Decoding lip language using triboelectric sensors with deep learning

Microbial biofilm-based hydrovoltaic pressure sensor with ultrahigh sensitivity for self-powered flexible electronics

Developing the integration of self-powered detection with both dynamic and static forces is a significant challenge in promoting intelligent technology systems. Herein, we introduce an innovative microbial biofilm based-hydrovoltaic pressure sensor (mBio-HPS) using whole-cell Geobacter sulfurreducens, which successfully combines self-powered functionality and static pressure detection within a single device. The mBio-HPS exhibited a sensitivity of up to 8968.7 kPa⁻1 (at 1 kPa) in the 0.4-25 kPa regime without external power supply. Moreover, the mBio-HPS demonstrated the fastest reported response speed to date, with a remarkable response time of 112.5 μs, enabling effective detection of both dynamic and static forces while maintaining stability during an extensive 30,000 s testing. Experimental validation using a sensor-integrated array showed its outstanding real-time detection capabilities for both dynamic and static pressure, highlighting its outstanding potential for electronic skin applications. This unprecedented concept of a hydrovoltaic pressure sensor also offers new insights into the development of high-performance self-powered electronics.

AI-Driven TENGs for Self-Powered Smart Sensors and Intelligent Devices

Triboelectric nanogenerators (TENGs) are emerging as transformative technologies for sustainable energy harvesting and precision sensing, offering eco-friendly power generation from mechanical motion. They harness mechanical energy while enabling self-sustaining sensing for self-powered devices. However, challenges such as material optimization, fabrication techniques, design strategies, and output stability must be addressed to fully realize their practical potential. Artificial intelligence (AI), with its capabilities in advanced data analysis, pattern recognition, and adaptive responses, is revolutionizing fields like healthcare, industrial automation, and smart infrastructure. When integrated with TENGs, AI can overcome current limitations by enhancing output, stability, and adaptability. This review explores the synergistic potential of AI-driven TENG systems, from optimizing materials and fabrication to embedding machine learning and deep learning algorithms for intelligent real-time sensing. These advancements enable improved energy harvesting, predictive maintenance, and dynamic performance optimization, making TENGs more practical across industries. The review also identifies key challenges and future research directions, including the development of low-power AI algorithms, sustainable materials, hybrid energy systems, and robust security protocols for AI-enhanced TENG solutions.

Exploration of Advanced Applications of Triboelectric Nanogenerator-Based Self-Powered Sensors in the Era of Artificial Intelligence

The integration of Deep Learning with sensor technologies has significantly advanced the field of intelligent sensing and decision making by enhancing perceptual capabilities and delivering sophisticated data analysis and processing functionalities. This review provides a comprehensive overview of the synergy between Deep Learning and sensors, with a particular focus on the applications of triboelectric nanogenerator (TENG)-based self-powered sensors combined with artificial intelligence (AI) algorithms. First, the evolution of Deep Learning is reviewed, highlighting the advantages, limitations, and application domains of several classical models. Next, the innovative applications of intelligent sensors in autonomous driving, wearable devices, and the Industrial Internet of Things (IIoT) are discussed, emphasizing the critical role of neural networks in enhancing sensor precision and intelligent processing capabilities. The review then delves into TENG-based self-powered sensors, introducing their self-powered mechanisms based on contact electrification and electrostatic induction, material selection strategies, novel structural designs, and efficient energy conversion methods. The integration of TENG-based self-powered sensors with Deep Learning algorithms is showcased through their groundbreaking applications in motion recognition, smart healthcare, smart homes, and human-machine interaction. Finally, future research directions are outlined, including multimodal data fusion, edge computing integration, and brain-inspired neuromorphic computing, to expand the application of self-powered sensors in robotics, space exploration, and other high-tech fields. This review offers theoretical and technical insights into the collaborative innovation of Deep Learning and self-powered sensor technologies, paving the way for the development of next-generation intelligent systems.

Soft Materials and Devices Enabling Sensorimotor Functions in Soft Robots

Sensorimotor functions, the seamless integration of sensing, decision-making, and actuation, are fundamental for robots to interact with their environments. Inspired by biological systems, the incorporation of soft materials and devices into robotics holds significant promise for enhancing these functions. However, current robotics systems often lack the autonomy and intelligence observed in nature due to limited sensorimotor integration, particularly in flexible sensing and actuation. As the field progresses toward soft, flexible, and stretchable materials, developing such materials and devices becomes increasingly critical for advanced robotics. Despite rapid advancements individually in soft materials and flexible devices, their combined applications to enable sensorimotor capabilities in robots are emerging. This review addresses this emerging field by providing a comprehensive overview of soft materials and devices that enable sensorimotor functions in robots. We delve into the latest development in soft sensing technologies, actuation mechanism, structural designs, and fabrication techniques. Additionally, we explore strategies for sensorimotor control, the integration of artificial intelligence (AI), and practical application across various domains such as healthcare, augmented and virtual reality, and exploration. By drawing parallels with biological systems, this review aims to guide future research and development in soft robots, ultimately enhancing the autonomy and adaptability of robots in unstructured environments.

Wireless, Multisensor Integrated Bioelectronics for Real-Time Monitoring and Assessment of Muscle Atrophy

Real-time monitoring and evaluation of muscle atrophy are vital for efficient treatment and effective diagnosis. Although some systems have been developed to monitor loss of muscle mass or strength, they are bulky and nonspecific, cannot be applied to the body, and cannot assess the degree of muscle atrophy. Our research focuses on developing a wireless multisensor wearable system (WMWS) for on-body monitoring and assessment of atrophy, which utilizes a single-electrode triboelectric nanogenerator (S-TENG) and electrochemical creatinine (CREA) sensor to achieve real-time acquisition of plantar pressure and interstitial fluid (ISF) CREA concentration. In addition, LSTM (long short-term memory) and SVM (support vector machine) machine learning (ML) algorithms also prove that our multisensor strategy can effectively improve the recognition rate of different degrees of atrophy (the highest accuracy reached 92.32%). Overall, our work makes it possible to monitor and grade muscle atrophy remotely in real time.

Machine learning-assisted wearable sensing systems for speech recognition and interaction

The human voice stands out for its rich information transmission capabilities. However, voice communication is susceptible to interference from noisy environments and obstacles. Here, we propose a wearable wireless flexible skin-attached acoustic sensor (SAAS) capable of capturing the vibrations of vocal organs and skin movements, thereby enabling voice recognition and human-machine interaction (HMI) in harsh acoustic environments. This system utilizes a piezoelectric micromachined ultrasonic transducers (PMUT), which feature high sensitivity (-198 dB), wide bandwidth (10 Hz-20 kHz), and excellent flatness (±0.5 dB). Flexible packaging enhances comfort and adaptability during wear, while integration with the Residual Network (ResNet) architecture significantly improves the classification of laryngeal speech features, achieving an accuracy exceeding 96%. Furthermore, we also demonstrated SAAS's data collection and intelligent classification capabilities in multiple HMI scenarios. Finally, the speech recognition system was able to recognize everyday sentences spoken by participants with an accuracy of 99.8% through a deep learning model. With advantages including a simple fabrication process, stable performance, easy integration, and low cost, SAAS presents a compelling solution for applications in voice control, HMI, and wearable electronics.

