Search Results (47)

This research presents a novel, real-time Pakistani Sign Language (PSL) recognition system utilizing a custom-designed sensory glove integrated with advanced machine learning techniques. The system aims to bridge communication gaps for individuals with hearing and speech impairments by translating hand gestures into readable text. At the core of this work is a smart glove engineered with five resistive flex sensors for precise finger flexion detection and a 9-DOF Inertial Measurement Unit (IMU) for capturing hand orientation and movement. The glove is powered by a compact microcontroller, which processes the analog and digital sensor inputs and transmits the data wirelessly to a host computer. A rechargeable 3.7 V Li-Po battery ensures portability, while a dynamic dataset comprising both static alphabet gestures and dynamic PSL phrases was recorded using this setup. The collected data was used to train two models: a Support Vector Machine with feature extraction (SVM-FE) and a Long Short-Term Memory (LSTM) deep learning network. The LSTM model outperformed traditional methods, achieving an accuracy of 98.6% in real-time gesture recognition. The proposed system demonstrates robust performance and offers practical applications in smart home interfaces, virtual and augmented reality, gaming, and assistive technologies. By combining ergonomic hardware with intelligent algorithms, this research takes a significant step toward inclusive communication and more natural human–machine interaction. Full article

Bone drilling is a widely encountered process in orthopedic surgeries and keyhole neuro surgeries. We are developing a sensor-integrated smart end-effector for drilling for robotic surgical applications. In manual surgeries, surgeons assess tool wear based on experience and force perception. In this work, we propose a machine-learning (ML)-based tool condition monitoring system based on multi-sensor data to preempt excessive tool wear during drilling in robotic surgery. Real-time data is acquired from the six-component force sensor of a collaborative arm along with the data from the temperature and multi-axis vibration sensor mounted on the bone specimen being drilled upon. Raw data from the sensors may have noises and outliers. Signal processing in the time- and frequency-domain are used for denoising as well as to obtain additional features to be derived from the raw sensory data. This paper addresses the challenging problem of identification of the most suitable ML algorithm and the most suitable features to be used as inputs to the algorithm. While dozens of features and innumerable machine learning and deep learning models are available, this paper addresses the problem of selecting the most relevant features, the most relevant AI models, and the optimal hyperparameters to be used in the AI model to provide accurate prediction on the tool condition. A unique framework is proposed for classifying tool wear that combines machine learning-based modeling with multi-sensor data. From the raw sensory data that contains only a handful of features, a number of additional features are derived using frequency-domain techniques and statistical measures. Using feature engineering, we arrived at a total of 60 features from time-domain, frequency-domain, and interaction-based metrics. Such additional features help in improving its predictive capabilities but make the training and prediction complicated and time-consuming. Using a sequence of techniques such as variance thresholding, correlation filtering, ANOVA F-test, and SHAP analysis, the number of features was reduced from 60 to the 4 features that will be most effective in real-time tool condition prediction. In contrast to previous studies that only examine a small number of machine learning models, our approach systematically evaluates a wide range of machine learning and deep learning architectures. The performances of 47 classical ML models and 6 deep learning (DL) architectures were analyzed using the set of the four features identified as most suitable. The Extra Trees Classifier (an ML model) and the one-dimensional Convolutional Neural Network (1D CNN) exhibited the best prediction accuracy among the models studied. Using real-time data, these models monitored the drilling tool condition in real-time to classify the tool wear into three categories of slight, moderate, and severe. Full article

The integration of robotics and mobile networks (5G/6G) through the Internet of Robotic Things (IoRT) is revolutionizing telemedicine, enabling remote physician participation in scenarios where specialists are scarce, where there is a high risk to them, such as in conflicts or natural disasters, or where access to a medical facility is not possible. Nevertheless, touching a human safely with a robotic arm in non-engineered or even out-of-hospital environments presents substantial challenges. This article presents a novel IoRT approach for healthcare in or from remote areas, enabling interaction between a specialist’s hand and a robotic hand. We introduce the IoRT-in-hand: a smart, lightweight end-effector that extends the specialist’s hand, integrating a medical instrument, an RGB camera with servos, a force/torque sensor, and a mini-PC with Internet connectivity. Additionally, we propose an open-source Android app combining MQTT and ROS for real-time remote manipulation, alongside an Edge–Cloud architecture that links the physical robot with its Digital Twin (DT), enabling precise control and 3D visual feedback of the robot’s environment. A proof of concept is presented for the proposed tele-robotic system, using a 6-DOF manipulator with the IoRT-in-hand to perform an ultrasound scan. Teleoperation was conducted over 2300 km via a 5G NSA network on the operator side and a wired network in a laboratory on the robot side. Performance was assessed through human subject feedback, sensory data, and latency measurements, demonstrating the system’s potential for remote healthcare and emergency applications. The source code and CAD models of the IoRT-in-hand prototype are publicly available in an open-access repository to encourage reproducibility and facilitate further developments in robotic telemedicine. Full article

