Abstract The U.S. Naval Research Laboratory (NRL) has been developing a space-rated 7 degree of freedom (DOF) robot arm with a high payload-to-mass ratio as an alternative design to motor-gear driven robotic manipulators. The robot arm employs antagonistic pairs of pneumatic artificial muscle (PAM) actuators to control each degree-of-freedom (DOF) to achieve large force outputs relative to the PAM component masses. A novel feature of the NRL PAM actuator was the integration of the pneumatic control components inside the pressure-bladder, which not only reduces the volume of the robotic arm hardware but also reduces the pressurized-gas actuation volume in the PAM enabling significant reductions in gas consumption during actuation. This multifunctional design enables reductions in launch-weight costs and increases in operational endurance for space applications. The integration of these PAMs into a well-designed robotic-arm structure, in tandem with a newly developed control algorithm, has the potential to exceed the performance metrics of traditional motor-driven robot arms. This paper describes the development of the improved efficiency PAM design that is advancing this technology towards space flight readiness.

Abstract Inflatable structures provide significant volume and weight savings for future space and soft robotic applications. Structural health monitoring (SHM) of these structures is essential to ensuring safe operation, providing early warnings of damage, and measuring structural changes over time. In this paper, we propose the design of a single flexible strain sensor for distributed monitoring of an inflatable tube, in particular, the detection and localization of a kink should that occur. Several commercially available conductive materials, including 3D-printing filaments, conductive paint, and conductive fabrics are explored for their strain-sensing performance, where the resistance change under uniaxial tension is measured, and the corresponding gauge factor (GF) is characterized. Flexible strain sensors are then fabricated and integrated with an inflatable structure fabric using screen-printing or 3D-printing techniques, depending on the nature of the raw conductive material. Among the tested materials, the conductive paint shows the highest stability, with GF of 15 and working strain range of 2.28%. Finally, the geometry of the sensor is designed to enable distributed monitoring of an inflatable tube. In particular, for a given deformation magnitude, the sensor output shows a monotonic relationship with the location where the deformation is applied, thus enabling the monitoring of the entire tube with a single sensor.

Abstract As a smart material thermal shape memory alloys (SMAs) feature actuator behavior combined with self-sensing capabilities. With their high energy density and design flexibility they are predestined to be used in soft robotics and the emerging field of morphing surfaces. Such shape changing surfaces can be used for novel human-machine interaction (HMI) elements based on mode-/situation-dependent interfaces that may be applied to all kind of machines, appliances and smart home devices as well as automotive interiors. Since many of those contain textile surfaces, it is of special interest to place SMA-based actuator-sensor-elements beneath a textile cover or integrated them in the textile itself. In this study, the unique features of SMAs are used to design a system which represents an active “morphing” button. It can lower into the surface it is integrated in, pops up to be used and shows a proportional signal output depending on the pushing stroke. The system is characterized concerning haptics and sensor technology. The button consists of a TPU structure, to which two NiTi wires are attached. When activated, the SMAs contract and the structure curves upwards. The user can now push on the device to use it as a button. In the future, the use of SMA wires and for example TPU fibers enables direct integration in the production process of a possible smart and functional textile.

Abstract Knitted textiles manufactured from shape memory alloy (SMA) monofilaments possess advanced capabilities for distributed and complex actuation and are suited for a range of emerging needs in aerospace, biomedical, and robotics applications. In general, high currents for short periods of time provide sufficient electroresistive (Joule) heat to cause SMA wires to transform to austenite. However, SMA knitted textiles are difficult to electroresistively heat because the interlocking knit structure short-circuits the flow of current, causing localized overheating and isolating the transformation of the material along the current path. One approach for heating SMA knitted textiles is to drive pulses of high current between pairs of electrodes positioned across horizontal courses (rows) of knitted loops. This research presents a preliminary experimental investigation of the effects of factors related to electroresistive heating for SMA knitted textiles. A design of experiments analysis with two levels of four factors was conducted using a 2 4–1 fractional factorial design. The factors included the voltage of the power supply connected to the current amplifiers; a geometric factor defining the horizontal spacing of the electrodes attached to the knit sample; and two waveform factors: On Cycles and Off/On Cycles, which defined the length of time each current amplifier was enabled and disabled. Actuation performance was quantified by the actuation displacement and actuation force of the knit sample. Preliminary results suggest that voltage is the most influential factor, but also indicate that interactions between the geometric and waveform factors have significant effects on the heating and actuation performance. The characterization of these factor interactions has the potential to inform optimal electroresistive heating approaches for SMA knitted textiles, enabling integration into applications such as wearable technologies where convective heating is not practical.