Deep-Learning-Based Analysis of Electronic Skin Sensing Data

E-skin is an integrated electronic system that can mimic the perceptual ability of human skin. Traditional analysis methods struggle to handle complex e-skin data, which include time series and multiple patterns, especially when dealing with intricate signals and real-time responses. Recently, deep learning techniques, such as the convolutional neural network, recurrent neural network, and transformer methods, provide effective solutions that can automatically extract data features and recognize patterns, significantly improving the analysis of e-skin data. Deep learning is not only capable of handling multimodal data but can also provide real-time response and personalized predictions in dynamic environments. Nevertheless, problems such as insufficient data annotation and high demand for computational resources still limit the application of e-skin. Optimizing deep learning algorithms, improving computational efficiency, and exploring hardware-algorithm co-designing will be the key to future development. This review aims to present the deep learning techniques applied in e-skin and provide inspiration for subsequent researchers. We first summarize the sources and characteristics of e-skin data and review the deep learning models applicable to e-skin data and their applications in data analysis. Additionally, we discuss the use of deep learning in e-skin, particularly in health monitoring and human-machine interactions, and we explore the current challenges and future development directions.

Emerging Wearable Acoustic Sensing Technologies

Sound signals not only serve as the primary communication medium but also find application in fields such as medical diagnosis and fault detection. With public healthcare resources increasingly under pressure, and challenges faced by disabled individuals on a daily basis, solutions that facilitate low-cost private healthcare hold considerable promise. Acoustic methods have been widely studied because of their lower technical complexity compared to other medical solutions, as well as the high safety threshold of the human body to acoustic energy. Furthermore, with the recent development of artificial intelligence technology applied to speech recognition, speech recognition devices, and systems capable of assisting disabled individuals in interacting with scenes are constantly being updated. This review meticulously summarizes the sensing mechanisms, materials, structural design, and multidisciplinary applications of wearable acoustic devices applied to human health and human-computer interaction. Further, the advantages and disadvantages of the different approaches used in flexible acoustic devices in various fields are examined. Finally, the current challenges and a roadmap for future research are analyzed based on existing research progress to achieve more comprehensive and personalized healthcare.

Triboelectric Bending Sensors for AI-Enabled Sign Language Recognition

Human-machine interfaces and wearable electronics, as fundamentals to achieve human-machine interactions, are becoming increasingly essential in the era of the Internet of Things. However, contemporary wearable sensors based on resistive and capacitive mechanisms demand an external power, impeding them from extensive and diverse deployment. Herein, a smart wearable system is developed encompassing five arch-structured self-powered triboelectric sensors, a five-channel data acquisition unit to collect finger bending signals, and an artificial intelligence (AI) methodology, specifically a long short-term memory (LSTM) network, to recognize signal patterns. A slider-crank mechanism that precisely controls the bending angle is designed to quantitively assess the sensor's performance. Thirty signal patterns of sign language of each letter are collected and analyzed after the environment noise and cross-talks among different channels are reduced and removed, respectively, by leveraging low pass filters. Two LSTM models are trained using different training sets, and four indexes are introduced to evaluate their performance, achieving a recognition accuracy of 96.15%. This work demonstrates a novel integration of triboelectric sensors with AI for sign language recognition, paving a new application avenue of triboelectric sensors in wearable electronics.

Luminescent Tactile Sensor System for Robots: Enhancing Human-Computer Interaction in Complex Dark Environments

In order to achieve interaction and collaboration with humans, robots need to have the ability for tactile perception of simulating human. Traditional methods use electrically connected sensors with complex arrays, leading to intricate wiring, high manufacturing costs, and demanding current environments. A flexible sensor with simple structure, easy preparation process, and low cost based on triboluminescence effect is proposed in this paper, which avoids the complex array and wiring of traditional sensors. The study discusses the relationship between luminescent intensity and factors such as luminescent particle content, luminescent layer thickness, encapsulation layer thickness, and friction layer thickness. It also analyzes the mechanism of luminescence. A micro charge-coupled device is configured for the luminescent unit to collect optical information and is integrated with the robot's manipulator for testing. A simple sensing system is constructed to demonstrate environmental perception, acquiring, and feeding back shape and size information of contact objects in the dark. The system successfully identifies and judges target objects in complex dark environments, offering insights for applications such as unmanned assembly lines. It overcomes the challenge of intricate electrical connections, paving new avenues for intelligent object recognition research in human-computer interaction.

Triboelectric sensor gloves for real-time behavior identification and takeover time adjustment in conditionally automated vehicles

The takeover issue, especially the setting of the takeover time budget, is a critical factor restricting the implementation and development of conditionally automated vehicles. The general fixed takeover time budget has certain limitations, as it does not take into account the driver's non-driving behaviors. Here, we propose an intelligent takeover assistance system consisting of all-round sensing gloves, a non-driving behavior identification module, and a takeover time budget determination module. All-round sensing gloves based on triboelectric sensors seamlessly detect delicate motions of hands and interactions between hands and other objects, and then transfer the electrical signals to the non-driving behavior identification module, which achieves an accuracy of 94.72% for six non-driving behaviors. Finally, combining the identification result and its corresponding minimum takeover time budget obtained through the takeover time budget determination module, our system dynamically adjusts the takeover time budget based on the driver's current non-driving behavior, significantly improving takeover performance in terms of safety and stability. Our work presents a potential value in the application and implementation of conditionally automated vehicles.

An intelligent hybrid-fabric wristband system enabled by thermal encapsulation for ergonomic human-machine interaction

Human-machine interaction has emerged as a revolutionary and transformative technology, bridging the gap between human and machine. Gesture recognition, capitalizing on the inherent dexterity of human hands, plays a crucial role in human-machine interaction. However, existing systems often struggle to meet user expectations in terms of comfort, wearability, and seamless daily integration. Here, we propose a handwriting recognition technology utilizing an intelligent hybrid-fabric wristband system. This system integrates spun-film sensors into textiles to form the smart fabric, enabling intelligent functionalities. A thermal encapsulation process is proposed to bond multiple spun-films without additional materials, ensuring the lightweight, breathability, and stretchability of the spun-film sensors. Furthermore, recognition algorithms facilitate precise accurate handwriting recognition of letters, with an accuracy of 96.63%. This system represents a significant step forward in the development of ergonomic and user-friendly wearable devices for enhanced human-machine interaction, particularly in the virtual world.

Piezoresistive Sensor Based on Porous Sponge with Superhydrophobic and Flame Retardant Properties for Motion Monitoring and Fire Alarm

Polyurethane sponge is frequently selected as a substrate material for constructing flexible compressible sensors due to its excellent resilience and compressibility. However, being highly hydrophilic and flammable, it not only narrows the range of use of the sensor but also poses a great potential threat to human safety. In this paper, a conductive flexible piezoresistive sensor (CHAP-PU) with superhydrophobicity and high flame retardancy was prepared by a simple dip-coating method using A-CNTs/HGM/ADP coatings deposited on the surface of a sponge skeleton and modified with polydimethylsiloxane. With great sensitivity and durability (>3000 cycles) as well as fast response/recovery time (152 ms/178 ms), the sensor is capable of monitoring human movement as a wearable device. The modified material surface has a hydrophobicity angle of 153°, which provides significant self-cleaning and weather resistance. Furthermore, the CHAP-PU sensor is able to respond stably to underwater movements. Importantly, when the sponge was directly exposed to an open flame, no flame spreading or dripping of molten material was detected, indicating excellent flame retardancy. Meanwhile, CHAP-PU was also equipped as a smart fire alarm system, and the results showed that an alarm signal was triggered within 2 s under flame erosion. Therefore, the flame-retardant superhydrophobic CHAP-PU sponge-based sensor shows great potential for human motion detection and fire alarm applications.