This paper introduces SmartBarrel, an innovative IoT-based sensory system that monitors and forecasts wine fermentation processes. At the core of SmartBarrel are two compact, attachable devices—the probing nose (E-nose) and the probing tongue (E-tongue), which mount directly onto stainless steel wine tanks. These devices periodically measure key fermentation parameters: the nose monitors gas emissions, while the tongue captures acidity, residual sugar, and color changes. Both utilize low-cost, low-power sensors validated through small-scale fermentation experiments. Beyond the sensory hardware, SmartBarrel includes a robust cloud infrastructure built on open-source Industry 4.0 tools. The system leverages the ThingsBoard platform, supported by a NoSQL Cassandra database, to provide real-time data storage, visualization, and mobile application access. The system also supports adaptive breakpoint alerts and real-time adjustment to the nonlinear dynamics of wine fermentation. The authors developed a novel deep learning model called V-LSTM (Variable-length Long Short-Term Memory) to introduce intelligence to enable predictive analytics. This auto-calibrating architecture supports variable layer depths and cell configurations, enabling accurate forecasting of fermentation metrics. Moreover, the system includes two fuzzy logic modules: a device-level fuzzy controller to estimate alcohol content based on sensor data and a fuzzy encoder that synthetically generates fermentation profiles using a limited set of experimental curves. SmartBarrel experimental results validate the SmartBarrel’s ability to monitor fermentation parameters. Additionally, the implemented models show that the V-LSTM model outperforms existing neural network classifiers and regression models, reducing RMSE loss by at least 45%. Furthermore, the fuzzy alcohol predictor achieved a coefficient of determination ($R^2$) of 0.87, enabling reliable alcohol content estimation without direct alcohol sensing. Full article

This study presents a systematic bibliometric review of wearable technologies aimed at vulnerable road user (VRU) safety, covering publications from 2000 to 2025. Guided by PRISMA procedures and a PICo-based search strategy, 58 records were extracted and analyzed in CiteSpace, yielding visualizations of collaboration networks, publication trajectories, and intellectual structures. The results indicate a clear evolution from single-purpose, stand-alone devices to integrated ecosystem solutions that address the needs of diverse VRU groups. Six dominant knowledge clusters emerged—street-crossing assistance, obstacle avoidance, human–computer interaction, cyclist safety, blind navigation, and smart glasses. Comparative analysis across pedestrians, cyclists and motorcyclists, and persons with disabilities shows three parallel transitions: single- to multisensory interfaces, reactive to predictive systems, and isolated devices to V2X-enabled ecosystems. Contemporary research emphasizes context-adaptive interfaces, seamless V2X integration, and user-centered design, and future work should focus on lightweight communication protocols, adaptive sensory algorithms, and personalized safety profiles. The review provides a consolidated knowledge map to inform researchers, practitioners, and policy-makers striving for inclusive and proactive road safety solutions. Full article

The concept of a digital twin (DT), rooted in mid-20th-century ideas, has recently gained significant traction even outside software simulation and engineering modeling. The recent advancements in computational power and the development of model integration methodologies have enabled the creation of virtual replicas of complex physical objects. The success of DTs in engineering has also pushed for the exploration of their use in other domains, especially where complex systems are at stake. One of these cases, which is the focus of this paper, is the modeling of cities and the way they are transformed via technologies into so-called smart cities. In these systems, the huge amount of data that are made accessible and constantly updated via sensor networks suggests that one can use DTs dedicated to the urban scenario as data-driven decision-making devices. However, the concept of a DT was not developed for socio-technical systems and requires careful analysis when applied to urban scenarios. While technologies and information systems have become integrated into city management, this has not reduced the complexity of the city. Relying only on sensory data for city modeling and management seems pretentious since detectable data (what is made accessible via sensor networks) do not seem suitable to inform on all important aspects of the city. Urban DTs hold promise, yet their development necessitates careful consideration of both opportunities and limitations. For this goal, it can be helpful to exploit an ontological analysis due to its neutral and systematic approach and to look at a city as a system of intertwined relationships across its components, such as places, agents, and knowledge. The variety of interactions that the components manifest highlights aspects of the city that the type of data we can collect today leaves unexplored. The paper presents a preliminary example of this issue by studying cases of city squares. The final part of this paper is a call to analyze DTs’ potential role in urban contexts and become aware of the intrinsic limitations of the data they rely upon. Full article

