materials at different size scales: 1) for large features ( > 0.1 micron relative to the 940 nm sensor wavelength in this research), more of the light escapes the core where the cladding is solid compared to where the cladding has a void. Therefore the light loss increases as pressure increases the solid fraction of the total contact area. Scattering and interference also occur at the interfaces between air and solid. 2) For sub-wavelength features (< 0.1 micron), scattering and interference become less of an issue, and the material behaves as a composite with a refractive index somewhere between solid and air, proportional to the solid volume fraction.

To characterize the optical pressure response of these structured materials, we developed an instrument to measure transmission vs. incidence angle as a function of pressure for soft films having thicknesses between 0.5 and 3 millimeters. Applied pressures were in the 0-0.1 MPa (0-15 PSI) range that humans typically use to manipulate objects. Optical microscopy of molded surfaces, for example a hexagonal array of 180-micron diameter silicone hemispheres (RI=1.41), showed that the solid contact area with a glass slide (RI=1.55) did increase under pressure. We found that increasing pressure on the sample increased the amount of light escaping the glass at incidence angles below the TIR angle for the glass-to-solid elastomer interface.

Changing the designs of the micromachined features might allow one to further tune the pressure response, for example buckling at a threshold pressure or responding to shear forces. The work here was performed on liquid-cure elastomers, but porogens incorporated into 3D printable filaments such as thermoplastic urethane could enable additive manufacturing of microstructured soft optical surfaces without high-resolution printing.

The electro-stiction phenomenon provides the frictional force between two contacting electrodes covered with high dielectric media, of which an applied electric field produces the Maxwell force to attract the two electrodes. Such an electro-stiction force is mechanically and electrically simple and has a fast response speed. But it requires a relatively high driving voltage. In this study, we developed robotic joints having adjustable linear and rotational stiffnesses by stacking many electro-stiction layers, composed of dielectric film/electrode/frame/electrode/dielectric film. Polyoxymethylene copolymer (POM-C) is used for the frame of thickness 1 mm, and Au is sputtered onto the POM-C frame as the electrode of thickness 24 nm. A relaxor ferroelectric polymer, poly(vinylidene fluoride-trifluoroethylene-chlorotrifluoroethylene (P(VDF-TrFE-CTFE)), is used as the dielectric film material since the electro-stiction force is proportional to the relative permittivity of the dielectric material. The P(VDF-TrFE-CTFE) film is first dip-coated on the Au electrode and frame, followed by lamination of the 5 μm thick P(VDF-TrFE-CTFE) film, which is fabricated by the adhesion-mediated film transfer technique (AMFTT). The multilayered structure is an effective way to produce high electro-stiction force with low driving voltage, but all layers must be kept at a constant distance to attract neighboring layers with the same force. In this study, by extending the edge of the structure, drilling a hole, and aligning many frames through an elastic string, the multilayered frames can be simultaneously controlled with the same applied voltage. Therefore, two bundles of frames can freely slide each other without applied voltage, while they can experience different linear and rotational stiffnesses when the applied voltage is changed.

Although various stretchable optoelectronic devices have been reported, omni-directionally stretchable transparent circuit lines have been a great challenge. We engineer the cracks and fabricate highly conductive patterned metal circuit lines in which Au grids are embedded. Au is deposited selectively in the cracks to form a grid without any junction between the grid lines. Since each grid line is expandable under stretching, the circuit lines are stretchable in all the directions. This study shows that a thin coating of Al on the oxide surface enables precise control of the cracks (crack density, crack depth) in the oxide layer. High optical transparency and high stretchability can be achieved simultaneously by controlling the grid density in the circuit line. We integrate light-emitting diodes (LEDs) directly on the circuit lines and demonstrate stable operation under 100% stretching.

Multifunctional soft actuators can impart soft robots with diverse properties in a limited space, enabling the integration of several functionalities in a single device. Recently, soft electroluminescent actuators have been reported, where a dielectric elastomer is combined with electroluminescent particles under combined AC and DC driving signals. Here, we present an electroluminescent soft actuator that utilizes a liquid dielectric as the active material with EL phosphor particles dispersed in it. By coating the surface of the ZnS with silica, we solved the problem of agglomeration of the ZnS particles in the dielectric liquid, resulting in uniform luminescence throughout the device active region. The soft actuators showed strain values of 34% and maximum EL values of 25 cd/m² for a single actuator and over 34 cd/m² for five stacked actuators. The strain values are comparable to previous works. We expect our EL actuator devices will be used as new smart components for soft robotics and interactive sensors where external stimulus can be displayed as color signals.