Triboelectric-Inertial Sensing Glove Enhanced by Charge-Retained Strategy for Human-Machine Interaction

As technology advances, human-machine interaction (HMI) demands more intuitive and natural methods. To meet this demand, smart gloves, capable of capturing intricate hand movements, are emerging as vital HMI tools. Moreover, triboelectric-based sensors, with their self-powered, cost-effective, and material various characteristics, can offer promising solutions for enhancing existing glove systems. However, a key limitation of these sensors is that charge leakage in the measurement circuit results in capturing only transient signals, rather than continuous changes. To address this issue, a charge-retained circuit that effectively prevents triboelectric signal attenuation is developed, enabling accurate measurement of continuous finger movements. This innovation forms the foundation of a highly integrated smart glove system, enhancing HMI functionality by combining continuous triboelectric signals with inertial sensor data. The system showcases a diverse range of applications, including complex robotic control, virtual reality interaction, smart home lighting adjustments, and intuitive interface operations. Furthermore, by leveraging artificial intelligence (AI) techniques, the system achieves accurate recognition of complex sign language with an impressive 99.38% accuracy. This work presents a charge-retained approach for continuous sensing with triboelectric-based sensors, offering valuable insights for developing future multifunctional HMI and sign language recognition systems. The proposed smart glove system, with its dual-mode sensing and AI integration, holds great potential for revolutionizing various domains and enhancing user experiences.

Highly stable and ultra-fast vibration-responsive flexible iontronic sensors for accurate acoustic signal recognition

Wearable verbal language servers function as sophisticated and effective tools for fostering intelligent interactions between humans and machines. In the realm of collecting acoustic vibration signals, flexible iontronic pressure sensors have demonstrated their efficacy by incorporating microstructures into the functional layer, resulting in heightened pressure sensitivity. However, the substantial viscosity of the integrated iontronic materials or the lack of bonding at the heterogeneous interface emerges as a significant hindrance to capacitance recovery, leading to sluggish response speeds and mechanical instability. Here, we address the issue by introducing hydrogen bonding between naturally microstructured protein micro-fibers and hydrophilic ionic hydrogel into the dielectric layer. Due to the good resilience of protein micro-fibers and the enahnced interfacial bonding, this flexible vibration sensor demonstrates outstanding performance characteristics, featuring exceptional signal stability, a high-pressure resolution of 522 pF kPa-1, an ultra-fast response time of 0.6 ms, and a relaxation time of 0.6 ms, with a limit of detection (LOD) of 0.12 Pa, making it well-suited for acoustic vibration acquisition. By using a one-dimensional convolutional neural network (1D-CNN) deep learning to process and recognize collected acoustic signals, our sensor achieved an impressive accuracy of 98.2%. These wearable vibration sensors exemplify promising versatile applications in biometric authentication, personalized services, and human-computer interaction.

Artificial intelligence enabled smart mask for speech recognition for future hearing devices

In recent years, Lip-reading has emerged as a significant research challenge. The aim is to recognise speech by analysing Lip movements. The majority of Lip-reading technologies are based on cameras and wearable devices. However, these technologies have well-known occlusion and ambient lighting limitations, privacy concerns as well as wearable device discomfort for subjects and disturb their daily routines. Furthermore, in the era of coronavirus (COVID-19), where face masks are the norm, vision-based and wearable-based technologies for hearing aids are ineffective. To address the fundamental limitations of camera-based and wearable-based systems, this paper proposes a Radio Frequency Identification (RFID)-based smart mask for a Lip-reading framework capable of reading Lips under face masks, enabling effective speech recognition and fostering conversational accessibility for individuals with hearing impairment. The system uses RFID technology to make Radio Frequency (RF) sensing-based Lip-reading possible. A smart RFID face mask is used to collect a dataset containing three different classes of vowels (A, E, I, O, U), Consonants (F, G, M, S), and words (Fish, Goat, Meal, Moon, Snake). The collected data are fed into well-known machine-learning models for classification. A high classification accuracy is achieved by individual classes and combined datasets. On the RFID combined dataset, the Random Forest model achieves a high classification accuracy of 80%.

Charge Generation and Enhancement of Key Components of Triboelectric Nanogenerators: A Review

The past decade has witnessed remarkable progress in high-performance Triboelectric nanogenerators (TENG) with the design and synthesis of functional dielectric materials, the exploration of novel dynamic charge transport mechanisms, and the innovative design of architecture, making it one of the most crucial technologies for energy harvesting. High output charge density is fundamental for TENG to expand its application scope and accelerate industrialization; it depends on the dynamic equilibrium of charge generation, trapping, de-trapping, and migration within its core components. Here, this review classifies and summarizes innovative approaches to enhance the charge density of the charge generation, charge trapping, and charge collection layers. The milestone of high charge density TENG is reviewed based on material selection and innovative mechanisms. The state-of-the-art principles and techniques for generating high charge density and suppressing charge decay are discussed and highlighted in detail, and the distinct charge transport mechanisms, the technologies of advanced materials preparation, and the effective charge excitation strategy are emphatically introduced. Lastly, the bottleneck and future research priorities for boosting the output charge density are summarized. A summary of these cutting-edge developments intends to provide readers with a deep understanding of the future design of high-output TENG.

Multi-Attribute Triboelectric Materials and Innovative Applications Via TENGs

Triboelectric nanogenerators (TENGs) as an avant-garde technology that transforms mechanical energy into electrical energy, offering a new direction for green energy and sustainable development. By means of high-efficiency TENGs, conventional materials as new triboelectric materials have exhibited multi-attribute characteristics, achieving innovative applications in the field of micro-nano energy harvesting and self-powered sensing. The progress of TENGs technology with the triboelectric materials is complementary and mutually promoting. On the one hand, one of the cruxes of TENGs lies in the triboelectric materials, which have a decisive impact on their performance. On the other hand, as the research and application of TENGs continue to deepen, higher demands are placed on triboelectric materials, which in turn promotes the advancement of the entire material system as well as the fields of materials science and physics. This work aims to delve into the characteristics, types, preferred choices, and modification treatments of triboelectric materials on the performances of TENGs, hoping to provide guidance and insights for future research and applications.

Ultrasensitive Wearable Pressure Sensors with Stress-Concentrated Tip-Array Design for Long-Term Bimodal Identification

The great challenges for existing wearable pressure sensors are the degradation of sensing performance and weak interfacial adhesion owing to the low mechanical transfer efficiency and interfacial differences at the skin-sensor interface. Here, an ultrasensitive wearable pressure sensor is reported by introducing a stress-concentrated tip-array design and self-adhesive interface for improving the detection limit. A bipyramidal microstructure with various Young's moduli is designed to improve mechanical transfer efficiency from 72.6% to 98.4%. By increasing the difference in modulus, it also mechanically amplifies the sensitivity to 8.5 V kPa-1 with a detection limit of 0.14 Pa. The self-adhesive hydrogel is developed to strengthen the sensor-skin interface, which allows stable signals for long-term and real-time monitoring. It enables generating high signal-to-noise ratios and multifeatures when wirelessly monitoring weak pulse signals and eye muscle movements. Finally, combined with a deep learning bimodal fused network, the accuracy of fatigued driving identification is significantly increased to 95.6%.

