Dielectric elastomer actuators (DEAs) are energy efficient, compact, and operate silently. DEAs consist of an elastomer membrane sandwiched between two stretchable electrodes. Optimizing the quality of both the elastomer and the stretchable electrodes is essential to improve the DEAs performance. Herein, novel strategies are reported to achieve fast, reliable, and low‐cost fabrication of DEAs. The strategies utilize a soft brush to directly pattern carbon nanotube (CNT) powder on the elastomer membrane and tune the mechanical and surface‐adhesiveness characteristics of a polydimethylsiloxane (PDMS) membrane by altering the mixing ratio of the curing agent and base polymer. A uniaxial engineering tensile test on PDMS indicates that a softer material is formed when less curing agent is used. The softer PDMS has a sticky surface, which allows the CNT powder to be physically bound to the surface. Field‐emission scanning electron microscopy (FE‐SEM) images prove that the strong CNT network is formed on the surface of the elastomer. The electromechanical investigation also indicates that the electrical conductivity is improved for a stickier PDMS surface. The optimal performance of PDMS 30‐1 in static and cyclic DEA tests shows that the brushing of CNT combined with soft and sticky elastomer membranes can increase the DEA performance.

To increase safety and reduce the electric circuit cost, dielectric elastomer actuators (DEAs) must operate below the kV range. The simplest strategy to reduce the voltage operation is to decrease the dielectric elastomer membrane thickness. This research aims to demonstrate DEAs with a nanometric uniform thickness that can operate at a low voltage (below 70 V) and a high frequency. We use the roll-to-roll process to fabricate a 600 nm thick stretchable PDMS (polydimethylsiloxane) nanosheet and a 200 nm thick conductive nanosheet. These nanosheet-DEAs are tested in high-frequency operations of DC voltage below 70 V and in a frequency range of 1–30 kHz.

Soft robotics and wearable devices are promising technologies due to their flexibility. As human-soft robot interaction technologies advance, the interest in stretchable sensor devices has increased. Currently, the main challenge in developing stretchable sensors is preparing high-quality sensors via a simple and cost-effective method. This study introduces the do-it-yourself (DIY)-approach to fabricate a carbon nanotube (CNT) powder-based stretchable sensor. The fabrication strategy utilizes an automatic brushing machine to pattern CNT powder on the elastomer. The elastomer ingredients are optimized to increase the elastomer compatibility with the brushing method. We found that polydimethylsiloxane-polyethyleneimine (PDMS-PEIE) is 50% more stretchable and 63% stickier than previously reported PDMS 30-1. With these improved elastomer characteristics, PDMS-PEIE/multiwalled CNT (PDMS-PEIE/MWCNT-1) strain sensor can realize a gauge factor of 6.2–8.2 and a responsivity up to 25 ms. To enhance the compatibility of the powder-based stretchable sensor for a wearable device, the sensor is laminated using a thin Ecoflex membrane. Additionally, system integration of the stretchable sensors are demonstrated by embedding it into a cotton-glove and a microcontroller to control a virtual hand. This cost-effective DIY-approach are expected to greatly contribute to the development of wearable devices since the technology is simple, economical, and reliable.

Stereolithography has emerged as a recent method in fabricating complex structures with high accuracy. Components using resin have poorer properties. The current study investigates the improvement in the properties of nano-graphite composites fabricated by the SLA technique. The properties are compared for plain resin and 0.2%, 0.5%, 1%, 3%, and 5% (w/v) of nano-graphite mixed with the UV-curable resin. Various analyses were conducted, including viscosity, UV spectroscopy, moisture content, water absorption, gel content, tensile, bending, hardness testing, and microscopic characterization. The results from the experiments showed a difference in the results of each percentage of the specimen tested, such as the specimen property, which shows that the greater the percentage of nano-graphite added (5%), the opaquer the specimen will appear and less light will be reflected. Viscosity testing shows that the greater the percentage of nano-graphite added to the resin, the greater the viscosity. UV spectroscopy testing produced information about the electronic structure and the structure of molecules, such as their composition, purity, and concentration. Observations from the moisture content analysis found that the moisture content in specimens with higher percentages of nano-graphite affected physical and mechanical properties, leading to easier warping, cracking, decreased strength, etc. Tensile and bending testing shows that the greater the percentage of nano-graphite added, the greater the effect on physical and mechanical properties, including fracture. However, certain tests did not consistently yield significant variations among specimens when different percentages of nano-graphite were added, as particularly evident in chemical resistance testing. This study offers valuable insights into the application of nano-graphite composites fabricated via the SLA method.

Human-movement recognition is a novel challenge in soft robotics. In recent years, there have been several attempts to develop soft wearable devices for supporting human-robot interfaces. Many algorithms and programming languages are available to integrate a wearable device with a soft robot. One such promising algorithm is reservoir computing (RC), which includes of a group of recurrently and randomly connected nodes. The RC model can easily process multidimensional signal data and can handle nonlineardata and has been extensively used in robotic control. It has been reported that the RC algorithm can speed up network training and solve complex data sets. However, the main existing limitations in hand-locomotion classification are the considerable run-time and the delayed response. In this study, we figure out the best machine learning algorithms to handle three-dimensional hand-gesture data. We employ a two-part strategy: a loopback filter is included in the preprocessing of the initial dataset to support the 3-dimensional (3D) signs of each hand posture; subsequently, the training dataset is applied to the machine learning algorithm which includes an artificial neural network (ANN), convolutional neural network (CNN), long short-term memory(LSTM), and reservoir computing(RC). Each training network is optimized with various hyperparameters. Furthermore, we compare the performance of several machine-learning algorithms in classifying the three-dimensional hand-signal posture data. The results show that the classification of nonlinear hand-locomotion signals by RC requires a comparatively shorter training duration (12 minutes for training times), and that optimal accuracy 94.17, precision 94.10, and recall 93.99 is realized for time series data.