Abstract Continuum robots are inspired by biological trunks, snakes and tentacles. Unlike conventional robot manipulators, there are no rigid structures or joints. Advantageous is the ease of miniaturization combined with high dexterity, since limiting components such as bearings or gears can be omitted. Most currently used actuation elements in continuum robots require a large drive unit with electric motors or similar mechanisms. Contrarily, shape memory alloys (SMAs) can be integrated into the actual robot. The actuation is realized by applying current to the wires, which eliminates the need of an additional outside drive unit. In the presented study, SMA actuator wires are used in variously scaled continuum robots. Diameters vary from 1 to 60 mm and the lengths of the SMA driven tentacles range from 75 to 220 mm. The SMAs are arranged on an annulus in a defined distance to the neutral fiber, whereby the used cores vary from superelastic NiTi rods to complex structures and also function as restoring unit. After outlining the theoretical basics for the design of an SMA actuated continuum robot, the design process is demonstrated exemplarily using a guidewire for cardiac catheterizations. Results regarding dynamics and bending angle are shown for the presented guidewire.

Abstract Electronic skins, or e-skins, are electronic devices capable of sensing physical interactions such as strain, temperature, or pressure. These e-skins are of interest in a variety of fields including robotics, structural health monitoring, and medicine. E-skins should measure strains over a larger range of elongation than traditional strain sensors could. This paper explores the synthesis of a flexible biaxial strain sensor for large surface strain measurement. The sensor is made by spraying an exfoliated graphite and latex mixture onto a latex substrate to form a 4 × 4 grid of electrically conductive strips. Electrodes are connected to each sensor to collect data on deformation induced voltage difference. Two setup geometries were characterized, the behavior of a single strip in each direction in a one by one configuration as well as the behavior of a four by four setup that can measure a two-dimensional strain field. The characteristics of the sensor is studied by attaching it on a tensile testing specimen. When the sensor is subjected to strain along one or both of the two measurement axes, the voltage difference can be recorded using Arduino. The voltage drop was normalized and used to construct a strain distribution plot in MATLAB to determine the highly strained location. In addition to characterizing the behavior of the sensor, the dispersion of the exfoliated graphite in the latex is also studied using optical microscopy. The sensor is made from inexpensive materials and was able to measure large strain that cannot be achieved with commercially available strain gauges.

Abstract The U.S. Naval Research Laboratory (NRL) has been developing a space-rated 7 degree of freedom (DOF) robot arm with a high payload-to-mass ratio as an alternative design to motor-gear driven robotic manipulators. The robot arm employs antagonistic pairs of pneumatic artificial muscle (PAM) actuators to control each degree-of-freedom (DOF) to achieve large force outputs relative to the PAM component masses. A novel feature of the NRL PAM actuator was the integration of the pneumatic control components inside the pressure-bladder, which not only reduces the volume of the robotic arm hardware but also reduces the pressurized-gas actuation volume in the PAM enabling significant reductions in gas consumption during actuation. This multifunctional design enables reductions in launch-weight costs and increases in operational endurance for space applications. The integration of these PAMs into a well-designed robotic-arm structure, in tandem with a newly developed control algorithm, has the potential to exceed the performance metrics of traditional motor-driven robot arms. This paper describes the development of the improved efficiency PAM design that is advancing this technology towards space flight readiness.

Abstract A key aspect of color change is altering perceived value or intensity. This paper presents a methodology to achieve value change through mechanical means via the deflection of bistable structures. We create mechanical pixel-based, reversible color change using 3D printed switchable bistability. Switchable bistability arises from the combination of pre-strain and shape memory, enabling us to access multiple elastically programmed shapes at elevated temperatures with fast morphing and low actuation forces, while retaining high stiffness at room temperature. Building on our previous study that achieved bistability through FDM printing with directional pre-stress, finite element analysis is conducted to design a pixel-like structure that acts as a unit cell with color change capabilities. Quantitative and qualitative analysis is conducted through image processing techniques in order to prove the viability of this approach to creating value change through geometric deformation of bistable structures. By leveraging this technique, there are numerous potential applications in fields including robotics, architecture, and interior design.

This article presents the experimental validation of a Direct Adaptive Control for angular position regulation of a lightweight robotic arm. The robotic arm is single degree-of-freedom (DOF) system, actuated by two Shape Memory Alloy (SMA) wires. The proposed adaptive control is capable of adapting itself to the hysteretic behavior of SMA wires and update its behavior to deal with the changing parameters of the material over time. The closed-loop approach is tested experimentally showing its effectiveness to deal with the highly nonlinear dynamics of the SMA wires. These results are discussed and compared with a classical control approach. The updated design and hardware development and modeling of the robotic arm are shown.