Enhancing human computer interaction with coot optimization and deep learning for multi language identification

Human-Computer Interaction (HCI) is a multidisciplinary field focused on designing and utilizing computer technology, underlining the interaction interface between computers and humans. HCI aims to generate systems that allow consumers to relate to computers effectively, efficiently, and pleasantly. Multiple Spoken Language Identification (SLI) for HCI (MSLI for HCI) denotes the ability of a computer system to recognize and distinguish various spoken languages to enable more complete and handy interactions among consumers and technology. SLI utilizing deep learning (DL) involves using artificial neural networks (ANNs), a subset of DL models, to automatically detect and recognize the language spoken in an audio signal. DL techniques, particularly neural networks (NNs), have succeeded in various pattern detection tasks, including speech and language processing. This paper develops a novel Coot Optimizer Algorithm with a DL-Driven Multiple SLI and Detection (COADL-MSLID) technique for HCI applications. The COADL-MSLID approach aims to detect multiple spoken languages from the input audio regardless of gender, speaking style, and age. In the COADL-MSLID technique, the audio files are transformed into spectrogram images as a primary step. Besides, the COADL-MSLID technique employs the SqueezeNet model to produce feature vectors, and the COA is applied to the hyperparameter range of the SqueezeNet method. The COADL-MSLID technique exploits the SLID process's convolutional autoencoder (CAE) model. To underline the importance of the COADL-MSLID technique, a series of experiments were conducted on the benchmark dataset. The experimentation validation of the COADL-MSLID technique exhibits a greater accuracy result of 98.33% over other techniques.

Toward an AI Era: Advances in Electronic Skins

Electronic skins (e-skins) have seen intense research and rapid development in the past two decades. To mimic the capabilities of human skin, a multitude of flexible/stretchable sensors that detect physiological and environmental signals have been designed and integrated into functional systems. Recently, researchers have increasingly deployed machine learning and other artificial intelligence (AI) technologies to mimic the human neural system for the processing and analysis of sensory data collected by e-skins. Integrating AI has the potential to enable advanced applications in robotics, healthcare, and human-machine interfaces but also presents challenges such as data diversity and AI model robustness. In this review, we first summarize the functions and features of e-skins, followed by feature extraction of sensory data and different AI models. Next, we discuss the utilization of AI in the design of e-skin sensors and address the key topic of AI implementation in data processing and analysis of e-skins to accomplish a range of different tasks. Subsequently, we explore hardware-layer in-skin intelligence before concluding with an analysis of the challenges and opportunities in the various aspects of AI-enabled e-skins.

A Conductive and Anti-impact Composite for Flexible Piezoresistive Sensors

Flexible piezoresistive sensors, which can convert specific mechanical information (such as compression, bending, tensile, and torsion) into a resistance value change signal through the piezoresistive effect, have attracted more and more attention. However, how to achieve the simple, low-cost fabrication of a piezoresistive sensor is still a challenge. Herein, we report a facile strategy that introduces conductive carbon black (CB) and shear thickening gel (SG) composite into a melamine sponge (MS) to generate an MS-SG-CB composite with a unique force-electric coupling effect. A flexible sensor derived from the MS-SG-CB composite can not only accurately identify deformation signals during static stretching and compression while monitoring human movement status in real time but also recognize electrical signals under dynamic impact in a very short time (6 ms). The 3 × 3 flexible array built on this basis can accurately identify the mass and position of heavy objects. Furthermore, based on the flame-retardant properties of MS, the flame-retardant ammonium polyphosphate (APP) is further introduced into MS-SG-CB to obtain MS-SG-CB-APP composite with excellent flame retardancy and stable temperature electrical response behavior, expanding its application in the field of high temperature trigger alarm.

Environmentally Robust Triboelectric Tire Monitoring System for Self-Powered Driving Information Recognition via Hybrid Deep Learning in Time-Frequency Representation

Developing a robust artificial intelligence of things (AIoT) system with a self-powered triboelectric sensor for harsh environment is challenging because environmental fluctuations are reflected in triboelectric signals. This study presents an environmentally robust triboelectric tire monitoring system with deep learning to capture driving information in the triboelectric signals generated from tire-road friction. The optimization of the process and structure of a laser-induced graphene (LIG) electrode layer in the triboelectric tire is conducted, enabling the tire to detect universal driving information for vehicles/robotic mobility, including rotation speeds of 200-2000 rpm and contact fractions of line. Employing a hybrid model combining short-term Fourier transform with a convolution neural network-long short-term memory, the LIG-based triboelectric tire monitoring (LTTM) system decouples the driving information, such as traffic lines and road states, from varied environmental conditions of humidity (10%-90%) and temperatures (50-70 °C). The real-time line and road state recognition of the LTTM system is confirmed on a mobile platform across diverse environmental conditions, including fog, dampness, intense sunlight, and heat shimmer. This work provides an environmentally robust monitoring AIoT system by introducing a self-powered triboelectric sensor and hybrid deep learning for smart mobility.

A data-efficient and easy-to-use lip language interface based on wearable motion capture and speech movement reconstruction

Lip language recognition urgently needs wearable and easy-to-use interfaces for interference-free and high-fidelity lip-reading acquisition and to develop accompanying data-efficient decoder-modeling methods. Existing solutions suffer from unreliable lip reading, are data hungry, and exhibit poor generalization. Here, we propose a wearable lip language decoding technology that enables interference-free and high-fidelity acquisition of lip movements and data-efficient recognition of fluent lip language based on wearable motion capture and continuous lip speech movement reconstruction. The method allows us to artificially generate any wanted continuous speech datasets from a very limited corpus of word samples from users. By using these artificial datasets to train the decoder, we achieve an average accuracy of 92.0% across individuals (n = 7) for actual continuous and fluent lip speech recognition for 93 English sentences, even observing no training burn on users because all training datasets are artificially generated. Our method greatly minimizes users' training/learning load and presents a data-efficient and easy-to-use paradigm for lip language recognition.

Displacement-pressure biparametrically regulated softness sensory system for intraocular pressure monitoring

High intraocular pressure (IOP) is one of the high-risk pathogenic factors of glaucoma. Existing methods of IOP measurement are based on the direct interaction with the cornea. Commercial ophthalmic tonometers based on snapshot measurements are expensive, bulky, and their operation requires trained personnel. Theranostic contact lenses are easy to use, but they may block vision and cause infection. Here, we report a sensory system for IOP assessment that uses a soft indentor with two asymmetrically deployed iontronic flexible pressure sensors to interact with the eyelid-eyeball in an eye-closed situation. Inspired by human fingertip assessment of softness, the sensory system extracts displacement-pressure information for soft evaluation, achieving high accuracy IOP monitoring (>96%). We further design and custom-make a portable and wearable ophthalmic tonometer based on the sensory system and demonstrate its high efficacy in IOP screening. This sensory system paves a way towards cost-effective, robust, and reliable IOP monitoring.