Human touch is critical to the human experience, especially those with low vision and sensory impaired disorders. The field of haptics aims to design materials that recreate tactile sensation. Sensations in fine touch derive from the dynamic frictional forces that arise at the interface between human skin and the surface of a material, and these forces depend on both pressure and sliding velocity. Current approaches rely on physical controls such as mechanical vibrations, bumps, or pins, but the mechanical forces in fine touch arise from not only physical surface features, but the chemical properties of the material itself. Crystallinity is a classic material structure property used to control the mechanical forces of a material and has wide applicability in many material systems. Semi-crystalline polymers can have both amorphous and crystalline regions, and the degree of crystallinity can be controlled through many processes including annealing the polymer at an ideal crystallization temperature. Crystallinity has well-defined effects on mechanical forces, most notably an increase in modulus, but also has effects on frictional forces. Increasing crystallinity correlates with higher molecular order and higher shear strengths, resulting in lower coefficients of friction and less sensitivity to contact pressure. To connect tactile perception with degree of crystallinity, chemically similar atactic and isotactic polystyrene films were prepared and then annealed to induce crystallization. Films were compared for tactile distinguishability with an in-house mechanical testing set-up and human testing. We demonstrated that humans could distinguish films which differed by only degree of crystallinity as predicted by frictional force curves and characterized through microscopy and GIWAXS. Connecting tactile discriminability with degree of crystallinity has immediate use in developing switchable haptics devices with materials such as liquid crystal elastomers or conjugated polymers.

Electrochemiluminescence (ECL) has been recently exploited to form alternative light-emitting devices. referred to as phenomenon electrochemiluminescence device (ECLD). Unlike conventional light-emitting devices, an ECLD requires mass transfer of oxidants/reductants as well as charge transfer reactions in an electrolytic environment. In this talk, we propose the electronic skin (e-skin) device exploiting the characteristic mass transfer phenomenon within ECLD, as an alternative strategy capable of transducing mechanical stimuli into visual readout rely on ECL. First, we investigate the influence of the mass transfer and electron transfer kinetics on emission of ECLDs. Then, we fabricated an e-skin based on ECL layer employing thermoplastic polyurethane as a matrix polymer. The proposed material platform shows the visco-poroelastic response to mechanical stress, which induces a change in the distribution of the ionic luminophore in the film referred to as the piezo-ionic effect. This piezo-ionic effect is exploited to develop a simple device containing the composite layer sandwiched between two electrodes, which is termed ECL skin. Emission from the ECL skin is examined, which increases with the applied normal-/tensile- stress. Additionally, locally applied stress to the ECL skin is spatially resolved and visualized without the use of spatially distributed arrays of pressure sensors. The simple fabrication and unique operation of the demonstrated ECL skin are expected to provide new insights into the design of materials for human-machine interactive electronic skins.

References
[1] Electrogenerated Chemiluminescence, Ed: Bard, A. J. Marcel Dekker, New York, 2004.
[2] Kang, M. S. et al. ACS Appl. Mater. Interfaces 2018, 10, 48, 41562–41569
[3] Kim, D. H. et al. Nat. Commun. 2019, 10, 1, 1-13
[4] Kang, M. S. et al. ACS Photonics 2018, 5, 2, 267–277

Keywords: electrochemiluminescence; light-emitting device; light-emitting electrochemical cells; ionic liquid

Reversible programming of 3D soft mesostructures is desired for many applications including soft robotics and biomedical devices. The large, reversible shape changes of liquid crystal elastomers (LCEs), which result from the coupling between the alignment of liquid crystal (LC) molecules and the macroscopic deformation of polymer networks, have attracted much attention for such applications. In this talk, I will discuss our recent effort in developing a facile and versatile strategy to create reconfigurable, freestanding 3D mesostructures of LCEs and magnetic LCE composites that are inaccessible with existing techniques via spatially programming LC molecules. Experimental and theoretical results of more than 20 reconfigurable 3D LCE mesostructures of diverse configurations and their large, reversible shape-switching behaviors over multiple cycles will be demonstrated. In addition, an LCE gripper and a robot of ferromagnetic LCE composites that simultaneously responds to magnetic and thermal stimuli for diverse biomimetic behaviors, especially crawling underneath a narrow crack, will be used to illustrate the integration of other functional materials to LCEs for multifunctional systems.

The sense of touch is vital for any system in which humans use machines to physically interact with the world. An artificial sense of touch is especially useful when complementary sensory modalities, such as vision, are limited or unavailable. I will highlight past and present work to enhance the functionality of artificial hands in human-robot systems. I will describe our efforts to develop multimodal tactile sensor skins and our experiences with a variety of deformable, low-cost and commercially available tactile sensor designs. I will discuss the use of machine learning to perceive tactile directionality and salient geometric features, along with a tactile perception pipeline for haptically locating objects buried in granular media. Finally, I will present work in which reinforcement learning is used for the tactile-driven manipulation of deformable objects, such as sheets of paper, ziplock bags, and computer cables. Real-time tactile perception and decision-making capabilities could be used to advance semi-autonomous robot systems and reduce the cognitive burden on human teleoperators for applications that include assistive technology, manufacturing, the handling of dangerous materials, and remote healthcare and work in space.