Due to the lucrative coffee market, this product is often subject to adulteration, as inferior or non-coffee materials or varieties are mixed in, negatively affecting its quality. Traditional sensory evaluations by expert tasters and chemical analysis methods, although effective, are time-consuming, costly, and require skilled personnel. The aim of this work was to evaluate the capacity of a smart electronic tongue (e-tongue) based on a polypyrrole sensor array as a tool for the rapid analysis of coffees elaborated from beans of different varieties. The smart e-tongue device was developed with a polypyrrole-based voltammetric sensor array and portable multi-potentiostat operated via smartphone. The sensor array comprised seven electrodes, each doped with distinct counterions to enhance cross-selectivity. The smart e-tongue was tested on five Arabica coffee varieties (Typica, Bourbon, Maragogype, Tabi, and Caturra). The resulting voltammetric signals were analyzed using principal component analysis assisted by neural networks (PCNN) and cluster analysis (CA), enabling clear discrimination among the coffee samples. The results demonstrate that the polypyrrole sensors can generate distinct electrochemical patterns, serving as “fingerprints” for each coffee variety. This study highlights the potential of polypyrrole-based smart e-tongues as a rapid, cost-effective, and portable alternative for coffee quality assessment and adulteration detection, with broader applications in the food and beverage industry. Full article

In this paper, which represents a fundamental step in ongoing research, a new smart low-energy dual-function half-mode substrate integrated waveguide cavity-interdigital capacitor (HMSIWC-DIC) antenna-based sensor is developed and investigated for remote frost and wildfire detection applications at 5.7 GHz. The proposed methodology exploits the HMSIW antenna-based sensor, a microfluidic channel (microliter water channel (50 μL)), interdigital capacitor technologies, and the resonance frequency parameters combined with machine learning algorithms. This allows for superior interaction between the water channel and the TE101 mode, resulting in high sensitivity (∆f/∆ε = 5.5 MHz/ε (F/m) and ∆f/∆°C = 1.83 MHz/°C) within the sensing range. Additionally, it exhibits high decision-making ability and immunity to interference, demonstrating a best-in-class sensory response to weather temperature across two ranges: positive (≥0 °C, including frost and wildfire) and negative (<0 °C, including ice accumulation). To address the challenges posed by the non-linear, unpredictable behavior of resonance frequency results, even when dealing with weak sensor antenna responses, an innovative sensory intelligent system was proposed. This system utilizes resonance frequency results as features to classify and predict weather temperature ranges into three environmental states: Early Frost, Normal, and Early Wildfire, achieving an accuracy of 96.4%. Several machine learning techniques are employed, including artificial neural networks (ANNs), random forests (RF), decision trees (DT), support vector machines (SVMs), and Gaussian processes (GPs). This sensor serves as an ideal solution for energy management through its utilization in RF-based weather temperature sensing applications. It boasts stable performance, minimal energy consumption, and real-time sensitivity, eliminating the necessity for manual data recording. Full article

Blockchain is recognized for its robust security features, and its integration with Internet of Things (IoT) systems presents scalability and operational challenges. Deploying Artificial Intelligence (AI) within blockchain environments raises concerns about balancing rigorous security requirements with computational efficiency. The prime motivation resides in integrating AI with blockchain to strengthen IoT security and withstand multiple variants of lethal threats. With the increasing number of IoT devices, there has also been a spontaneous increase in security vulnerabilities. While conventional security methods are inadequate for the diversification of IoT devices, adopting AI can assist in identifying and mitigating such threats in real time, whereas integrating AI with blockchain can offer more intelligent decentralized security measures. The paper contributes to a three-layered architecture encompassing the device/sensory, edge, and cloud layers. This structure supports a novel method for assessing legitimacy scores and serves as an initial security measure. The proposed scheme also enhances the architecture by introducing an Ethereum-based data repositioning framework as a potential trapdoor function, ensuring maximal secrecy. To complement this, a simplified consensus module generates a conclusive evidence matrix, bolstering accountability. The model also incorporates an innovative AI-based security optimization utilizing an unconventional neural network model that operates faster and is enhanced with metaheuristic algorithms. Comparative benchmarks demonstrate that our approach results in a 48.5% improvement in threat detection accuracy and a 23.5% reduction in processing time relative to existing systems, marking significant advancements in IoT security for smart cities. Full article