Predicting concrete strength early age using a combination of machine learning and electromechanical impedance with nano-enhanced sensors

To ensure the structural integrity of concrete and prevent unanticipated fracturing, real-time monitoring of early-age concrete's strength development is essential, mainly through advanced techniques such as nano-enhanced sensors. The piezoelectric-based electro-mechanical impedance (EMI) method with nano-enhanced sensors is emerging as a practical solution for such monitoring requirements. This study presents a strength estimation method based on Non-Destructive Testing (NDT) Techniques and Long Short-Term Memory (LSTM) and artificial neural networks (ANNs) as hybrid (NDT-LSTMs-ANN), including several types of concrete strength-related agents. Input data includes water-to-cement rate, temperature, curing time, and maturity based on interior temperature, allowing experimentally monitoring the development of concrete strength from the early steps of hydration and casting to the last stages of hardening 28 days after the casting. The study investigated the impact of various factors on concrete strength development, utilizing a cutting-edge approach that combines traditional models with nano-enhanced piezoelectric sensors and NDT-LSTMs-ANN enhanced with nanotechnology. The results demonstrate that the hybrid provides highly accurate concrete strength estimation for construction safety and efficiency. Adopting the piezoelectric-based EMI technique with these advanced sensors offers a viable and effective monitoring solution, presenting a significant leap forward for the construction industry's structural health monitoring practices.

Deep Learning Model Coupling Wearable Bioelectric and Mechanical Sensors for Refined Muscle Strength Assessment

Muscle strength (MS) is related to our neural and muscle systems, essential for clinical diagnosis and rehabilitation evaluation. Although emerging wearable technology seems promising for MS assessment, problems still exist, including inaccuracy, spatiotemporal differences, and analyzing methods. In this study, we propose a wearable device consisting of myoelectric and strain sensors, synchronously acquiring surface electromyography and mechanical signals at the same spot during muscle activities, and then employ a deep learning model based on temporal convolutional network (TCN) + Transformer (Tcnformer), achieving accurate grading and prediction of MS. Moreover, by combining with deep clustering, named Tcnformer deep cluster (TDC), we further obtain a 25-level classification for MS assessment, refining the conventional 5 levels. Quantification and validation showcase a patient's postoperative recovery from level 3.2 to level 3.6 in the first few days after surgery. We anticipate that this system will importantly advance precise MS assessment, potentially improving relevant clinical diagnosis and rehabilitation outcomes.

Flexible Self-Powered Low-Decibel Voice Recognition Mask

In environments where silent communication is essential, such as libraries and conference rooms, the need for a discreet means of interaction is paramount. Here, we present a single-electrode, contact-separated triboelectric nanogenerator (CS-TENG) characterized by robust high-frequency sensing capabilities and long-term stability. Integrating this TENG onto the inner surface of a mask allows for the capture of conversational speech signals through airflow vibrations, generating a comprehensive dataset. Employing advanced signal processing techniques, including short-time Fourier transform (STFT), Mel-frequency cepstral coefficients (MFCC), and deep learning neural networks, facilitates the accurate identification of speaker content and verification of their identity. The accuracy rates for each category of vocabulary and identity recognition exceed 92% and 90%, respectively. This system represents a pivotal advancement in facilitating secure and efficient unobtrusive communication in quiet settings, with promising implications for smart home applications, virtual assistant technology, and potential deployment in security and confidentiality-sensitive contexts.

Thermodynamic driving forces in contact electrification between polymeric materials

Contact electrification, or contact charging, refers to the process of static charge accumulation after rubbing, or even simple touching, of two materials. Despite its relevance in static electricity, various natural phenomena, and numerous technologies, contact charging remains poorly understood. For insulating materials, even the species of charge carrier may be unknown, and the direction of charge-transfer lacks firm molecular-level explanation. Here, we use all-atom molecular dynamics simulations to investigate whether thermodynamics can explain contact charging between insulating polymers. Based on prior work suggesting that water-ions, such as hydronium and hydroxide ions, are potential charge carriers, we predict preferred directions of charge-transfer between polymer surfaces according to the free energy of water-ions within water droplets on such surfaces. Broad agreement between our predictions and experimental triboelectric series indicate that thermodynamically driven ion-transfer likely influences contact charging of polymers. Furthermore, simulation analyses reveal how specific interactions of water and water-ions proximate to the polymer-water interface explain observed trends. This study establishes relevance of thermodynamic driving forces in contact charging of insulators with new evidence informed by molecular-level interactions. These insights have direct implications for future mechanistic studies and applications of contact charging involving polymeric materials.

High Signal-to-Noise Ratio Event-Driven MEMS Motion Sensing

Two solutions for improving MEMS triboelectric vibration sensors performance in contact-separation mode are reported experimentally and analytically. Triboelectric sensors have mostly been studied in the mesoscale. The gap variation between the electrodes induces a potential difference that represents the external vibration. Miniaturizing the device limits the sensor output because of the limited gap. This work offers a warped MEMS diaphragm constrained on its edges. The dome-shaped structure provides one order of magnitude larger displacement after contact-separation than standard designs resulting in one order of magnitude greater voltage and signal-to-noise-ratio. Second, micro triboelectric sensors do not operate unless the external vibration is sufficiently forceful to initiate contact between layers. The proposed constraints on the edge of the diaphragm provide friction during periodic motion and generate charges. The combination of the warped diaphragm and boundary constraints instead of serpentine springs increases the charge density and voltage generation. The mechanical properties and electrical output are thoroughly investigated including nonlinearity, sensitivity, and signal-to-noise ratio. A sensitivity of 250 mV g-1 and signal-to-noise-ratio of 32 dB is provided by the presented device at resonance, which is very promising for event-driven motion sensors because it does not require signal conditioning and therefore simplifies the sensing circuitry.

Integrated visual transformer and flash attention for lip-to-speech generation GAN

Lip-to-Speech (LTS) generation is an emerging technology that is highly visible, widely supported, and rapidly evolving. LTS has a wide range of promising applications, including assisting speech impairment and improving speech interaction in virtual assistants and robots. However, the technique faces the following challenges: (1) Chinese lip-to-speech generation is poorly recognized. (2) The wide range of variation in lip-speaking is poorly aligned with lip movements. Addressing these challenges will contribute to advancing Lip-to-Speech (LTS) technology, enhancing the communication abilities, and improving the quality of life for individuals with disabilities. Currently, lip-to-speech generation techniques usually employ the GAN architecture but suffer from the following problems: The primary issue lies in the insufficient joint modeling of local and global lip movements, resulting in visual ambiguities and inadequate image representations. To solve these problems, we design Flash Attention GAN (FA-GAN) with the following features: (1) Vision and audio are separately coded, and lip motion is jointly modelled to improve speech recognition accuracy. (2) A multilevel Swin-transformer is introduced to improve image representation. (3) A hierarchical iterative generator is introduced to improve speech generation. (4) A flash attention mechanism is introduced to improve computational efficiency. Many experiments have indicated that FA-GAN can recognize Chinese and English datasets better than existing architectures, especially the recognition error rate of Chinese, which is only 43.19%, the lowest among the same type.