The fields of soft robotics and human-machine interfaces are rapidly evolving but are still limited by the commercial viability of soft actuator technologies. When compared against other soft actuators (e.g. shape-memory alloys, pneumatics/hydraulics), dielectric elastomers provide a superior balance between performance metrics like actuation frequency, stroke length, and implementation hardware but fall short when comparing force output. Additional benefits of dielectric elastomers are that they are compliant, low profile, lightweight, and scalable, making them ideal in wearable, e-textile, and haptic interface platforms. The force output of dielectric elastomer artificial muscles depends on a few key parameters (e.g. electrode area, dielectric fluid volume, applied voltage, geometry of thin-films); it is well known that increasing the dielectric constant of the electrically insulating materials within the actuator has a nominally linear relationship to force output, but it is challenging to access high-dielectric polymers that can be used in artificial muscle fabrication. Diverging away from homo- and co-polymers, high dielectric constants can be alternatively accessed through nanomaterial composite films. This methodology allows for the use of high-dielectric inorganic materials (which are typically very rigid) in soft robotic applications. Inorganic nanomaterial composites using titanium dioxide, Fe₃O₄, and barium titanate were formed in a sandwich structure with polymer layers of biaxially-oriented polypropylene and poly(ethylacrylate acrylamide). Composites consisting of conductive, organic nanomaterials (e.g. graphene, carbon nanotubes) and their combinations with inorganic materials were also fabricated. The microstructure, composition, and electrical properties of these nanomaterial composite films were analyzed using scanning electron microscopy (SEM), Raman and FT-IR spectroscopy, and capacitance measurements. Changes in the dielectric constant were measured indirectly through material property analysis and mechanical studies of the artificial muscle actuator performance. These modifications showed artificial muscle maximum force output improvements greater than 2.5x compared to control muscles without nanomaterial incorporation. Actuator performance gains were in spite of any added thickness to the electrical insulators from the addition of nanomaterial. Efforts to decrease dielectric relaxation time and increase the actuator duty cycle by using a graphene-doped titanium dioxide nanomaterial composite found that conductive carbon can have a dramatic impact on performance. These nanomaterial composite advancements are particularly attractive because their drop-in quality allows for application in artificial muscle technologies without impacting existing actuator design. Such fundamental material insights push the field of soft robotics towards improving real-world human-machine interactions.

Haptic actuator research presents a challenge correlating fundamental knowledge between the properties of the materials and their behaviour on haptic systems. Specifically, the analysis of the magnetic properties of magnetic polymers produced by assisting the synthesis process with a magnetic field is still promising. In this work, magnetic polymer films (thickness: ∼0.020 mm) were made by the in situ synthesis of iron oxide nanoparticles (Fe₂O₃) in a polymeric matrix of polyvinyl butyral (PVB). This synthesis is performed by applying a constant magnetic field during the nucleation and growth of the Fe₂O₃ nanoparticles. Consequently, an increase in the reaction time was also necessary during this process. Therefore, a change in the magnetic properties (and the magnetic response) of the resulting magnetic polymer films was obtained. The magnetic characterization of the resulting PVB/Fe₂O₃ magnetic polymer films was performed using a vibrating sample magnetometer (VSM). The magnetization curves showed that in samples produced without the magnetic field, the maximum magnetic magnetization is similar (about 0.89 emu/g) and, in samples obtained with the magnetic field, the maximum magnetic magnetization is different (from 0.32 emu/g to 0.66 emu/g). Thereby, the orientation of the magnetic field could have affected the maximum magnetic magnetization. These results will improve the use of magnetic actuators in haptic systems like soft robots for manipulating actuators, aquatic actuators, or mimic animals.

Flexible sensors are being used in a variety of applications such as biomedical, textile robotics, and wearable devices. The ability of the sensors to function while deformed proves their adaptability for being implemented into various soft robotic structures, which makes them more desirable than conventional rigid sensors. Therefore, it is important to study the mechanisms through which the sensors reach their functional limits and begin to fail. In this study, we used PDMS as the matrix and silver-based ink for printing linear strain sensors. The sensors with varying vol.% of binder, dispersant, and solvent were embedded using two methods: lay-up and direct-writing. The failure mechanisms of these sensors were then studied based on the electro-mechanical performances, deformation tolerance, and durability while under varying levels of tension and compression using a universal testing machine. The desirable ink composition and fabrication method were evaluated based on the sensor’s conductivity, stretchability, bendability, and longevity.

Mechanosensors in biological somatosensory system encode mechanical stimuli into electrical pulse (i.e, action potential) for tactile perception and learning. Inspired by the nature of biological action potentials, we present a soft neuromorphic mechanosensor that enable robust tactile application even in aquatic environments. The soft ferromagnetic material in the mechanosensors generates electrical pulses under external stimuli. The decoded responses from the specific pulse sequences exhibit a highly linear function with the mechanical stimuli. Assisted with machine learning algorithms, we achieved high accuracy multi-pixel differentiation and underwater object recognition in robotic and flexible human-machine interface applications. We anticipate that this new soft neuromorphic mechanosensor will potentially enable new forms of robots and human-machine interfaces.