Recently, the research community has shown significant interest in the continuous temporal data obtained from motion sensors in wearable devices. These data are useful for classifying and analysing different human activities in many application areas such as healthcare, sports and surveillance. The literature has presented a multitude of deep learning models that aim to derive a suitable feature representation from temporal sensory input. However, the presence of a substantial quantity of annotated training data is crucial to adequately train the deep networks. Nevertheless, the data originating from the wearable devices are vast but ineffective due to a lack of labels which hinders our ability to train the models with optimal efficiency. This phenomenon leads to the model experiencing overfitting. The contribution of the proposed research is twofold: firstly, it involves a systematic evaluation of fifteen different augmentation strategies to solve the inadequacy problem of labeled data which plays a critical role in the classification tasks. Secondly, it introduces an automatic feature-learning technique proposing a Multi-Branch Hybrid Conv-LSTM network to classify human activities of daily living using multimodal data of different wearable smart devices. The objective of this study is to introduce an ensemble deep model that effectively captures intricate patterns and interdependencies within temporal data. The term “ensemble model” pertains to fusion of distinct deep models, with the objective of leveraging their own strengths and capabilities to develop a solution that is more robust and efficient. A comprehensive assessment of ensemble models is conducted using data-augmentation techniques on two prominent benchmark datasets: CogAge and UniMiB-SHAR. The proposed network employs a range of data-augmentation methods to improve the accuracy of atomic and composite activities. This results in a 5% increase in accuracy for composite activities and a 30% increase for atomic activities. Full article

The decentralized nature of blockchain enables data traceability, transparency, and immutability as complementary security features to the existing Internet of Things (IoT) systems. These Blockchain-based IoT (BIoT) systems aim to mitigate security risks such as malicious control, data leakage, and dishonesty often found in traditional cloud-based, vendor-specific IoT networks. As we steadily advance into the era of situation-aware IoT, the use of machine learning (ML) techniques has become essential for synthesizing situations based on sensory contexts. However, the challenge to integrate learning-based situation awareness with BIoT systems restricts the full potential of such integration. This is primarily due to the conflicts between the deterministic nature of smart contracts and the non-deterministic nature of machine learning, as well as the high costs of conducting machine learning on blockchain. To address the challenge, we propose a framework named Situ-Oracle. With the framework, a computation oracle of the blockchain ecosystem is leveraged to provide situation analysis as a service, based on Recurrent Neural Network (RNN)-based learning models tailored for the Situ model, and specifically designed smart contracts are deployed as intermediary communication channels between the IoT devices and the computation oracle. We used smart homes as a case study to demonstrate the framework design. Subsequently, system-wide evaluations were conducted over a physically constructed BIoT system. The results indicate that the proposed framework achieves better situation analysis accuracy (above 95%) and improves gas consumption as well as network throughput and latency when compared to baseline systems (on-chain learning or off-chain model verification). Overall, the paper presents a promising approach for improving situation analysis for BIoT systems, with potential applications in various domains such as smart homes, healthcare, and industrial automation. Full article

Bridges are critical components of transportation networks, and their conditions have effects on societal well-being, the economy, and the environment. Automation needs in inspections and maintenance have made structural health monitoring (SHM) systems a key research pillar to assess bridge safety/health. The last decade brought a boom in innovative bridge SHM applications with the rise in next-generation smart and mobile technologies. A key advancement within this direction is smartphones with their sensory usage as SHM devices. This focused review reports recent advances in bridge SHM backed by smartphone sensor technologies and provides case studies on bridge SHM applications. The review includes model-based and data-driven SHM prospects utilizing smartphones as the sensing and acquisition portal and conveys three distinct messages in terms of the technological domain and level of mobility: (i) vibration-based dynamic identification and damage-detection approaches; (ii) deformation and condition monitoring empowered by computer vision-based measurement capabilities; (iii) drive-by or pedestrianized bridge monitoring approaches, and miscellaneous SHM applications with unconventional/emerging technological features and new research domains. The review is intended to bring together bridge engineering, SHM, and sensor technology audiences with decade-long multidisciplinary experience observed within the smartphone-based SHM theme and presents exemplary cases referring to a variety of levels of mobility. Full article