Stable Flexible Electronic Devices under Harsh Conditions Enabled by Double-Network Hydrogels Containing Binary Cations

Hydrogels are increasingly used in flexible electronic devices, but the mechanical and electrochemical stabilities of hydrogel devices are often limited under specific harsh conditions. Herein, chemically/physically cross-linked double-network (DN) hydrogels containing binary cations Zn2+ and Li+ are constructed in order to address the above challenges. Double networks of chemically cross-linked polyacrylamide (PAM) and physically cross-linked κ-Carrageenan (κ-CG) are designed to account for the mechanical robustness while binary cations endow the hydrogels with excellent ionic conductivity and outstanding environmental adaptability. Excellent mechanical robustness and ionic conductivity (25 °C, 2.26 S·m-1; -25 °C, 1.54 S·m-1) have been achieved. Utilizing the DN hydrogels containing binary cations as signal-converting materials, we fabricated flexible mechanosensors. High gauge factors (resistive strain sensors, 2.4; capacitive pressure sensors, 0.82 kPa-1) and highly stable sensing ability have been achieved. Interestingly, zinc-ion hybrid supercapacitors containing the DN hydrogels containing binary cations as electrolytes have achieved an initial capacity of 52.5 mAh·g-1 at a current density of 3 A·g-1 and a capacity retention rate of 82.9% after 19,000 cycles. Proper working of the zinc-ion hybrid supercapacitors at subzero conditions and stable charge-discharge for more than 19,000 cycles at -25 °C have been demonstrated. Overall, DN hydrogels containing binary cations have provided promising materials for high-performance flexible electronic devices under harsh conditions.

A Review of Recent Advances in Human-Motion Energy Harvesting Nanogenerators, Self-Powering Smart Sensors and Self-Charging Electronics

In recent years, portable and wearable personal electronic devices have rapidly developed with increasing mass production and rising energy consumption, creating an energy crisis. Using batteries and supercapacitors with limited lifespans and environmental hazards drives the need to find new, environmentally friendly, and renewable sources. One idea is to harness the energy of human motion and convert it into electrical energy using energy harvesting devices-piezoelectric nanogenerators (PENGs), triboelectric nanogenerators (TENGs) and hybrids. They are characterized by a wide variety of features, such as lightness, flexibility, low cost, richness of materials, and many more. These devices offer the opportunity to use new technologies such as IoT, AI or HMI and create smart self-powered sensors, actuators, and self-powered implantable/wearable devices. This review focuses on recent examples of PENGs, TENGs and hybrid devices for wearable and implantable self-powered systems. The basic mechanisms of operation, micro/nano-scale material selection and manufacturing processes of selected examples are discussed. Current challenges and the outlook for the future of the nanogenerators are also discussed.

Viscoelastic blood coagulation testing system enabled by a non-contact triboelectric angle sensor

Thromboelastography (TEG) remains a convenient and effective viscoelastic blood coagulation testing device for guiding blood component transfusion and assessing the risk of thrombosis. Here, a TEG enabled by a non-contact triboelectric angle sensor (NTAS) with a small size (∼7 cm3) is developed for assessing the blood coagulation system. With the assistance of a superelastic torsion wire structure, the NTAS-TEG realizes the detection of blood viscoelasticity. Benefiting from a grating and convex design, the NTAS holds a collection of compelling features, including accurate detection of rotation angles from -2.5° to 2.5°, high linearity (R 2 = 0.999), and a resolution of 0.01°. Besides, the NTAS exhibits merits of low cost and simplified fabrication. Based on the NTAS-TEG, a viscoelastic blood coagulation detection and analysis system is successfully constructed, which can provide a graph and parameters associated with clot initiation, formation, and stability for clinicians by using 0.36 mL of whole blood. The system not only validates the feasibility of the triboelectric coagulation testing sensor, but also further expands the application of triboelectric sensors in healthcare.

Controlled Preparation of Highly Stretchable, Crack-Free Wrinkled Surfaces with Tunable Wetting and Optical Properties

Constructing wrinkles by utilizing strain-driven surface instability in film-substrate systems is a general method to prepare micronano structures, which have a wide range of applications in smart surfaces and devices such as flexible electronics, reversible wetting, friction, and optics. However, cracks generated during the preparation and use process significantly affect the uniformity of wrinkled surfaces and degrade the functional properties of the film devices. The realization of crack-free wrinkles with high stretchability in hard film systems is still a great challenge. Here, we report on a facile technique for controllable preparation of large-area, highly stretchable, crack-free wrinkled surfaces by ultraviolet ozone (UVO) treatment of Ecoflex. The thickness dependence of the wrinkles and the in situ wrinkling process during mechanical loading are investigated. The wrinkles including striped, labyrinth-like, herringbone, and transitional structures are controllable by changing strain mode (uniaxial or biaxial), loading history (simultaneous or sequential), strain anisotropy, and gradient loading. The wrinkled surfaces obtained using UVO-treated Ecoflex have tunable wetting and optical properties and can maintain excellent mechanical stability under large strains. This study provides a facile method for the preparation of large-area, crack-free wrinkles, which is simple, fast, low-cost, and robust. The resulting wrinkled surfaces remain stable under high stretching, which is beneficial for many practical applications, especially in the cases of large strains.

Encoding of multi-modal emotional information via personalized skin-integrated wireless facial interface

Human affects such as emotions, moods, feelings are increasingly being considered as key parameter to enhance the interaction of human with diverse machines and systems. However, their intrinsically abstract and ambiguous nature make it challenging to accurately extract and exploit the emotional information. Here, we develop a multi-modal human emotion recognition system which can efficiently utilize comprehensive emotional information by combining verbal and non-verbal expression data. This system is composed of personalized skin-integrated facial interface (PSiFI) system that is self-powered, facile, stretchable, transparent, featuring a first bidirectional triboelectric strain and vibration sensor enabling us to sense and combine the verbal and non-verbal expression data for the first time. It is fully integrated with a data processing circuit for wireless data transfer allowing real-time emotion recognition to be performed. With the help of machine learning, various human emotion recognition tasks are done accurately in real time even while wearing mask and demonstrated digital concierge application in VR environment.

Synergizing Machine Learning Algorithm with Triboelectric Nanogenerators for Advanced Self-Powered Sensing Systems

The advancement of the Internet of Things (IoT) has increased the demand for large-scale intelligent sensing systems. The periodic replacement of power sources for ubiquitous sensing systems leads to significant resource waste and environmental pollution. Human staffing costs associated with replacement also increase the economic burden. The triboelectric nanogenerators (TENGs) provide both an energy harvesting scheme and the possibility of self-powered sensing. Based on contact electrification from different materials, TENGs provide a rich material selection to collect complex and diverse data. As the data collected by TENGs become increasingly numerous and complex, different approaches to machine learning (ML) and deep learning (DL) algorithms have been proposed to efficiently process output signals. In this paper, the latest advances in ML algorithms assisting solid-solid TENG and liquid-solid TENG sensors are reviewed based on the sample size and complexity of the data. The pros and cons of various algorithms are analyzed and application scenarios of various TENG sensing systems are presented. The prospects of synergizing hardware (TENG sensors) with software (ML algorithms) in a complex environment and their main challenges for future developments are discussed.