Mobile phones, tablets, and watches have become a normal part of everyday life. The next generation of devices will enable touch feedback, become flexible, and extend increasingly rich and immersive interactions into virtual reality. Remote touch interactions will convey an object’s compliance, or softness, as with fruits such as plums and tissues such as skin. In addition to applications in entertainment and personal productivity, such interactions will enable surgeons to distinguish gallbladder and prostate tissue and ducts from fat and bone, small children to feel a parent’s hand, and consumers to inspect and compare products, clothing, and work pieces. While promising new possibilities, the displays under development do not yet feel natural. They also face severe limitations in terms of weight, power, and actuation range. To inform devices to replicate naturalistic interactions of this sort, we have been working to define how the finger pad must deform in space and time to adequately convey a natural sense of compliance.

Object softness, of which compliance is one major dimension, is thought to be encoded by relationships of force, displacement, and contact area at the finger pad. This talk will describe work to understand how time-dependent cues, or information in the rate of change of skin over a spatial field, govern the encoding of compliance. To do so, we have first developed experimental devices to control the force, indentation, and or their rates in passive touch. We concurrently developed an ink-based method to measure biomechanical contact of the finger pad. Human-subjects studies show that temporal, force-rate cues are more efficient in discriminating compliances, by reducing the amount of deformation of the finger pad and stimulus, which aligns with ongoing skin remodeling. To further study force-rate cues, but under an active touch paradigm, we developed a compliance illusion by introducing spheres of various elasticities and curvatures, derived from finite element simulations of the finger pad. The illusion is that participants cannot discriminate, in passive touch, spheres that are small and compliant versus large and stiff, which generate equivalent contact areas. In active touch, however, participants easily discriminate these spheres, and may utilize cues tied to proprioceptive joint angle, the rate of change of contact area, or surface penetration depth. This special case tells us that our perception of softness is a product of both sensation and volition, and depends upon both afferents in skin and proprioception. To further evaluate these cues, we developed a stereo imaging technique to capture skin deformation while interacting with objects of varied elasticity. We have also standardized a way to characterize the elasticity of test stimuli used by others in similar studies and begun to study ecological interactions with naturalistic fruits in order to reproduce observed skin deformation with dynamic displays.

Open Discussion

Roughness at the micrometer scale has been identified as a key tactile dimension in the haptic perception of materials. There is a longstanding tradition to create well-defined surface structures with the goal to elucidate mechanisms underlying roughness perception. We will discuss results of two tactile perception studies with micro-structured stimuli.
Randomly rough surfaces were created by 3D printing with independently varied topography and spectral distribution of roughness. We found that the tactile perception of similarity follows the amplitude of roughness at smaller length scale rather than the topographic resemblance. The interaction between surface asperities and fingertip skin led to higher friction for higher microscale roughness. Individual friction data allowed us to construct a psychometric curve which relates similarity decisions to differences in friction [1].
We also produced fibrillar samples from materials with varying elastic modulus. The perception of similarity between samples with different fibril length and different elastic modulus followed the bending of the sub-millimeter scale fibrils, which were arranged in regular arrays on the surface.
[1] R. Sahli, A. Prot, A. Wang, M.H. Müser, M. Piovarči, P. Didyk, R. Bennewitz, Tactile perception of randomly rough surfaces, Scientific Reports, 10 (2020) 15800.

Haptic devices can elicit sensations remotely or in virtual environments. Sensing and actuation are required in order to digitize real world interactions and transmit it to an actuation device or system. In this talk, I will discuss strategies to enable soft electromagnetic materials to sense the environment and simultaneously actuate. Such materials can potentially be made self-healable, and allow for mechanical recovery in performance despite mechanical damage. Repair can be effected simply by mechanical contact of the damage areas. It is envisioned that such materials can be useful in the growing field of haptic devices and systems.

Soft sensory devices that mimic the human skin with multiple sensors spatially distributed to extract information in a temporal manner are technically challenging yet promising for many emerging applications from haptic interfaces, to medical robotics and to skin prosthesis among others. Although drastic advances already been made in developing various soft devices with various functions, one of the long-lasting challenges is to have biology-similar materials and structures to physically mimic the human skin. This presentation will introduce rubbery electronics, which is a new type of electronics with skin/tissue-like softness and mechanical stretchability, constructed all based on elastic rubbery electronic materials. The innovations in rubbery electronic materials and devices set the foundation for rubbery electronics, integrated system and their broad usages, such as active matrix based tactile sensory skins, soft neurorobotics, etc. This presentation will introduce fully rubbery based active matrix tactile sensory skin that is able to sense the touch spatiotemporally that could be equipped for various robotics and robotic interfaces. This presentation will also demonstrate a soft neurorobots which is constructed with neurologic function integrated skin on soft robotics to decode the external interactions into neurological signals for decision making.