Shape Memory Alloys (SMAs) are often used for robotic, biomedical, and aerospace applications because of their unique ability to undergo large amounts of stress and strain during thermomechanical loading compared to traditional metals. While SMAs such as NiTi have been used in wire, plate, and tubular forms, NiTi as a woven dry fabric has yet to be analyzed for use as protective materials and actuators. Applications of SMA fabric as a “passive” material include shields, seatbelts, watchbands and window screens. Applications as an “active” material include robotic actuators, wearable medical and therapy devices, and self-healing shields and screens. This paper applies a macro-mechanical model from composites analysis to NiTi plain woven fabric to determine the effective elastic constants. The fabric model is based on actual weave geometry, including the presence of open gaps and wire cross-sectional area, and with the same diameter and alloy in the warp and weft. A woven NiTi ribbon has been manufactured (Figure 1) using a narrow weaving machine and has been tested in uniaxial tension. Planar fabric constants were measured at a range of temperatures. The analytically and experimentally derived constants for various weave patterns and cover factor combinations are presented and compared. It was determined that in uniaxial tension the fabric behaves like a collection of unidirectional wires, but has 78% of the rigidity, on average, across all test temperatures. This result is predicted by the fabric model with a 16% error, demonstrating that the proposed analytical model offers a useful tool for design and simulation of SMA fabrics.

This paper presents an initial step towards a new class of soft robotics materials, where localized, geometric patterning of smart materials can exhibit discrete levels of stiffness through the combinations of smart materials used. This work is inspired by a variety of biological systems where actuation is accomplished by modulating the local stiffness in conjunction with muscle contractions. Whereas most biological systems use hydrostatic mechanisms to achieve stiffness variability, and many robotic systems have mimicked this mechanism, this work aims to use smart materials to achieve this stiffness variability. Here we present the compositing of the low melting point Field’s metal, shape memory alloy Nitinol, and a low melting point thermoplastic Polycaprolactone (PCL), composited in simple beam structure within silicone rubber. The comparison in bending stiffnesses at different temperatures, which reside between the activation temperatures of the composited smart materials demonstrates the ability to achieve discrete levels of stiffnesses within the soft robotic tissue.

Self-fitting is the ability of a wearable, garment or body-mounted object to recover the exact shape and size of the human body. Self-fitting is highly desirable for wearable applications, ranging from medical and recreational health monitoring to wearable robotics and haptic feedback, because it enables complex devices to achieve accurate body proximity, which is often required for functionality. While garments designed with compliant fabrics can easily accomplish accurate fit for a range of body shapes and sizes, integrated actuators and sensors require fabric stiffness to prevent drift and deflection from the body surface. This paper merges smart materials and structures research with anthropometric analysis and functional apparel methodologies to present a novel, functionally gradient self-fitting garment designed to address the challenge of achieving accurate individual and population fit. This fully functional garment, constructed with contractile SMA knitted actuator fabrics, exhibits tunable %-actuation contractions between 4–50%, exerts minimal on-body pressure (≤ 1333Pa or 10 mmHg), and can be designed to actuate fully self-powered with body heat. The primary challenge in the development of the proposed garment is to design a functionally gradient system that does not exert significant pressure on part of the leg and/or remain oversized in others. Our research presents a new methodology for the design of contractile SMA knitted actuator garments, describes the manufacture of such self-fitting garments, and concludes with an experimental analysis of the garment performance evaluated through three-dimensional marker tracking.

Additive manufacturing has emerged as an alternative to traditional manufacturing technologies. In particular, industries like fluid power, aviation and robotics have the potential to benefit greatly from this technology, due to the design flexibility, weight reduction and compact size that can be achieved. In this work, the design process and advantages of using 3D printing to make soft linear actuators were studied and highlighted. This work explored the limitations of current additive manufacturing tolerances to fabricate a typical piston-cylinder assembly, and how enclosed bellow actuators could be used to overcome high leakage and friction issues experienced with a piston-cylinder type actuator. To do that, different 3D printing technologies were studied and evaluated (stereolithorgraphy and fused deposition modeling) in the pursuit of high-fidelity, cost-effective 3D printing. The initial attempt consisted of printing the soft actuators directly using flexible materials in a stereolithography-type 3D printer. However, these actuators showed low durability and poor performance. The lack of a reliable resin resulted in the replacement of this material by EcoFlex ® 00-30 silicone and the use of a 3D printed mold to cast the actuators. These molds included a 3-D printed dissolvable core inside the cast actuator in order to finish the manufacturing process in one single step. An experimental setup to evaluate the capabilities of these actuators was developed. Results are shown to assess the steady-state and the dynamic characteristics of these actuators. These tests resulted into the stroke-pressure and stroke-time responses for a specific load given different proportional valve inputs.