Micropyramid Array Bimodal Electronic Skin for Intelligent Material and Surface Shape Perception Based on Capacitive Sensing

Developing electronic skins (e-skins) that are comparable to or even beyond human tactile perception holds significant importance in advancing the process of intellectualization. In this context, a machine-learning-motivated micropyramid array bimodal (MAB) e-skin based on capacitive sensing is reported, which enables spatial mapping applications based on bimodal sensing (proximity and pressure) implemented via fringing and iontronic effects, such as contactless measurement of 3D objects and contact recognition of Braille letters. Benefiting from the iontronic effect and single-micropyramid structure, the MAB e-skin in pressure mode yields impressive features: a maximum sensitivity of 655.3 kPa-1 (below 0.5 kPa), a linear sensitivity of 327.9 kPa-1 (0.5-15 kPa), and an ultralow limit of detection of 0.2 Pa. With the assistance of multilayer perceptron and convolutional neural network, the MAB e-skin can accurately perceive 6 materials and 10 surface shapes based on the training and learning using the collected datasets from proximity and pressure modes, thus allowing it to achieve the precise perception of different objects within one proximity-pressure cycle. The development of this MAB e-skin opens a new avenue for robotic skin and the expansion of advanced applications.

Implantable Triboelectric Nanogenerators for Self-Powered Cardiovascular Healthcare

Triboelectric nanogenerators (TENGs) have gained significant traction in recent years in the bioengineering community. With the potential for expansive applications for biomedical use, many individuals and research groups have furthered their studies on the topic, in order to gain an understanding of how TENGs can contribute to healthcare. More specifically, there have been a number of recent studies focusing on implantable triboelectric nanogenerators (I-TENGs) toward self-powered cardiac systems healthcare. In this review, the progression of implantable TENGs for self-powered cardiovascular healthcare, including self-powered cardiac monitoring devices, self-powered therapeutic devices, and power sources for cardiac pacemakers, will be systematically reviewed. Long-term expectations of these implantable TENG devices through their biocompatibility and other utilization strategies will also be discussed.

A Metamaterial Computational Multi-Sensor of Grip-Strength Properties with Point-of-Care Human-Computer Interaction

Grip strength is a biomarker of frailty and an evaluation indicator of brain health, cardiovascular morbidity, and psychological health. Yet, the development of a reliable, interactive, and point-of-care device for comprehensive multi-sensing of hand grip status is challenging. Here, a relation between soft buckling metamaterial deformations and built piezoelectric voltage signals is uncovered to achieve multiple sensing of maximal grip force, grip speed, grip impulse, and endurance indicators. A metamaterial computational sensor design is established by hyperelastic model that governs the mechanical characterization, machine learning models for computational sensing, and graphical user interface to provide visual cues. A exemplify grip measurement for left and right hands of seven elderly campus workers is conducted. By taking indicators of grip status as input parameters, human-computer interactive games are incorporated into the computational sensor to improve the user compliance with measurement protocols. Two elderly female schizophrenic patients are participated in the real-time interactive point-of-care grip assessment and training for potentially sarcopenia screening. The attractive features of this advanced intelligent metamaterial computational sensing system are crucial to establish a point-of-care biomechanical platform and advancing the human-computer interactive healthcare, ultimately contributing to a global health ecosystem.

Conformable Multifunctional Space Fabric by Metal 3D Printing for Collision Hazard Protection and Self-Powered Monitoring

The monitoring of space debris assumes paramount significance to ensure the sustainability and security of space activities as well as underground bases in outer space. However, designing a wide range monitoring system with easy fabrication, low power, and high precision remains an urgent challenge under the scarcity of materials and extreme environment conditions of outer space. Here, we designed a one-piece, robust, but flexible, and repairable 3D metal-printed triboelectric nanogenerator (FR-TENG) by incorporating the advantages of standardization and customization of outer space 3D metal printing. Inspired by the structure of hexagonal and pangolin scales, a curved structure is ingeniously applied in the design of 3D printed metal to adapt different curved surfaces while maintaining superior compressive strength, providing excellent flexibility and shape adaptability. Benefiting from the unique structural design, the FR-TENG has a minimum length of 1 cm with a weight of only 3.5 g and the minimum weight resolution detected of 9.6 g, with a response time of 20 ms. Furthermore, a multichannel self-powered collision monitoring system has been developed to monitor minor collisions, providing warnings to determine potential impacts on the space station and bases surfaces. The system may contribute to ensuring the successful completion of space missions and providing a safer space environment for the exploration of extraterrestrial life and the establishment of underground protective bases.

Design and application of a flexible nano cardiac sound sensor based on P(VDF-TrFE)/KNN/GR composite piezoelectric film for heart disease diagnosis

The paper proposes a flexible micro-nano composite piezoelectric thin film. This flexible piezoelectric film is fabricated through electrospinning process, utilizing a combination of 12 wt% poly(vinylidene fluoride-co-trifluoroethylene)(P(VDF-TrFE)), 8 wt% potassium sodium niobate (KNN) nanoparticles, and 0.5 wt% graphene (GR). Under cyclic loading, the composite film demonstrates a remarkable increase in open-circuit voltage and short-circuit current, achieving values of 36.1 V and 163.7 uA, respectively. These values are 5.8 times and 3.6 times higher than those observed in the pure P(VDF-TrFE) film. The integration of this piezoelectric film into a wearable flexible heartbeat sensor, coupled with the RepMLP classification model, facilitates heartbeat acquisition and real-time automated diagnosis. After training and validation on a dataset containing 2000 heartbeat samples, the system achieved an accuracy of approximately 99% in two classification of heart sound signals (normal and abnormal). This research substantially enhances the output performance of the piezoelectric film, offering a novel and valuable solution for the application of flexible piezoelectric films in physiological signal detection.

Decoding silent speech commands from articulatory movements through soft magnetic skin and machine learning

Silent speech interfaces have been pursued to restore spoken communication for individuals with voice disorders and to facilitate intuitive communications when acoustic-based speech communication is unreliable, inappropriate, or undesired. However, the current methodology for silent speech faces several challenges, including bulkiness, obtrusiveness, low accuracy, limited portability, and susceptibility to interferences. In this work, we present a wireless, unobtrusive, and robust silent speech interface for tracking and decoding speech-relevant movements of the temporomandibular joint. Our solution employs a single soft magnetic skin placed behind the ear for wireless and socially acceptable silent speech recognition. The developed system alleviates several concerns associated with existing interfaces based on face-worn sensors, including a large number of sensors, highly visible interfaces on the face, and obtrusive interconnections between sensors and data acquisition components. With machine learning-based signal processing techniques, good speech recognition accuracy is achieved (93.2% accuracy for phonemes, and 87.3% for a list of words from the same viseme groups). Moreover, the reported silent speech interface demonstrates robustness against noises from both ambient environments and users' daily motions. Finally, its potential in assistive technology and human-machine interactions is illustrated through two demonstrations - silent speech enabled smartphone assistants and silent speech enabled drone control.

Can language models be used for real-world urban-delivery route optimization?