The field related to triboelectric nanogenerator (TENG) devices is emerging rapidly by continuously presenting new original and creative TENG concepts for convertion of mechanical energy into electricity.¹ The working principle of TENG is straightforward - upon contact-separation or sliding of two triboelectric material (often polymer) coated electrodes surface charges are formed on the triboelectric materials, which induce an electrostatic charge on the conductive electrodes. As electrodes move, a potential difference is created causing a current flow in the external electric circuit. Evidently, TENG devices can be used as excelent self-powered force sensors and haptic sensor arrays.^2,3 Up to now devices have been integrated into a range of objects - fabrics,⁴ wearables,⁵ interior objects,⁶ and even implantable devices.⁷
Additionally embedding ferroelectric dipoles in contacting polymer layers is known to effectively enhance the performance of TENG. However, dipoles can form also between the opposite triboelectric surface charges on the contacting ferroelectric films. Up to now this similarity has been neglected in the relevant studies. Our research demonstrates that proper attention to the alignment of the distinct dipoles present between two contacting surfaces and in composite polymer/BaTiO₃ ferroelectric films can lead to up to four times higher energy and power density output compared with cases when dipole arrangement is mismatched. For example, TENG device based on PVAc/BaTiO₃ shows energy density increase from 32.4 μJ m⁻² to 132.9 μJ m⁻² when comparing devices with matched and mismatched triboelectric and ferroelectric dipoles. The presented strategy and understanding of resulting stronger electrostatic induction in the contacting layers enable the development of TENG devices with greatly enhanced properties. Greater generated energy allows easier integration with MEMS and since the output signal is stronger - more precise electromechanical response and sensing.

References:
1. Sci. Technol. Adv. Mater. 2019; 20: 758-773
2. Nano Energy 2021; 79: 105431
3. Nano Energy 2020; 78: 105266
4. ACS Appl. Mater. Interfaces. 2014; 6: 14695-14701
5. Adv. Mater. 2015; 27: 2367-2376
6. Sci. Rep. 2016; 6: 22253
7. Nat. Commun. 2018; 9: 5349

The concept of ‘kirigami’ has been of interest in the field of stretchable electronics because the kirigami patterned structure can increase the degree of freedom for deformation beyond the intrinsic limits of the original material. Furthermore, shape-reconfiguration of tunable electronics can be achieved by the introduction of kirigami-engineered design for programmable materials such as azo-benzene functionalized liquid crystalline polymer networks (azo-LCN). In this presentation, we introduce a facile coating process for patterned reduced graphene oxide (rGO) as well as integration of the patterned rGO along with shape-reconfigurable azo-LCN. The rGO patterned azo-LCN (azo-LCN/rGO) demonstrates about 500 % higher storage modulus of 6.4 GPa than neat azo-LCN (1.3 GPa) and an electrical conductivity of 380 S cm^-1 at 4 times the number of rGO coating cycles. In addition, multi-stimuli (i.e. UV, NIR, focused solar ray, and direct heat) responsive actuation of azo-LCN/rGO is achieved by photochemical isomerization of azobenzene moiety in conjunction with the coefficient of thermal expansion (CTE) mismatch via photothermal effects. In particular, the convolution of photochemical and photothermal effects provides enhanced actuation performance of the azo-LCN/rGO bilayer from the neat azo-LCN monoliths. Finally, the kirigami pattern allows the azo-LCN/rGO to endure a higher strain beyond the strain capacity of azo-LCN/rGO without permanent damages to its own body or deterioration of electrical performance. The kirigami-engineered azo-LCN/rGO demonstrates passive, active and hybrid types of shape-morphing under mechanical tension, UV, and simultaneous stimuli of UV and tension, respectively. Toward shape-reconfigurable and stretchable electronics, we demonstrate UV-induced shape-morphing of the kirigami-engineered azo-LCN followed by 60 % of mechanical strain by tension without increase in electrical resistance.

Current virtual reality (VR) techniques can provide synthesized graphics and sounds that allow the user to explore the virtual world but permit comparatively limited haptic feedback. One challenge frequently encountered by wearable haptics designers is the dilemma between the need for generating sufficiently large haptic signals and the desire for keeping the device small and lightweight for better wearability. Indeed, a bulky and heavy haptic device may constrain the movement of the user and cause fatigue. Actuators made from smart materials can have high power-to-weight ratio and therefore can provide sufficient haptic feedback with limited size and weight.