This paper explores the development of a smart Structural Health Monitoring (SHM) platform tailored for long-span bridge monitoring, using the Forth Road Bridge (FRB) as a case study. It discusses the selection of smart sensors available for real-time monitoring, the formulation of an effective data strategy encompassing the collection, processing, management, analysis, and visualization of monitoring data sets to support decision-making, and the establishment of a cost-effective and intelligent sensor network aligned with the objectives set through comprehensive communication with asset owners. Due to the high data rates and dense sensor installations, conventional processing techniques are inadequate for fulfilling monitoring functionalities and ensuring security. Cloud-computing emerges as a widely adopted solution for processing and storing vast monitoring data sets. Drawing from the authors’ experience in implementing long-span bridge monitoring systems in the UK and China, this paper compares the advantages and limitations of employing cloud- computing for long-span bridge monitoring. Furthermore, it explores strategies for developing a robust data strategy and leveraging artificial intelligence (AI) and digital twin (DT) technologies to extract relevant information or patterns regarding asset health conditions. This information is then visualized through the interaction between physical and virtual worlds, facilitating timely and informed decision-making in managing critical road transport infrastructure. Full article

Traditional laboratory-based water quality monitoring and testing approaches are soon to be outdated, mainly because of the need for real-time feedback and immediate responses to emergencies. The more recent wireless sensor network (WSN)-based techniques are evolving to alleviate the problems of monitoring, coverage, and energy management, among others. The inclusion of the Internet of Things (IoT) in WSN techniques can further lead to their improvement in delivering, in real time, effective and efficient water-monitoring systems, reaping from the benefits of IoT wireless systems. However, they still suffer from the inability to deliver accurate real-time data, a lack of reconfigurability, the need to be deployed in ad hoc harsh environments, and their limited acceptability within industry. Electronic sensors are required for them to be effectively incorporated into the IoT WSN water-quality-monitoring system. Very few electronic sensors exist for parameter measurement. This necessitates the incorporation of artificial intelligence (AI) sensory techniques for smart water-quality-monitoring systems for indicators without actual electronic sensors by relating with available sensor data. This approach is in its infancy and is still not yet accepted nor standardized by the industry. This work presents a smart water-quality-monitoring framework featuring an intelligent IoT WSN monitoring system. The system uses AI sensors for indicators without electronic sensors, as the design of electronic sensors is lagging behind monitoring systems. In particular, machine learning algorithms are used to predict E. coli concentrations in water. Six different machine learning models (ridge regression, random forest regressor, stochastic gradient boosting, support vector machine, k-nearest neighbors, and AdaBoost regressor) are used on a sourced dataset. From the results, the best-performing model on average during testing was the AdaBoost regressor (a $\overline{MAE}$ of 14.37 counts/100 mL), and the worst-performing model was stochastic gradient boosting (a $\overline{MAE}$ of 42.27 counts/100 mL). The development and application of such a system is not trivial. The best-performing water parameter set (Set A) contained pH, conductivity, chloride, turbidity, nitrates, and chlorophyll. Full article

This study has designed and developed a smart data glove based on five-channel flexible capacitive stretch sensors and a six-axis inertial measurement unit (IMU) to recognize 25 static hand gestures and ten dynamic hand gestures for amphibious communication. The five-channel flexible capacitive sensors are fabricated on a glove to capture finger motion data in order to recognize static hand gestures and integrated with six-axis IMU data to recognize dynamic gestures. This study also proposes a novel amphibious hierarchical gesture recognition (AHGR) model. This model can adaptively switch between large complex and lightweight gesture recognition models based on environmental changes to ensure gesture recognition accuracy and effectiveness. The large complex model is based on the proposed SqueezeNet-BiLSTM algorithm, specially designed for the land environment, which will use all the sensory data captured from the smart data glove to recognize dynamic gestures, achieving a recognition accuracy of 98.21%. The lightweight stochastic singular value decomposition (SVD)-optimized spectral clustering gesture recognition algorithm for underwater environments that will perform direct inference on the glove-end side can reach an accuracy of 98.35%. This study also proposes a domain separation network (DSN)-based gesture recognition transfer model that ensures a 94% recognition accuracy for new users and new glove devices. Full article