Stretchable strain sensors with large strain range, high sensitivity, and excellent reliability are of great interest for applications in soft robotics, wearable devices, and structure-monitoring systems. Unlike conventional template lithography-based approaches, 3D-printing can be used to fabricate complex devices in a simple and cost-effective manner. In this paper, we report 3D-printed stretchable strain sensors that embeds a flexible conductive composite material in a hyper-plastic substrate. Three commercially available conductive filaments are explored, among which the conductive thermoplastic polyurethane (ETPU) shows the highest sensitivity (gauge factor of 5), with a working strain range of 0%–20%. The ETPU strain sensor exhibits an interesting behavior where the conductivity increases with the strain. In addition, an experiment for measuring the wind speed is conducted inside a wind tunnel, where the ETPU sensor shows sensitivity to the wind speed beyond 5.6 m/s.

Robotic Materials are materials that have sensing, computation and, possibly actuation, distributed throughout the bulk of the material. In such a material, we envision semiconducting polymer based sensing, actuation, and information processing for on-board decision making to be designed, in tandem, with the smart product that will be implemented with the smart material. Prior work in printing polymer semiconductors for sensing and cognition have focused on highly energetic inkjet printing. Alternatively, we are developing liquid polymer extrusion processes to work hand-in-hand with existing solid polymer extrusion processes (such as Fused Deposition Manufacturing - FDM) to simultaneously deposit sensing, computation, actuation and structure. We demonstrate the successful extrusion printing of conductors and capacitors to impedance-match a new, higher-performance organic transistor design that solves the cascading problem of the device previously reported and is more amenable to liquid extrusion printing. Consequently, these printed devices are integrated into a sheet material that is folded into a 3-D, six-legged walking machine with attached electric motor.

This study aims to examine the coiling and uncoiling motion of a soft pneumatic actuator reinforced with tilted helix fibers. Coiling motion can be quite useful for robotic manipulation and locomotion purposes. This research proposes and investigates a novel actuator that is inspired and derived from the unique cell wall architecture in the seed appendage of Stork’s Bill plant (Erodium Gruinum). These plant cells are reinforced by cellulose fibers distributed in a tilted helix pattern — helixes that are tilted at a certain angle with respect to the longitudinal axis of the cell. As a result, the seed appendage can coil and uncoil via a combination of twisting and bending. This paper discusses the design, fabrication, and testing of a soft actuator that can mimic this sophisticated motion. This actuator consists of Kevlar fiber thread wrapped around a silicon rubber body that has the shape of a tube. The tube will be capped at both ends so that it can be pressurized internally to induce motion. Once the design parameter has been chosen, the soft actuator are fabricated by 1) designing and 3D printing molds, 2) tube casting and fiber wrapping, and 3) creating the end caps for pressure sealing. Carefully executing these fabrication steps is essential because any errors could give undesired deformation. Several soft actuators prototypes are fabricated based on different design choices regarding the actuator radius, tube wall thickness, and the number of tilted helix fibers (aka. fiber coverage). Proof-of-concept tests show that these actuator prototypes can indeed exhibit a combined twisting and bending under internal pressurization: all are the necessary receipts to achieve the coiling and uncoiling motion. Result of this paper can pave the way for a new family of soft actuators capable of unprecedented and sophisticated actuation motions, which are particularly appealing for soft robot application.

Photochemical actuation systems, those that employ coupled photo-stimuli and chemical reactions to power and control mechanical motion, have the potential to combine the benefits of precise light driven control with chemical energy storage. Furthermore, these systems are inherently soft, making them ideal for use in the emerging field of soft robotics. However, such systems have received comparatively little attention, perhaps due to the poor cycle life and limited activation time of past systems. Here we address these two challenges by switching from the technique of past systems, that of aqueous photoacid solutions and pH-responsive hydrogel actuators, to one employing organic solvents instead. While this switch of solvents successfully eliminates cycle life constraints and allows for tuning of the activation recovery time it also shifts the relative activation point of the hydrogel actuator in such a way that actuation is no longer observed. Several options for addressing this are discussed, with the prospect of using the lessons learned within to make a more informed selection of a different photoacid compound considered the most feasible. While the exploration of photochemical actuation systems is still in a nascent stage, we have great hope for such systems to form the basis of future smart machines with unique functionality.