Here, we present our recent design of a wearable haptic device made from electrohydraulic actuators [1]. The device consists of a flat dielectric pouch filled with liquid dielectric. Six pairs of hydrogel electrodes are attached to opposite sides of the pouch and enclosed by insulating rubber. Made from soft and compliant materials, the device can be attached to the skin as a body-worn haptic interface. Voltage applied to any pair of hydrogel electrodes generates electrostatic force, which closes the gap and displaces the liquid dielectric between them. Activation of multiple pairs of electrodes can produce both static and dynamic spatial patterns to provide feedback to a large area of the skin. The device is thin (< 3.5 mm), lightweight (< 35 grams), and capable of producing substantial displacement (> 2 mm) and forces (> 0.8 N) for haptic feedback. The wearability of the device allows long-term usage in VR systems. Since the feedback provides a softness feeling, potential applications of the device include wearable interfaces for social haptics and medical training simulators in VR. Currently, in our unpublished work, we further improved the design to enable more complex haptic feedback patterns, larger displacement and force output, and full portability of the whole system.

[1] Yitian Shao, Siyuan Ma, Sang Ho Yoon, Yon Visell, and James Holbery, "Surfaceflow: Large area haptic display via compliant liquid dielectric actuators". In 2020 IEEE Haptics Symposium (HAPTICS), pp. 815-820.

Haptics refers to our capacities for touch perception and physical interaction with our surroundings. Haptic technologies hold the potential to transform how we interact within environments that are increasingly interwoven with digital information. However, technologies for furnishing tactile information to the skin, our largest sensory organ, are rudimentary when compared with biological capabilities. I will describe the challenges involved in characterizing and engineering for the human sense of touch. I will discuss recent research in my lab that aims to meet these challenges through research on haptics and emerging material technologies, including several projects that elucidate key synergies between materials, mechanics, and capabilities of biological and engineered systems, and how this research is yielding new technologies for haptics and robotics.

Haptic sensors are devices for measuring any physical quantities that can be received from a tactile sense. They measure from temperature, pressure to their advanced concepts such as roughness, vibration, pain, etc. Most haptic sensors are designed to be placed close to the skin for the purpose of haptic. This can be attributed to the tremendous development of 2D flexible devices. Despite such advanced haptic sensors, haptic sensors always suffer from motion noise. Even though the sensor reads the necessary data, it is mixed with noise and cannot transmit meaningful information. Accordingly, many researchers tried to solve the problem with various ideas such as strain/bending insensitive pressure sensors, uni-directional sensors, decoupled strain-temperature sensors, etc., and published quite meaningful research results. However, another difficulty arises when the different sensors studied in their own methods need to be combined once more to be used in the complex sensing system of the haptic.
Unlike previous studies, we did not focus on the sensor itself, but rather focused on the surrounding environment where the haptic sensors are located. Haptic sensors are mostly in the form of a flexible film in order to be close to the skin, and are affected by the relaxation, contraction, and bending of the skin. Here, we have focused on stress-transformation in this skin. In an environment where only one-way normal force of contraction and relaxation is applied, such as skin, stress transformation shows that there is a region where stress in a specific direction does not exist. At this time, it is necessary to focus on the fact that most flexible sensors are in the form of a 2D film and have little reactivity to normal stress acting perpendicular to the 2D film unless they have a bi-layer structure. From the perspective of research up to now, it was taken for granted that the flexible sensor was located on the same surface as the skin surface due to the morphological factor of the 2D film. This is the most vulnerable positioning method to the force transmitted by contraction and relaxation. However, even though it is a 2D film, it is a different story if it maintains a different angle from the skin surface. The force transmitted from the skin can be made to act in a direction in which the sensor does not respond. Taking advantage of this, we found a specific area that stress from the dynamic motion does not affect the device. In this area, the strain sensor is insensitive to contraction, tension, and bending, but reacts sensitively to the external pressure that must be read as a real tactile sense. Furthermore, since this study is not limited to sensors, it can be applied to universal 2d flexible sensors. Therefore, we performed a measurement experiment with a 2d temperature sensor in the corresponding area. It was possible to measure the desired temperature without being affected by motion noise.
Our research was more about the environment surrounds the strain sensor rather than the sensor itself. It means it is applicable to many other already-made sensors without an additional device.

Structural and chemical heterogeneity around nanoparticles have strong influences on mechanical properties of nanocomposites. The phase stability of polymer grafted nanoparticles in polymer melt is governed by the mixing of chemically different polymers in the particle-polymer interfacial regions. In this work, poly(methyl methacrylate) (PMMA)-grafted Fe₃O₄ nanoparticles with the same grafting density but with two different chain lengths are synthesized and dispersed in poly(methyl acrylate) (PMA) at different molecular weights. The mechanical responses of the composites are measured in rheometer. The grafted particle composites exhibit the higher rubber plateau modulus than adsorbed samples. This is attributed to the higher number of segments interacting with the matrix chains, and this conformational factor on interfacial layers on nanoparticles tune thermal-stiffening and reinforcement of composites. It was found that intermediate molecular weight of grafted chains reinforces the composite better at high temperatures. For both long and short grafted particles, rubber plateau modulus and entanglement density decreased due to phase separation of grafted particles. Moreover, the interface entanglements recover and re-entangle more with the large deformation experiments with the short graft and matrix chains.