Language models have contributed to breakthroughs in interdisciplinary research, such as protein design and molecular dynamics understanding. In this study, we reveal that beyond language, representations of other entities, such as human behaviors, that are mappable to learnable sequences can be learned by language models. One compelling example is the real-world delivery route optimization problem. We here propose a novel approach based on the language model to optimize delivery routes on the basis of drivers' historical experiences. Although a broad range of optimization-based approaches have been designed to optimize delivery routes, they do not capture the implicit knowledge of complex delivery operating environments. The model we propose integrates this knowledge in the route optimization process by learning from driving behaviors in experienced drivers. A real-world delivery route that preserves drivers' implicit behavioral patterns is first analogized to a sentence in natural language. Through unsupervised learning, we then learn the vector representations of words and infer the drivers' delivery chains on the basis of the tailored chain-reaction-based algorithm. We also provide insights into the fusion of language models and operations research methods. In our approach, language models are applied to learn drivers' delivery behaviors and infer new deliveries at the delivery zone level, while the classic traveling salesman problem (TSP) model is embedded into the hybrid framework for intra-zone optimization. Numerical experiments performed on real-world data from Amazon's delivery service demonstrate that the proposed approach outperforms pure optimization, supporting the effectiveness, efficiency, and extensibility of our model. As a versatile approach, the proposed framework can easily be extended to various disciplines in which the data follow certain grammar rules. We anticipate that our work will serve as a stepping stone toward the understanding and application of language models in tackling interdisciplinary research problems.

Recent Progress of Advanced Materials for Triboelectric Nanogenerators

Triboelectric nanogenerators (TENGs) have received intense attention due to their broad application prospects in the new era of internet of things (IoTs) as distributed power sources and self-powered sensors. Advanced materials are vital components for TENGs, which decide their comprehensive performance and application scenarios, opening up the opportunity to develop efficient TENGs and expand their potential applications. In this review, a systematic and comprehensive overview of the advanced materials for TENGs is presented, including materials classifications, fabrication methods, and the properties required for applications. In particular, the triboelectric, friction, and dielectric performance of advanced materials is focused upon and their roles in designing the TENGs are analyzed. The recent progress of advanced materials used in TENGs for mechanical energy harvesting and self-powered sensors is also summarized. Finally, an overview of the emerging challenges, strategies, and opportunities for research and development of advanced materials for TENGs is provided.

Sensing in Soft Robotics

Soft robotics is an exciting field of science and technology that enables robots to manipulate objects with human-like dexterity. Soft robots can handle delicate objects with care, access remote areas, and offer realistic feedback on their handling performance. However, increased dexterity and mechanical compliance of soft robots come with the need for accurate control of the position and shape of these robots. Therefore, soft robots must be equipped with sensors for better perception of their surroundings, location, force, temperature, shape, and other stimuli for effective usage. This review highlights recent progress in sensing feedback technologies for soft robotic applications. It begins with an introduction to actuation technologies and material selection in soft robotics, followed by an in-depth exploration of various types of sensors, their integration methods, and the benefits of multimodal sensing, signal processing, and control strategies. A short description of current market leaders in soft robotics is also included in the review to illustrate the growing demands of this technology. By examining the latest advancements in sensing feedback technologies for soft robots, this review aims to highlight the potential of soft robotics and inspire innovation in the field.

Eye tracking and eye expression decoding based on transparent, flexible and ultra-persistent electrostatic interface

Eye tracking provides valuable insight for analyzing visual attention and underlying thinking progress through the observation of eye movements. Here, a transparent, flexible and ultra-persistent electrostatic sensing interface is proposed for realizing active eye tracking (AET) system based on the electrostatic induction effect. Through a triple-layer structure combined with a dielectric bilayer and a rough-surface Ag nanowire (Ag NW) electrode layer, the inherent capacitance and interfacial trapping density of the electrostatic interface has been strongly enhanced, contributing to an unprecedented charge storage capability. The electrostatic charge density of the interface reached 1671.10 μC·m^-2 with a charge-keeping rate of 96.91% after 1000 non-contact operation cycles, which can finally realize oculogyric detection with an angular resolution of 5°. Thus, the AET system enables real-time decoding eye movements for customer preference recording and eye-controlled human-computer interaction, supporting its limitless potentiality in commercial purpose, virtual reality, human computer interactions and medical monitoring.

Electromyogram-Based Lip-Reading via Unobtrusive Dry Electrodes and Machine Learning Methods

Lip-reading provides an effective speech communication interface for people with voice disorders and for intuitive human-machine interactions. Existing systems are generally challenged by bulkiness, obtrusiveness, and poor robustness against environmental interferences. The lack of a truly natural and unobtrusive system for converting lip movements to speech precludes the continuous use and wide-scale deployment of such devices. Here, the design of a hardware-software architecture to capture, analyze, and interpret lip movements associated with either normal or silent speech is presented. The system can recognize different and similar visemes. It is robust in a noisy or dark environment. Self-adhesive, skin-conformable, and semi-transparent dry electrodes are developed to track high-fidelity speech-relevant electromyogram signals without impeding daily activities. The resulting skin-like sensors can form seamless contact with the curvilinear and dynamic surfaces of the skin, which is crucial for a high signal-to-noise ratio and minimal interference. Machine learning algorithms are employed to decode electromyogram signals and convert them to spoken words. Finally, the applications of the developed lip-reading system in augmented reality and medical service are demonstrated, which illustrate the great potential in immersive interaction and healthcare applications.

Worm Generator: A System for High-Throughput in Vivo Screening

Large-scale screening of molecules in organisms requires high-throughput and cost-effective evaluating tools during preclinical development. Here, a novel in vivo screening strategy combining hierarchically structured biohybrid triboelectric nanogenerators (HB-TENGs) arrays with computational bioinformatics analysis for high-throughput pharmacological evaluation using Caenorhabditis elegans is described. Unlike the traditional methods for behavioral monitoring of the animals, which are laborious and costly, HB-TENGs with micropillars are designed to efficiently convert animals' behaviors into friction deformation and result in a contact-separation motion between two triboelectric layers to generate electrical outputs. The triboelectric signals are recorded and extracted to various bioinformation for each screened compound. Moreover, the information-rich electrical readouts are successfully demonstrated to be sufficient to predict a drug's identity by multiple-Gaussian-kernels-based machine learning methods. This proposed strategy can be readily applied to various fields and is especially useful in in vivo explorations to accelerate the identification of novel therapeutics.

Anatomically Designed Triboelectric Wristbands with Adaptive Accelerated Learning for Human-Machine Interfaces

Recent advances in flexible wearable devices have boosted the remarkable development of devices for human-machine interfaces, which are of great value to emerging cybernetics, robotics, and Metaverse systems. However, the effectiveness of existing approaches is limited by the quality of sensor data and classification models with high computational costs. Here, a novel gesture recognition system with triboelectric smart wristbands and an adaptive accelerated learning (AAL) model is proposed. The sensor array is well deployed according to the wrist anatomy and retrieves hand motions from a distance, exhibiting highly sensitive and high-quality sensing capabilities beyond existing methods. Importantly, the anatomical design leads to the close correspondence between the actions of dominant muscle/tendon groups and gestures, and the resulting distinctive features in sensor signals are very valuable for differentiating gestures with data from 7 sensors. The AAL model realizes a 97.56% identification accuracy in training 21 classes with only one-third operands of the original neural network. The applications of the system are further exploited in real-time somatosensory teleoperations with a low latency of <1 s, revealing a new possibility for endowing cyber-human interactions with disruptive innovation and immersive experience.