In this paper, we present characterization results for thermal, mechanical, and electrical properties of a 3D-printed conductive polylactic acid (PLA) composite material. The material exhibits electrically controllable stiffness, allowing for the fabrication of novel robotic and biomedical devices. In particular, an applied voltage induces a Joule heating effect, which modulates the material stiffness. Dumbbell samples are 3D-printed and loaded into a universal testing machine (UTM) to measure their Young’s moduli at different temperatures. The conductive PLA composite shows 98.6% reduction of Young’s modulus, from 1 GPa at room temperature to 13.6 MPa at 80 °C, which is fully recovered when cooled down to its initial temperature. Measurements with differential scanning calorimeter (DSC) and thermal diffusivity analyzer are conducted to investigate the thermal behavior of this material. Electrical conductivity of the material is measured under different temperatures, where the resistivity increases about 60% from 30 °C to 100 °C and hysteresis between the resistivity and the temperature is observed. These tests have shown that the conductive PLA composite has a glass transition temperature (Tg) of 56.7 °C, melting point (Tm) of 153.8 °C, and thermal conductivity of 0.366 W/(mK). The obtained results can be used as design parameters in finite element models and computational tools to rapidly simulate multi-material components for several applications such as object manipulation, grasping, and flow sensing.

Artificial muscle systems have the potential to impact many technologies ranging from advanced prosthesis to miniature robotics. Recently, it has been shown that twisting drawn polymer monofilaments, such as nylon fishing line or sewing thread, can result in a biomimetic thermally activated torsional actuator. The actuation phenomenon in these twisted polymer actuators (TPAs) is thought to be a result of an untwisting that occurs about the fiber’s axis due to an anisotropic thermal expansion. Before being twisted, the precursor fibers are comprised of polymer chains that are aligned axially. During fabrication of TPAs, the polymer chains reorient as the precursor fiber is twisted about the central axis of the monofilament. At the end of the fabrication process, the TPA is annealed in order to relieve internal stresses and to keep the fiber in the twisted configuration. The mechanism of untwisting actuation is generally thought to be a result of radial expansion and axial contraction. After being twisted, these radial and axial expansion relationships remain relatively unchanged, but the polymer chain direction is no longer axially aligned. Thus, upon heating the twisted fibers of the TPA, the fibers untwist and torsional actuation occurs. This actuation phenomenon has been used in the past to create linear actuators, but can also be use directly as a torsional actuator. Compared to other torsional actuators TPAs are low cost, lightweight, and can actuate reasonably high torques per unit volume. However, because TPAs are thermally activated, they may not be suitable for all applications. In this work, we present a novel TPA design for use as a torsional actuator for miniature actuation and artificial muscle applications. Our design bundles twisted monofilaments to increase the torque. Both fabrication and testing methods of the new design are presented. Results for temperature versus torsional displacement under various loads give insights as to how these actuators may be used and the reversibility of the actuation process under different fabrication loads. Additionally, comparisons are made between these bundled actuators and similarly loaded single TPA monofilament actuation.

As the soft robotics industry continues to grow, the need for new materials and simplified manufacturing techniques are essential. Of interest is the development of highly flexible strain sensors that are easily integrated into these robotic components. Current strain sensing solutions using piezoresistive materials often involve complex fabrication techniques with multiple steps. Recent work by the authors has shown that thermoplastic polyurethane/multiwall carbon nanotubes (TPU/MWCNT) has good piezoresistive behavior and can be easily fabricated into strain sensors using Fused Deposition Modeling (FDM). This work expands upon that effort to characterize the mechanical properties of FDM-printed TPU/MWCNT as a function of the FDM processing parameters. In this study, the air gap, raster orientation, and MWCNT weight percent were varied and tensile tests performed. The stress-strain behavior, modulus of elasticity, and ultimate tensile strength (UTS) are compared to assess the influence of the processing conditions. Optical microscopy was also carried out to correlate the mechanical behavior to the printed mesostructures. The results show that with increased MWCNT content, the UTS decreased by as much at 47% for 2wt.%MWCNT, while the modulus of elasticity increased by 54%, compared to those of pure TPU. The results of this work provide an understanding of the mechanical performance in relation to the print parameters and sets the base to tune the mechanical properties of printed flexible functional nanocomposites.