Open Discussion

Our coordination and convergence challenges in industrial soft robotics are defined by task uniqueness (no one has ever tried to do what we do) and expansive multidisciplinarity (...→chemistry→materials→mechanics→design→...). Such an environment makes it hard to scope and align research for the individual scientists, engineers, and designers, and makes prioritizing the work ambiguous for the academic and industrial organization.

The history of scientific discovery highlights certain research teams (Bell Labs, PARC, Rad Lab) that implemented strategic organizational and tactical behavioral innovations that lead to a culture of convergence. All research may be considered on a two dimensional space defined by the axes of Fundamental↔Applied and Curiosity-Driven↔Use-Inspired. Industrial soft roboticists are working in the quadrant of Fundamental, Use-Inspired research — called Pasteur’s Quadrant — that is neither pure basic nor pure applied research.

Much like academia, industrial research into soft robotics is based on the fundamental questions that push each subsystem forward. The anatomy of soft robots spans actuators, sensors, control, power, and the mechanical structural (skin, bone, etc). The divergence between academia and industry is in terms of value assigned to system integration and to reliability/replicability. System integration is often considered technical and without intellectual merit, yet the gaps in human knowledge around holistic system design is vast -- especially in soft robotics. Risk assessment, balancing, and mitigation drive much of the industrial research portfolio management -- and a major attribute of this process is assessing the reliability and replicability of novel technology. The current academic incentive for novelty and first-to-press discounts the value of replicability/reliability that drives industrial research. Strengthening the alignment and bridging the divergence between academy-industry must be our goal in soft robotics research if it is to broadly translate its impact to society.

The use of 3D printing for rapid prototyping of functional and biocompatible polymeric materials enables new functionality of polymeric materials for devices useful in flexible and wearable electronics. The use of thermoplastics, thermosets, and elastomers
based on their outstanding thermo-mechanical properties have to be married with their 3D printability. In addition, for actual biomedical device applications, their in-vivo or in-vitro functions have to be considered. A number of 3D printing methodologies (FDM, SLA, SLS, VSP)
can make use of new blended or formulated compositions that incorporated stimuli-responsiveness. Herein, we demonstrate the ability to deploy 3D and 4D printing methods of designed (materials and geometry) polymer materials. This will be demonstrated in the following: 1) biomedical grade thermoplastic polyurethanes (TPU) with compatibility to mammalian NIH 3T3 cells and elastomeric response. 2) Shape-memory polybenzoxazine-epoxy polymers that enable T and P response, and 3) Actuator devices to measure mechanical response in motion detection. The 4D printing allows the design of new materials and applications based on integrating the chemistry of conversion with the printing mode. For example, these materials have been printed via viscous extrusion printing (VEP) or VSP and then converted to an elastomeric actuating material with very high cyclic compressibility. Other work based on the use of SLA, SLS, FDM, towards high
strength nanocomposite and biomaterials will also be discussed.

Textile- and clothing-based haptic systems remain a persistent challenge for research and industry. Integrating mechanical actuation into structures with variable and unpredictable geometry and mechanics is a new level of complexity for system design. The field of flexible electronics has afforded a degree of conformability to electronic devices, but does not yet meet the 3D drapability or comfort requirements expected of clothing-like wearability, particularly for systems that require simultaneous access to many parts of the body. Developing effective wearable haptic systems requires accommodating many facets of user and system needs, which are often in conflict. This talk will explore challenges and current approaches related to: haptic modalities in clothing; fabrication and manufacture of soft actuation devices; understanding and affording perceptibility of actuators in clothing systems; sizing and fit across a user population; and balancing the comfort/functionality tradeoffs in system design.

Printed electronics are expected to be implemented as a set of industrial technologies that will facilitate the on-demand fabrication of imperceptible^[1] and shapeable^[2] devices. Conductive pastes are typically composed of polymeric matrices with embedded conductive fillers. The properties of the fillers can be exploited to deliver functional devices as printed transistors^[3], displays^[4], and sensors^[5]. The smart integration of such elements will allow task-specific integration in consumer electronics and even personalized wearable devices.

Aiming to develop on-skin printed interfaces, it is necessary to ponder mechanical, performance, and health safety considerations. Integrating magnetic sensors on interactive platforms is attractive due to their touchless, action-at-distance nature, which increases the reliability of the devices^[6]. In the past, solution processable magnetic field sensors have been fabricated from composite pastes embedding magnetic particles. Among the previous reports on printable magnetic sensors^[7], there are not examples of devices able to maintain high performance sensing during usual skin deformations. Concomitantly, there is a lack of research on skin-compliant printed magnetic sensors able to perform below the 40 mT safety continuous exposure threshold established by the World Health Organization^[8].

Here, we will present the fabrication and implementation of stretchable printed magnetic field sensors. They are based on composite pastes with embedded flakes of [Py/Cu]₃₀ Giant Magnetoresistance (GMR) thin-film stacks. We demonstrated printed GMR sensors on ultrathin (3-µm-thick Mylar) foils which are skin compliant, and with maximum sensitivity at 0.88 mT. The stretchable sensors maintained stable sensing performance at 16 µm bending radius and 100 % strain which corresponds to two orders of magnitude increase with respect to previous reports. We demonstrate the implementation of the technology on interactive applications after laminating the printed sensors on the user's skin to navigate through digital maps and scroll through text documents. The ability of the sensor to comply with the skin creases and deformations, and to detect field changes in the safe threshold limit, place this technology as a prospective method for fabricating on-demand printed human-machine interfaces^[9].

[1] M. Melzer et al., Nat. Commun. 6, 6080 (2015)
[2] D. Makarov et al., Appl. Phys. Rev. 3, 011101 (2016)
[3] J.A. Lim et al., Adv. Func. Mater. 20, 3292 (2010)
[4] S. Cho et al. ACS Appl. Mater. Interfaces 9, 44096 (2017)
[5] X. Wang et al. ACS Appl. Mater. Interfaces 10, 7371 (2018)
[6] P. Makushsko et al., Adv. Func. Mater, 2101089 (2021)
[7] E.S. Oliveros Mata, et al. Appl. Pys. A 127, 280 (2021)
[8] Static Fields. World Health Organization. (2006)
[9] M. Ha, E.S. Oliveros Mata, et al. Adv. Mater. 33, 2005521 (2021)

Haptic devices allow touch-based information transfer between humans and intelligent systems, enabling communication in a salient but private manner that frees other sensory channels. For such devices to become ubiquitous, they must be intuitive and unobtrusive – properties that can be achieved using soft materials that interface comfortably with the body. The amount of information that can be transmitted through touch is limited in large part by the location, distribution, and sensitivity of human mechanoreceptors. Not surprisingly, many haptic devices are designed to be held or worn at the highly sensitive fingertips, yet stimulation using a device attached to the fingertips precludes natural use of the hands. In this talk, we will explore the design of haptic feedback mechanisms that can be worn on the body, which achieve compelling haptic stimulation using a combination of soft robotics approaches and an understanding of human haptic illusions.

In creating wearable soft haptic devices, one can consider an approach that transmits information through many contacts, taking advantage of “unused real estate” on the surface of the skin to deliver haptic cues over a large area. One example uses a sequence of discrete contacts delivered to the forearm by pneumatic pouches attached to a growing haptic device constructed from flexible low-density polyethylene, called HapWRAP. Controlled air flow through tubes and pouches allows HapWRAP to grow out of a compact housing unit and provide a combination of directional and force feedback to a user. When activated, HapWRAP grows up and around the forearm; its loops form a temporary sleeve. After growth, pneumatic actuators inflate and deflate to stimulate mechanoreceptors in the skin at distinguishable locations, or in sequence to allow the haptic illusion of a traveling contact on the skin.

Alternatively, a wearable haptic device can use a single contact, but apply multi-modal feedback and/or multiple degrees of freedom of simulation to transmit information to the user. We created a multi-modal haptic device with a rigid rotational housing and three soft fiber-constrained linear pneumatic actuators. Soft pneumatic actuators were used because of their compliance, light weight, and simplicity, while rigid components provided robust and precise control. The soft pneumatic actuators provide linear horizontal and vertical movements, and the rigid housing, affixed to a motor, provides rotational movement of the tactor. The device can produce normal, shear, vibration, and torsion skin deformation cues by combining the movement of the soft pneumatic actuators with the rotational housing. The device generates haptic illusions in that the skin deformation, which is inherently tactile in nature, can generate the sensation of a world-grounded force applied to the arm.

This talk will present an information-theoretical approach to maximizing information transmission through haptic displays. While the talk will not focus on soft materials per se, it will provide the engineering specifications for actuation materials and mechanisms based on psychophysical data. Our past research over several decades have provided empirical evidence that the human skin is capable of receiving complex information at a rate of at least 12 bits/s, a rate that is significantly above the demonstrated performance of most haptic displays. A major challenge is the lack of suitable actuation mechanisms for wearable haptic displays.

I will cover the following topics in this talk:
1) The need for an information-theoretical approach to haptic interface research;
2) The derivation of 12 bits/s from natural tactile speech communication methods;
3) Guidelines for maximizing information transmission based on empirical evidence;
4) Recent success in transmitting English words via a tactile sleeve with an embedded array of broadband actuators.

I will conclude the talk with a set of engineering specifications for soft materials and mechanisms capable of delivering broadband actuation on the skin surface to achieve high information transmission rates.